1: \documentclass{emulateapj}

2: \slugcomment{{\sc Accepted to ApJ:} September 10, 2008} 

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

4: %\usepackage{natbib}

5: \def\arcsec{$^{\prime\prime}$}

6: \bibliographystyle{apj}

7: \newcommand\degree{{^\circ}}

8: \newcommand\surfb{$\mathrm{mag}/\square$\arcsec}

9: \newcommand\kms{\rm{~km~s}$^{-1}$}

10: \newcommand\Gyr{\rm{~Gyr}}

11: \newcommand\msun{\rm{M}_\odot}

12: \newcommand\ha{$\rm{H}\alpha$}

13: \newcommand\hb{$\rm{H}\beta$}

14: 

15: %% Emission Line Names

16: \def\oiii{[O\,{\sc iii}]}

17: \def\halpha{\hbox{H$\alpha$}}

18: \def\hbeta{\hbox{H$\beta$}}

19: \def\ni{\hbox{[N\,{\sc i}]}}

20: \def\nii{\hbox{[N\,{\sc ii}]}}

21: \def\hi{\hbox{H\,{\sc i}}}

22: \def\hii{\hbox{H\,{\sc ii}}}

23: \def\nii{[N\,{\sc ii}]}

24: \def\mgb{Mg~{\emph b}}

25: 

26: \shorttitle{Lick Indicies of Thick Disks}

27: \shortauthors{Yoachim \& Dalcanton}

28: 

29: 

30: \begin{document}

31: 

32: 

33: \title{Lick Indices in the Thin and Thick Disks of Edge-On Disk Galaxies}

34: 

35: \author{Peter Yoachim\altaffilmark{1,2} 

36:   \&  Julianne J. Dalcanton\altaffilmark{2,3}}

37: 

38: 

39: \altaffiltext{1}{Department of Astronomy and McDonald Observatory, University of Texas, Austin, TX 78712; {yoachim@astro.as.utexas.edu}}

40: \altaffiltext{2}{Department of Astronomy, University of Washington, Box 351580,

41: Seattle WA, 98195}

42: \altaffiltext{3}{Tom and Margo Wyckoff Fellow}

43: 

44: 

45: 

46: 

47: 

48: \begin{abstract}

49: %

50: We have measured Lick index equivalent widths to derive luminosity

51: weighted stellar ages and metallicities for thin and thick disk

52: dominated regions of 9 edge-on disk galaxies with the ARC 3.5 meter

53: telescope at Apache Point Observatory.  In all cases, the thick disks

54: are confirmed to be old stellar populations, with typical ages between

55: 4 and 10 Gyr.  The thin disks are uniformly younger than the thick

56: disks, and show strong radial age gradients, with the outer regions of

57: the disks being younger than 1 Gyr.  We do not detect any significant

58: metallicity differences or $\alpha$-element enhancement in the thick

59: disk stars compared to the thin disk, due to the insensitivity of the

60: Lick indices to these differences at low metallicity.  We compare

61: these results to thick disks measured in other systems and to

62: predictions from thick disk formation models.

63: %

64: \end{abstract}

65: 

66: \keywords{galaxies: stellar content --- galaxies: spiral --- galaxies: structure}

67: 

68: \section{Introduction}

69: 

70: Old stellar populations preserve a fossil record of a galaxy's early

71: formation and evolution.  In the Milky Way, the stellar thick disk and

72: halo have been recognized as the oldest stellar components and have

73: been studied extensively.  Stars in the MW thick disk are old

74: ($\sim$8-12 Gyr), and are metal poor when compared to local thin disk

75: stars \citep{Reid93, Chiba00}.  Their chemical composition shows that

76: they are enhanced in $\alpha$-elements compared to thin disk stars,

77: suggesting a rapid formation timescale ($<$1 Gyr)

78: \citep{Reddy06,Brewer06,Bensby03,Bensby05, Prochaska00}.

79: 

80: Because stellar thick disks and halos are intrinsically low surface

81: brightness features, they have been observed in detail in only a

82: handful of galaxies.  Thick disks have been photometrically detected

83: as an excess flux at large galactic latitudes across a range of Hubble

84: types \citep{Dalcanton02,Burstein79, Tsikoudi79,deGrijs96, deGrijs97b,

85: Pohlen04,Kruit84, Shaw89, vanDokkum94, Morrison97, Wu02,Abe99,

86: Neeser02}, as well as detected in star counts using resolved stellar

87: populations from HST \citep{Seth05b,Mould05,Tikhonov05,Tikhonovo5b,Tik08}.

88: 

89: Three general classes of formation mechanisms have been put forward to

90: explain thick disks. In the first, a previously thin disk is

91: kinematically heated.  In this scenario, stars form in a thin disk and

92: increase their velocity dispersion with time.  This vertical heating

93: can be rapid, due to interactions and mergers

94: \citep{Quinn93,Walker96,Velazquez99,Chen01,Robin96,Villalobos08} or

95: gradual due to scattering off giant molecular clouds, spiral arms,

96: and/or dark matter substructure

97: \citep{Villumsen85,Hann02,Benson04,Hayashi06,kaz07}.  In the second

98: formation scenario, stars ``form thick'' with star formation occurring

99: above the midplane of the galaxy \citep{Brook04} or form with large

100: initial velocity dispersions in massive stellar clusters

101: \citep{Kroupa02}.  In the final class of models, thick disk stars are

102: directly accreted from satellite galaxies.  Numerical simulations have

103: shown that stars in disrupted satellite galaxies can be deposited onto

104: thick disk like orbits \citep{Abadi203, Martin04, Bekki01, Gilmore02,

105: Navarro04, Statler88,Read08}.  While these models were originally

106: developed to explain the origin of the MW thick disk, they should work

107: equally well for thick disks in other massive galaxies.

108: 

109: On the other hand, some of these mechanisms are likely to be less

110: effective in lower mass galaxies which have lower density disks,

111: little or no spiral structure, and fewer satellites hosting stars.

112: Lower mass galaxies also have different formation times, environments,

113: and gravitational potentials that may also lead to variation in the

114: mass, age, and metallicity of thick disks of low mass galaxies.

115: 

116: To test these formation models, detailed comparisons of thin and thick

117: disk properties are required across a range of galaxy masses.  In

118: particular, the relative ages and chemical enrichment patterns of the

119: thin and thick disks should differ among these formation models.  If

120: the thick disk is just a gradually kinematically heated thin disk,

121: there should be a smooth age and enrichment gradient between the two.

122: In contrast, if the thick disk is formed from accreted stars we should

123: expect the ages and metallicities of the thin and thick disks to be

124: only weakly correlated.  We may also expect to see variations with the

125: mass of the galaxies, with less massive galaxies being more

126: susceptible to external heating and more massive galaxies being better

127: able to tidally disrupt satellites.

128: 

129: Measuring the ages and metallicities of thick disks outside the MW has

130: proved to be challenging.  Ages and metallicities can be derived from

131: isochrone fits to thin and thick disk stars resolved with HST,

132: provided that the host galaxies are sufficiently close and oriented

133: edge-on to the line of sight.  The systems studied this way show older

134: populations at large scale heights but little vertical metallicity

135: gradient, at least for the low mass galaxies ($\rm{V}_{\rm{c}}<100$

136: \kms) which dominate these samples \citep{Seth05b,Tikhonov05,Mould05}.

137: 

138: 

139: For the systems that are farther away, only broadband colors have been

140: used to estimate the ages and metallicities of the thick disks.  When

141: thick disks are photometrically detected, they typically have very red

142: colors ($B-R\sim1.3-1.5$) \citep{Yoachim06, Dalcanton02, vanDokkum94}

143: suggestive of an old population.  However, stellar parameters are

144: notoriously difficult to derive from broadband colors due to the

145: age-metallicity degeneracy in the optical colors, and a lack of

146: IR-colors at the low surface brightnesses of the thick disk region.

147: 

148: Some progress has been made in measuring the

149: metallicity of the thin disk using high signal-to-noise emission lines

150: \citep[e.g.][]{Tremonti04, Zaritsky94, vanzee98}, and in measuring

151: metallicity gradients \citep[e.g.,][]{Zaritsky94,vanzee98}.  While the

152: emission lines studies can constrain the total chemical enrichment a

153: galaxy has experienced, they tell us nothing about the underlying

154: stellar populations in the galaxies.  Emission lines are also unable

155: to provide constraints on the properties of extra-planar stellar

156: populations, which are likely to be dominated by old, dormant stellar

157: populations.

158: 

159: 

160: To better measure the ages and metallicities of thick disks in a

161: larger sample of galaxies, we turn to the integrated spectrum of these

162: galaxies and use the Lick/IDS absorption line system to derive

163: luminosity-weighted stellar population properties.  The Lick indices

164: were originally developed for studying older stellar populations

165: \citep{Burstein84, Faber85}, and have been used extensively in

166: analyzing elliptical galaxies and globular clusters

167: \citep[e.g.,][]{Trager98, Trager00b, SB07}.  \citet{Worthey94} showed

168: that using a combination of age sensitive (Balmer lines) and

169: metallicity sensitive (\mgb\ and Fe) indices, one can lift the

170: age-metallicity degeneracy for a stellar population.  Stellar spectral

171: libraries have now been used to create SSP models over a large range

172: of metallicity and age combinations \citep{Worthey94b,Vazdekis99},

173: including models with variable $\alpha$-element enhancement

174: \citep{Thomas03}.  The tools also now exist to calculate expected Lick

175: indices for composite stellar populations \citep{Bruzual03}.

176: 

177: 

178: Despite the development of stellar synthesis codes that can be

179: extended to younger stellar populations, relatively few studies have

180: attempted to observe Lick indices in disk galaxies.  Studies using

181: tunable filters have been able to detect Mg and Fe index gradients in

182: disks \citep{Beauchamp97,Ryder05}, but have not been combined with

183: measurements of age-sensitive indices.  Studies of disk systems have

184: tended to focus on the high surface brightness bulge components

185: \citep{Moorthy06,Peletier07,Prugniel01,Proctor00}, and fail to reach

186: very far into the disks.  In the most extensive study observing Lick

187: indices in disk galaxies, \citet{MacArthur06} observed Lick indices in

188: 8 galaxy disks, including several late-type galaxies out to $\sim$1

189: scale length.  These observations probed to $\sim$1 scale length.

190: However, all of these galaxies were fairly face-on, preventing thick

191: and thin disk components from being separated.

192: 

193: In this paper, we target regions of edge-on galaxies that are

194: dominated by either the stellar thick disk and thin components.  We then

195: compare ages and metallicities derived from Lick indices both

196: between the two components and from galaxy-to-galaxy.

197: 

198: 

199: 

200: 

201: 

202: 

203: 

204: \section{Observations}

205: 

206: The original sample of 49 galaxies was selected from the Flat Galaxy

207: Catalog \citep{Karachentsev93} and imaged in $B$, $R$, and $K_s$

208: \citep{Dalcanton00}.  \citet{Dalcanton02} used this imaging to

209: demonstrate the ubiquity of thick disks around late-type galaxies.  We

210: have since used two-dimensional decompositions of the galaxy images to

211: measure structural parameters for the thick and thin disks

212: \citep{Yoachim06}.  We have also measured kinematic properties of the

213: thick and thin disks using GMOS on the Gemini telescopes

214: \citep{Yoachim05,Yoachim07inp2}.

215: 

216: For this study, we selected a subset of galaxies where the photometric

217: decompositions suggested we would be able to isolate thin and thick

218: disk regions while obtaining adequate signal.  This limits

219: us to observing predominantly lower-mass galaxies, as

220: \citet{Yoachim06} found these are the galaxies with proportionally

221: larger thick disks.

222: 

223: Observations were made using the Dual Imaging Spectrograph (DIS) on

224: the ARC 3.5 meter telescope at Apache Point Observatory (APO) between

225: June 2003 and February 2006 during a total of 34 half-nights of

226: observing.  Although the telescope aperture is much smaller than used

227: for previous studies (3.5m compared to 8m), we were able to reach the

228: needed signal-to-noise using a novel slit design.  Spectroscopy of

229: extended low surface brightness objects has traditionally been limited

230: by the need to collect a large number of photons while maintaining

231: adequate spectral resolution.  However, since the Lick indices are

232: defined at low spectral resolution ($\sim8$ \AA) and cover a limited

233: wavelength range, it is possible to capture more photons by using a

234: much wider slit to feed a spectrograph with a higher resolution

235: grating.  We therefore built a custom 5\arcsec x 4$^\prime$ slit to

236: allow the largest possible amount of light to reach the detector.  On

237: the blue side, we used a grating with 1200 lines/mm and on the red

238: side, 830 lines/mm.  The resulting spectral resolution was 4.9(7.1)

239: \AA\ FWHM as measured from calibration lamps for the blue(red) chip.

240: The blue chip gave wavelength coverage from 4390-5430 \AA\ sufficient

241: to measure the age sensitive \hb\ and metallicity sensitive Mg,

242: Fe5270, and Fe5335 Lick indices.  The red side covered 6140-7860~\AA,

243: which allowed us to measure the strength of the \ha\ emission line.

244: The CCDs were binned by 2 in the spectral direction to reduce read

245: noise.  Combining a large aperture slit with a high resolution grating

246: is an unconventional setup, but the increased throughput of the wide

247: slit allows us to observe the low surface brightness regions where the

248: thick disk dominates.

249: 

250: The slit camera on DIS allowed us to accurately place the slit by

251: centroiding nearby bright stars.  The slit camera has a plate-scale of

252: 0.298 \arcsec/pixel, and we were typically able to place the slit

253: within 1 pixel of our target region.  Slit camera images were taken

254: throughout long exposures to ensure accurate tracking.  For

255: each galaxy, we gathered spectra at the midplane as well as at several

256: vertical scale heights above the midplane where photometric

257: decompositions imply the majority of flux should be supplied by the

258: thick disk component.  Images of the slit placement are shown in

259: Figure~\ref{slit_place} and the observation log is listed in

260: Table~\ref{fgc_obs}.

261: 

262: We found that there were several prominent skylines near the relevant

263: indices that would slowly fade for several hours after sunset.  Most

264: problematic was a skyline of OH at 5200 \AA, which often contaminated

265: the \mgb\ index.  To avoid the skylines, we took advantage of the APO

266: 3.5m scheduling system which allocates observations in half-night

267: intervals and scheduled most of our observations for the second half

268: of the night.

269: 

270: %\clearpage

271: \begin{figure}

272: \epsscale{.75}

273: \plotone{f1a.eps}

274: \plotone{f1b.eps}

275: \plotone{f1c.eps}

276: \plotone{f1d.eps}\\

277: \plotone{f1e.eps}\\

278: \plotone{f1f.eps}\\

279: \plotone{f1g.eps}\\

280: \plotone{f1h.eps}\\

281: \plotone{f1i.eps}\\

282: \caption{$R$-band images of our galaxy sample with the APO longslit

283: positions overlayed.  All images are stretched to include $20 < \mu_R<

284: 24.2$ \surfb.  \label{slit_place}}

285: \end{figure}

286: %\clearpage

287: 

288: 

289: %{

290: %\plotone{f1d.eps}\\

291: %\plotone{f1e.eps}\\

292: %\plotone{f1f.eps}\\

293: %}

294: %\centerline{Fig. 1. --- Continued.}

295: %\clearpage

296: %{

297: %\plotone{f1g.eps}\\

298: %\plotone{f1h.eps}\\

299: %\plotone{f1i.eps}\\

300: %}

301: %\centerline{Fig. 1. --- Continued.}

302: %\clearpage

303: %\thispagestyle{empty}

304: 

305: 

306: 

307: 

308: \begin{deluxetable*}{ l l c c c c c c c }

309: %\tabletypesize{\small \footnotesize \scriptsize}

310: %\rotate

311: \tablewidth{0pt}

312: %\tablenum{num}

313: %\tablecolumns{num}

314: %\tableheadfrac{num}

315: \tablecaption{Edge-On Galaxy Observations \label{fgc_obs}}

316: \tablehead{  

317: \colhead{Galaxy } & \colhead{Date Range} & \multicolumn{2}{c}{Midplane} & \multicolumn{2}{c}{Offplane} & \multicolumn{2}{c}{Offsets} & \colhead{Scale Length}\\

318: \colhead{ FGC}  & \colhead{Observed} &\colhead{exposures} & 

319:   \colhead{total time (min)} & \colhead{exposures} & \colhead{total time (min)}  & \colhead{arcsec} & \colhead{kpc\tablenotemark{1}} & \colhead{arcsec}}

320: \startdata

321: 227  & 10/2004 to 10/2005  &    7 &    150 &     20 &    480  &4.2  & 1.7 & 10.2 \\

322: 913  & 02/2004 to 03/2004  &    7 &     80 &     32 &    698 &5.1  &  1.5 & 9.0 \\

323: 1285 & 02/2004 to 04/2004  &    8 &    105 &     18 &    510 & 10.2 & 0.9 & 19.7 \\

324: 1440 & 03/2004 to 02/2006  &    4 &     80 &     10 &    270 & 5.8 & 2.0  & 15.9 \\

325: 1642 & 06/2003 to 05/2005  &    5 &    100 &      5 &    150 & 5.1 &  0.9 & 12.5 \\

326: 1948 & 06/2003 to 05/2005  &    5 &     95 &     15 &    405& 6.5 &  1.2  & 12.3 \\

327: 2131 & 06/2005 to 06/2005  &    3 &     45 &     17 &    338&  4.7&  1.0  & 10.0 \\

328: 2369 & 10/2004 to 08/2005  &    5 &    105 &     25 &    521 & 3.6 &  1.0 & 8.7 \\

329: 2548 & 10/2003 to 10/2003  &    5 &     75 &     27 &    536 & 5.6 &  1.5 & 9.9  \\

330: \enddata

331: \tablenotetext{1}{Distances from \citet{Kara00}}

332: \end{deluxetable*}

333: 

334: 

335: 

336: 

337: 

338: 

339: \begin{deluxetable}{ l c c}

340: %\tabletypesize{\small \footnotesize \scriptsize}

341: %\rotate

342: \tablewidth{0pt}

343: %\tablenum{num}

344: %\tablecolumns{num}

345: %\tableheadfrac{num}

346: \tablecaption{Elliptical Galaxies Observed \label{ell_obs}}

347: \tablehead{\colhead{Galaxy} & \colhead{exposures} &\colhead{ total time (min)} }

348: \startdata

349: NGC 1453 & 2 &   25.  \\

350: NGC 1600 & 2 &   20.  \\

351: NGC 2778 & 3 &    9.  \\

352: NGC 3379 & 1 &    5.  \\

353: NGC 5638 & 2 &   10.  \\

354: NGC 6127 & 2 &   20.  \\

355: NGC 6702 & 2 &   10.  \\

356: NGC 6703 & 2 &   20.  \\

357: NGC 7052 & 1 &    5.  

358: \enddata

359: \end{deluxetable}

360: 

361: 

362: 

363: %\clearpage

364: 

365: 

366: 

367: 

368: 

369: 

370: %###########################DATA REDUCTION#########################

371: \section{Data Reduction}

372: Data were processed with standard IRAF/PyRAF routines along with

373: several custom scripts written in IDL.  Reduction steps for the red

374: and blue side were identical unless otherwise noted.  All the frames

375: were bias corrected by subtracting the mean from the overscan region.

376: Any residual bias structure was removed by subtracting a bias frame

377: constructed from 5-10 bias exposures taken every night.  For one half

378: night (Feb 11, 2004 observations of FGC 913), the blue chip suffered

379: from 60 Hz noise which resulted in diagonal streaks across the images.

380: This pattern was removed by shifting and subtracting a single bias

381: frame that had similar variation in the readout pattern.  The images

382: were flat-fielded using spectra of a quartz lamp.  Spectra of the

383: twilight sky showed no need for an illumination correction.  The

384: spectra were wavelength calibrated using He, Ne, and Ar arc lamps

385: along with night-sky emission lines \citep{Oster96}.  Arc spectra were

386: taken interspersed with the science exposures throughout the night,

387: approximately every hour.  Stellar spectra were used to correct the

388: spatial distortions (tilt) in the observations.  Observations of

389: standard stars and standard atmospheric extinction curves were used to

390: flux calibrate the spectra.  Few of the observations were made in

391: photometric conditions, thus our flux calibration is primarily used as

392: a first order removal of the instrumental sensitivity profile.

393: Because we are primarily interested in measuring equivalent widths,

394: the exact flux normalization is not crucial.  The sky was subtracted

395: using a second order polynomial fit to regions dominated by the sky.

396: The spectra were finally corrected for motion relative to the Local

397: Standard of Rest, scaled to a common flux level, spatially aligned,

398: and combined rejecting cosmic ray hits.

399: 

400: Our final spectra have a spatial scale of 0.42 arcsec/pixel on both

401: the red and blue chips, and a wavelength solution of 1.24 \AA/pixel in

402: the blue and 1.68 \AA/pixel in the red.  Measurement of arc lamp lines

403: showed a FHWM resolution of $4.9$ \AA\ on the blue chip and 7.1 \AA\

404: on the red chip.  \citet{Worthey97} report that the resolution (FWHM)

405: of the original Lick indices of interest as 8.4 \AA.  We therefore

406: broaden our blue spectra with a Gaussian kernel with $\sigma=2.9$ \AA\

407: to match the Lick system resolution.

408: 

409: 

410: Systematic and rotational velocities were removed by cross-correlating

411: a (logarithmically-binned) stellar template plus Gaussian emission

412: lines.  The shifts were accurate to within 1 pixel ($\sim 1.2$ \AA).

413: The midplane rotation curve was used for both the midplane and

414: offplane spectra, as any difference between the two should be small at

415: our resolution \citep{Yoachim07inp2}.

416: 

417: 

418: After the 2D spectra were broadened to match the Lick resolution, any

419: foreground stars were masked and 1D spectra were extracted by summing

420: in the spatial direction.  For both the midplane and offplane, we

421: extracted spectra from the region of $\pm$1 scale length ($h_R$), as

422: measured in \citet{Yoachim05}.  

423: 

424: 

425: 

426: 

427: We tested how the SNR of our spectra affects the accuracy of the

428: measured Lick indices.  Adding artificial noise to template stellar

429: spectra, we find the RMS error in a measured Lick index scales with

430: S/N as $\sigma_{index}\sim 9(SNR)^{-1}$\AA.  Therefore, to achieve an

431: Lick EW uncertainty of $\pm0.2$\AA\ requires a SNR of $\sim45$ per

432: \AA, which agrees with previous determinations by

433: \citet{Trager00}.

434: 

435: Throughout our analysis, we will calculate the uncertainties in our

436: derived ages and metallicities using the SNR of the extracted spectra.

437: As an additional check, we have extracted 1D spectra from $0<R/h_R<1$

438: and $-1<R/h-R<0$ to find the variance in the Lick indices from one side

439: of the galaxy to the other and note cases where there are large

440: discrepancies.  This procedure allows us to flag systems where

441: systematic errors are likely to dominate over random

442: uncertainties.

443: 

444: %\clearpage

445: 

446: \begin{figure}

447: \plotone{f2.eps}

448: \caption{Examples of our blue and red spectra.  From top to bottom:

449: Midplane of FGC 1285, Offplane of FGC 1285, Elliptical galaxy NGC

450: 5638, Lick standard star HD114762.  All the blue spectra have been smoothed

451: to a resolution of 8.4 \AA\ and the Lick indices of interest have been

452: labeled.

453: \label{example_spec}}

454: \end{figure}

455: 

456: %\clearpage

457: 

458: \section{Moving Onto the Lick System}

459: 

460: 

461: While we have matched the Lick resolution reported in

462: \citet{Worthey97}, this is not enough to ensure that we are on the

463: Lick system.  Because the Lick indices were originally defined from

464: spectra that had not been flux-calibrated, additional corrections are

465: needed.  To do so, we made 144 observations of 37 unique Lick standard

466: stars.  Stars for which we have repeat observations have a mean RMS

467: error of 0.09 \AA\ for each index.  Our standard star EWs are compared

468: to values listed in \citet{Worthey94}, and are shown in

469: Figure~\ref{resids2}.  The derived zero point corrections are

470: listed in Table~\ref{zp_table}.  In general, the agreement is quite

471: good, with a typical Gaussian scatter of 0.26 \AA\ and little systematic

472: offset.  The notable exception is for \mgb, for which the APO system

473: measures low EW for metal rich systems with \mgb$>3$ \AA.  We do not

474: explicitly correct for this offset as we are observing metal poor

475: galaxies which have \mgb\ indices $<2$ \AA.  We also compare our standard

476: star EWs with stars in common with \citet{Schiavon06} and find

477: a spread of 0.2-0.4 \AA.  \citet{Schiavon06} points out that

478: measurements of EWs of bright stars are disturbingly inconsistent, and

479: that there can be surprisingly large variations between observations.

480: \citet{Schiavon06} argue that these offsets are caused by errors in

481: the flat-fielding, which dominate the errors of the bright stars,

482: unlike fainter galaxy spectra which are background limited.

483: 

484: 

485: %\clearpage

486: 

487: \begin{deluxetable}{c c c c c}

488: %\tabletypesize{\small \footnotesize \scriptsize}

489: %\rotate

490: \tablewidth{0pt}

491: %\tablenum{num}

492: %\tablecolumns{num}

493: %\tableheadfrac{num}

494: \tablecaption{Zero point conversions \label{zp_table}}

495: \tablehead{index& H$\beta$ & \mgb & Fe 5270 & Fe 5335 }

496: \startdata

497:  zero point\tablenotemark{1} (\AA) & 0.13 & 0.00 & -0.01 & -0.05  \\

498:  RMS (\AA) & 0.23 &  0.26 &  0.35 &  0.29 

499: \enddata

500: \tablenotetext{1}{Value subtracted from our measured indices to move onto the Lick system}

501: \end{deluxetable}

502: 

503: %\clearpage

504: 

505: \begin{figure}

506: \plotone{f3.eps}

507: \caption{Comparison of Lick indices measured with DIS compared to the

508: published equivalent widths in \citet{Worthey94}.  Dashed horizontal

509: lines show the mean offsets while the dotted horizontal lines show the

510: average Worthey et al. $\pm 3 \sigma$ uncertainties.  Dashed vertical

511: lines show the range of the line indices measured in our galaxy sample.

512: Corresponding means and scatters are given in Table~\ref{zp_table}.

513: \label{resids2}}

514: \end{figure}

515: 

516: 

517: 

518: 

519: \begin{figure}

520: \plotone{f4.eps}

521: \caption{ Lick indices measured with APO compared with values

522: published in \citet{Trager00}.  Dashed lines show the median offset

523: between our measurements and those in \citet{Trager00}.  The agreement

524: is good, even with our larger slit and lack of velocity dispersion

525: correction.  The large errors for the Fe5335 index values are caused

526: by the feature approaching the DIS dichroic cut-off.

527: \label{ell_resids}}

528: \end{figure}

529: 

530: %\clearpage

531: 

532: We also observed a sample of elliptical galaxies which have reported

533: Lick measurements in the literature.  The ellipticals were observed

534: when light cloud cover made observing faint regions of disks

535: impossible, or during brief periods when the primary targets were not

536: visible.  We observed 9 galaxies in common with \citet{Trager00}

537: listed in Table~\ref{ell_obs}.  Our indices are not directly

538: comparable to the ones listed in \citet{Trager00}, as we used our

539: large 5\arcsec\ slit which samples more flux from the outer regions of

540: the galaxies, and we have not replicated the corrections for velocity

541: dispersion and emission line fill-in as done in \citet{Trager00}.

542: Thus, some amount of scatter is expected.  The residuals between our

543: measurements and those in \citet{Trager00} are plotted in

544: Figure~\ref{ell_resids}.  For the H$\beta$, \mgb, and Fe 5270 indices,

545: we measure RMS scatters of 0.16-0.31 \AA\ around the \citet{Trager00}

546: values.  There is a much larger scatter for Fe 5335 because the

547: feature approaches the DIS dichroic cut-off for galaxies with a

548: redshift greater than 4000 \kms\, and becomes very low

549: signal-to-noise.

550: 

551: 

552: 

553: 

554: \subsection{Emission Line Removal}

555: Unlike observations of gas-poor ellipticals and globular clusters, we must

556: remove any emission lines that contaminate the index passbands before

557: we can derive ages and metallicities from the absorption features.

558: The removal of emission lines is not trivial.  We would like to remove

559: the \hb\ emission, but to do so we would need to know the underlying

560: shape of the stellar continuum.  This shape in turn depends strongly

561: on the stellar population's age and metallicity, which is what we set

562: out to measure.  We therefore must turn to other parts of the spectrum

563: to estimate the amount of \hb\ emission.

564: 

565: Lick indices, particularly \hb, are often contaminated with emission

566: lines.  \citet{Gonzalez93} popularized using the correction

567: \hb$_{emission}$=0.7[O\,{\sc iii}]$\lambda$5007.  This correction was derived

568: by fitting a stellar template to the underlying spectrum of their

569: elliptical galaxies to isolate the emission feature.  Unfortunately,

570: the shape of the template can determine the magnitude of the emission

571: correction and the ratio of $F_{\rm{H}\beta}/F_{[\rm{O}III]}$ is

572: highly metallicity sensitive \citep[e.g.,][]{Kewley02}, making this correction

573: inappropriate for low metallicity systems with varying stellar ages.

574: \citet{Moorthy06} make a correction by fitting a Gaussian to the

575: emission peak inside the \hb\ Lick index.  This correction probably

576: underestimates the \hb\ correction since the Gaussian fit will not be

577: sensitive to the fraction of \hb\ emission that has filled the

578: absorption feature.  This correction is also very sensitive to the

579: resolution of the measured spectrum.  When we try to fit a Gaussian to

580: the emission feature after broadening, we find that the correction is

581: 0.4 \AA\ smaller than when we make the correction before broadening.

582: \citet{MacArthur06} takes the most extreme measure of masking regions

583: where the spectrum is contaminated by emission lines.

584: 

585: 

586: 

587: Using the [O\,{\sc iii}] emission line to estimate \hb\ is a fine

588: approximation when the \hb\ emission correction is small, as in

589: elliptical galaxy spectra.  However, our midplane spectra are clearly

590: dominated by the \hb\ emission (Figure~\ref{example_spec}), and we

591: must use a more accurate technique to remove the emission.  To do so,

592: we measure the EW of the \ha\ emission line from the spectra taken

593: simultaneously with the red DIS spectrograph and use this to estimate

594: the EW of the \hb\ emission line.

595: 

596: Our procedure for removing the \hb\ emission is as follows.  We first

597: assume case B recombination with

598: $F_{\rm{H}\alpha}=2.86F_{\rm{H}\beta}$ \citep{Osterbrock89}.  This

599: correction is used in \citet{Rampazzo05} and \citet{Denicol05} when

600: measuring Lick indices in elliptical galaxies.  However, because the

601: \ha\ and \hb\ emission lines were measured on different CCDs, we are

602: hesitant to use the measured flux ratios.  While both CCDs are

603: calibrated using the same flux standard stars, the subsequent scaling

604: and co-adding of frames was done independently and could skew the

605: absolute flux calibration between the two.  Instead, we make

606: corrections based on equivalent widths as follows. 

607: 

608: The definition of the EW is

609: \begin{equation}

610: {\rm{EW}}=\int_{\lambda_1}^{\lambda_2}\left(1-\frac{F_{I\lambda}}

611: {F_{C\lambda}}\right)d\lambda 

612: \end{equation}

613: where $F_{I\lambda}$ is the spectrum in the index passband and $F_{C\lambda}$ is the continuum spectrum calculated from the flanking pseudo-continuum regions.  Our goal is to remove the contaminating emission feature to recover the equivalent-width of the underlying Lick absorption feature.

614: \begin{equation}

615: \rm{EW_{observed}}=\rm{EW_{Lick}}+\rm{EW_{emission}}

616: \end{equation}

617: 

618:     In the case where the continuum is constant, the EW from the emission alone is given by

619: \begin{equation}

620: {\rm{EW}_{emission}}=\bigtriangleup\lambda-\frac{F_{em}+

621: F_{C\lambda}\bigtriangleup\lambda}{F_{C\lambda}}=

622: -\frac{F_{em}}{F_{C\lambda}}

623: \end{equation}

624: where $F_{em}$ is the integrated flux in the emission line and $F_{C\lambda}$ is the continuum level.  We can then write the expected EW ratio for the Balmer emission lines as

625: \begin{equation}

626: \frac{\rm{EW(H}\alpha)_{em}}{\rm{EW(H}\beta)_{em}}=

627: \frac{F_{\rm{H}\alpha}}{F_{\rm{H}\beta}}

628: \frac{F_{C\lambda,\rm{H}\beta}}{F_{C\lambda,\rm{H}\alpha}}.

629: \end{equation}

630: This in turn can be modified to account for differential extinction between the stars and HII regions.

631: 

632: \begin{equation}

633: \frac{\rm{EW(H}\alpha)_{em}}{\rm{EW(H}\beta)_{em}}=

634: \frac{F_\alpha}{F_\beta}

635: \frac{F_{C\lambda,\beta}}{F_{C\lambda,\alpha}} 

636: \frac{10^{0.4E(B-V)_{gas}k(\rm{H}\beta) - k(\rm{H}\alpha)}}{10^{0.4E(B-V)_{stars}k(\rm{H}\beta) - k(\rm{H}\alpha)}}

637: \end{equation}

638: \citet{Calzetti01} lists $ k ({\rm{H}}\beta) - k ({\rm{H}}\alpha)=1.163$ and finds that  $E(B-V)_{stars}=0.44E(B-V)_{gas}$.  The difference in reddening is due to the geometrically clumpy distribution of HII regions compared to the smoother distribution of stars.  Finally, we get the relation

639: \begin{equation}

640: \frac{\rm{EW(H}\alpha)_{em}}{\rm{EW(H}\beta)_{em}}=

641: \frac{F_{\rm{H}\alpha}}{F_{\rm{H}\beta}}

642: \frac{F_{C\lambda,\rm{H}\beta}}{F_{C\lambda,\rm{H}\alpha}} 

643: 10^{0.26E(B-V)_{gas}}.

644: \end{equation}

645: 

646: For the offplane spectra, we assume there is negligible dust

647: extinction, a fairly flat continuum level

648: ($F_{C\lambda,\rm{H}\beta}\approx F_{C\lambda,\rm{H}\alpha}$), and

649: case B recombination, resulting in the standard correction of

650: ${\rm{EW(H}}\beta)_{em}=y {\rm{EW(H}}\alpha)_{em}$ with $y=2.86$.  In

651: the case of the midplane spectra, there is a younger bluer stellar

652: population and the possibility of dust extinction.  We adopt case B

653: recombination, a continuum ratio of 1.1, and $E(B-V)\sim0.1$ resulting

654: in $y=3.3$.  Our value of $E(B-V)$ is lower than typical values for

655: the centers of edge-on disks (\citet{Matthews99} find $E(B-V)\sim0.2$

656: in an edge-on disk similar to our sample galaxies), however, light off

657: the midplane and at larger radii experience significantly less

658: extinction.  For elliptical galaxies identified as dusty,

659: \citet{Denicol05} adopt a correction of $y=3.0$.

660: 

661: These adopted corrections give results consistent with what we expect

662: from the broadband colors and SSP models.  Most of the corrected \hb\

663: Lick EWs fall within the range expected from the model grids, and we

664: find younger ages where the integrated light is blue and older

665: ages where it is red.  In \S\ref{elcorr}, we calculate how a 10\%

666: change in our adopted value of $y$ would propagate to the derived ages

667: and metallicities.  This uncertainty corresponds to extinction values $0.08 <

668: E(B-V) < 0.42$ or continuum ratios $1.05

669: <F_{C\lambda,\rm{H}\alpha}/F_{C\lambda,\rm{H}\beta} < 1.3$.

670: 

671: 

672: While we consider how dust affects the emission line EW ratios, the

673: Lick absorption features themselves are fairly insensitive to dust

674: \citep{MacArthur05} and require no extra corrections even if the

675: galaxies are dusty.

676: 

677: %\clearpage

678: 

679: \begin{deluxetable*}{ r c c c c c c c   }

680: \tabletypesize{\small}% \footnotesize \scriptsize}

681: %\rotate

682: \tablewidth{0pt}

683: %\tablenum{num}

684: %\tablecolumns{num}

685: %\tableheadfrac{num}

686: \tablecaption{Measured Lick indices in the range $-1 < h_R <1$\label{lick_table}}

687: \tablehead{  

688: \colhead{Galaxy } & \colhead{SNR/Pixel} &  \colhead{\hb$_{\rm{raw}}$} & \colhead{\ha} & \colhead{\hb$_{\rm{corr}}$} & \colhead{\mgb} &\colhead{Fe 5270} & \colhead{Fe 5335}

689: }

690: \startdata

691: FGC1285 midplane &  272.18 &   -0.02$\pm 0.0$ &   -9.82$\pm 0.1$ &    2.92$\pm 0.0$ &    1.16$\pm 0.0$ &

692:    1.45$\pm 0.0$ &    1.54$\pm 0.0$  \\

693:  offplane &   91.21 &    1.68$\pm 0.1$ &   -1.43$\pm 0.1$ &    2.18$\pm 0.1$ &    1.89$\pm 0.1$ &    1.21$\pm 0.1$ &

694:    1.27$\pm 0.1$  \\

695: FGC1440 midplane &  200.25 &   -0.92$\pm 0.0$ &  -11.83$\pm 0.1$ &    2.62$\pm 0.0$ &    2.29$\pm 0.0$ &

696:    1.83$\pm 0.1$ &    1.23$\pm 0.1$  \\

697:  offplane &   36.24 &    0.59$\pm 0.2$ &   -4.28$\pm 0.1$ &    2.09$\pm 0.2$ &    2.63$\pm 0.2$ &    3.36$\pm 0.3$ &

698:    6.75$\pm 0.5$  \\

699: FGC1642 midplane &   76.15 &    0.35$\pm 0.1$ &  -11.28$\pm 0.1$ &    3.73$\pm 0.1$ &    0.78$\pm 0.1$ &

700:    0.68$\pm 0.1$ &    0.40$\pm 0.2$  \\

701:  offplane &   60.95 &   -0.24$\pm 0.1$ &   -8.29$\pm 0.1$ &    2.66$\pm 0.1$ &    1.45$\pm 0.1$ &    1.45$\pm 0.2$ &

702:    0.84$\pm 0.3$  \\

703: FGC1948 midplane &   77.36 &   -2.30$\pm 0.1$ &  -22.40$\pm 0.1$ &    4.41$\pm 0.1$ &    0.53$\pm 0.1$ &

704:   -0.16$\pm 0.1$ &    0.87$\pm 0.2$  \\

705:  offplane &   15.83 &    0.05$\pm 0.5$ &   -8.99$\pm 0.1$ &    3.19$\pm 0.5$ &   -0.20$\pm 0.6$ &   -0.31$\pm 0.6$ &

706:   -1.35$\pm 0.6$  \\

707: FGC2131 midplane &   98.92 &   -3.82$\pm 0.1$ &  -29.31$\pm 0.1$ &    4.96$\pm 0.1$ &    0.68$\pm 0.1$ &

708:    0.94$\pm 0.1$ &    1.82$\pm 0.1$  \\

709:  offplane &   14.71 &    0.16$\pm 0.5$ &   -8.13$\pm 0.1$ &    3.01$\pm 0.5$ &    1.47$\pm 0.6$ &    1.51$\pm 0.7$ &

710:    0.88$\pm 0.7$  \\

711: FGC227 midplane &  104.72 &   -0.58$\pm 0.1$ &  -16.33$\pm 0.1$ &    4.30$\pm 0.1$ &    0.77$\pm 0.1$ &

712:    1.24$\pm 0.1$ &    0.87$\pm 0.1$  \\

713:  offplane &   21.54 &   -1.60$\pm 0.3$ &  -10.25$\pm 0.1$ &    1.98$\pm 0.3$ &    2.03$\pm 0.4$ &    2.45$\pm 0.4$ &

714:    8.97$\pm 0.6$  \\

715: FGC2369 midplane &   51.41 &   -0.70$\pm 0.1$ &  -14.38$\pm 0.1$ &    3.60$\pm 0.1$ &    1.05$\pm 0.1$ &

716:    0.99$\pm 0.2$ &   -0.31$\pm 0.3$  \\

717:  offplane &    9.37 &    1.64$\pm 0.9$ &   -4.84$\pm 0.1$ &    3.33$\pm 0.9$ &    3.43$\pm 0.9$ &    0.19$\pm 0.9$ &

718:    0.33$\pm 1.1$  \\

719: FGC2548 midplane &   81.29 &    0.64$\pm 0.1$ &  -11.25$\pm 0.1$ &    4.00$\pm 0.1$ &   -0.12$\pm 0.1$ &

720:    0.99$\pm 0.1$ &   -0.04$\pm 0.2$  \\

721:  offplane &   28.38 &    0.67$\pm 0.2$ &   -3.66$\pm 0.1$ &    1.95$\pm 0.2$ &    0.12$\pm 0.3$ &   -0.32$\pm 0.4$ &

722:   -1.74$\pm 0.6$  \\

723: FGC913 midplane &   71.87 &   -2.43$\pm 0.1$ &  -22.14$\pm 0.1$ &    4.20$\pm 0.1$ &    1.37$\pm 0.1$ &

724:    1.30$\pm 0.2$ &   -0.79$\pm 0.3$  \\

725:  offplane &   35.48 &   -0.57$\pm 0.2$ &   -9.45$\pm 0.1$ &    2.73$\pm 0.2$ &    2.86$\pm 0.2$ &    1.19$\pm 0.3$ &

726:   -1.94$\pm 0.4$  

727: 

728: \enddata

729: \tablecomments{All Lick index values are measured in angstroms.  Negative values indicate emission.}

730: 

731: \end{deluxetable*}

732: 

733: %\clearpage

734: 

735: %###########################MODEL GRIDS#######################

736: \subsection{Deriving Ages and Metallicities}

737: 

738: 

739: 

740: Ideally, we would fit full star formation and chemical evolution

741: histories for our galaxies.  However, we have chosen to focus on only

742: 4 of the most prominent Lick indices, limiting the total number of

743: parameters we can hope to fit.  We therefore choose to interpret the

744: spectra using SSP models.  While our galaxies are clearly more

745: complicated than single-burst stellar populations (given that we see

746: old stellar populations along with signs of current star formation),

747: the SSP models will still give a reasonable first-order approximation

748: of the luminosity weighted mean ages, metallicities, and

749: $\alpha$-element enhancements.

750: 

751: 

752: 

753: To convert our measured indices to ages we start with the model grids

754: of \citet{Thomas03} and interpolate them to a finer grid with age

755: steps of $\delta t=0.1$ Gyr and metallicity steps of

756: $\delta$[Z/H]=0.01.  We exclude the very young age grid points (age $<

757: 0.1$ Gyr) as the Lick indices expected for such young ages are

758: degenerate with older populations.  To calculate the uncertainties, we

759: interpolate ages and metallicities for 100 Lick index pairs

760: distributed according to the equivalent width uncertainties.  We also

761: test for possible systematic errors caused by incorrect emission line

762: corrections and calculate how a 10\% error in the emission line

763: correction for H$\beta$ propagates to errors in age and metallicity.

764: 

765: Figure~\ref{grid1} shows the Lick indices measured for each of our

766: galaxies, placed on the model grids of \citet{Thomas03}, assuming no

767: $\alpha$-enhancement.  For each galaxy, we calculate the age and

768: metallicity by interpolating onto grids of \hb+\mgb, \hb+Fe 5270, and

769: \hb+Fe 5335.  

770: 

771: Three of the galaxies (FGC 1948, 2369, and 2548) have Lick indices

772: that consistently fall far outside the model grids.  This is not

773: surprising, as the offplane SNR for FGC 1948 and 2369 were so low we

774: should not have expected to measure Lick indices

775: (Table~\ref{lick_table}), and FGC 2548 is only on the cusp of having

776: adequate SNR.  For the remaining six galaxies, we plot the average

777: interpolated ages and metallicities for the midplane and offplane in

778: Figure~\ref{age_met} and plot cumulative distributions of the age and

779: metallicity differences between the thin and thick disk components in

780: Figure~\ref{hists}.  For consistency, we use only the metallicities

781: measured from the \mgb\ and Fe 5270 indices because several galaxies

782: have very low SNR at the Fe 5335 index.

783: 

784: For the 6 galaxies with high SNR, we find the thick disks have a

785: median age of 7.2 Gyr and metallicity of [Z/H]=-0.6.  The thin disks

786: have a similar median metallicity as expected for their low mass, but

787: are uniformly much younger, with a median age of 4.6 Gyr.  This age

788: determination for the thin disk is derived after making very large

789: emission line corrections and represents a flux-weighted average of

790: all the stars within a radius of one scale length.  It is certainly

791: possible that the oldest central regions of the thin disk contain

792: stars whose ages are similar to those of the thick disks.  We return

793: to these results in \S\ref{sec_radgrad} below.

794: 

795: %\clearpage

796: 

797: \begin{figure}

798: \epsscale{.6}

799: \plotone{f5a.eps}

800: \plotone{f5b.eps}

801: \caption{Lick index measurements for our observations.  Top plots:

802: Model grids from \citet{Thomas03} along with our observed points.

803: Blue points are from midplane observations and red points are for

804: offplane measurements measured from spectra extracted from $R<|h_R|$.

805: Ellipses are drawn that encompass indices measured from the regions $0

806: < R < h_R$ and $-h_R < R < 0$ .  Small ellipses imply the measured

807: indices are robust, while large ellipses show that the spectra on

808: either side of the galaxy are not particularly consistent.  Bottom

809: Plots: The interpolated ages and metallicities for all the above

810: indices.  Large filled points are used for spectra from the region

811: $-h_R < R < h_R$ that fall inside the model grids.  The ellipses from

812: the \hb-\mgb\ plane are propagated to the bottom plots.

813: Throughout, blue circles are used for midplane observations while red

814: triangles are used for offplane observations.  Error bars on the

815: points show the statistical uncertainties based on the SNR of the

816: spectra.

817:   \label{grid1} }

818: \end{figure}

819: \clearpage

820: {

821: \plotone{f5c.eps}\\

822: \plotone{f5d.eps}\\

823: }

824: \centerline{Fig. 5. --- Continued.}

825: %\clearpage

826: {

827: \plotone{f5e.eps}\\

828: \plotone{f5f.eps}\\

829: }

830: \centerline{Fig. 5. --- Continued.}

831: %\clearpage

832: {

833: \plotone{f5g.eps}\\

834: \plotone{f5h.eps}\\

835: }

836: \centerline{Fig. 5. --- Continued.}

837: %\clearpage

838: {

839: \plotone{f5i.eps}\\

840: \plotone{f5j.eps}\\

841: }

842: \centerline{Fig. 5. --- Continued.}

843: %\clearpage

844: {

845: \plotone{f5k.eps}\\

846: \plotone{f5l.eps}\\

847: }

848: \centerline{Fig. 5. --- Continued.}

849: %\clearpage

850: 

851: 

852: 

853: \begin{figure}

854: \plotone{f6.eps}

855: \caption{Average ages and metallicities as measured by Lick indices \hb, \mgb, and Fe 5270.  Blue circles show midplane (thin disk) observations while red triangles show offplane (thick disk) observations. \label{age_met}}

856: \end{figure}

857: 

858: 

859: \begin{figure}

860: \plotone{f7.eps}

861: \caption{Cumulative distribution plots showing the differences between the thin disk and thick disk ages and metallicities.  Positive ages mean the thick disk is older while negative metallicity differences mean the thick disk is metal poor in comparison to the thin disk. \label{hists}}

862: \end{figure}

863: 

864: %\clearpage

865: 

866: 

867: \subsection{Radial Gradients}\label{sec_radgrad}

868: 

869: The observations of the midplane of FGC 1440 were deep enough that we

870: can extract radial gradients of the Lick indices.  We extracted

871: spectra by binning 6.3\arcsec\ spatially in a sliding region across

872: the galaxy, and moving the observations onto the Lick system as

873: before.  This gives us a SNR of 200 in the central regions of the

874: galaxy and 50 at 1.3 $h_R$.  The radial variation of the Lick indices

875: along with the interpolated ages and metallicities are plotted in

876: Figures~\ref{rad_ind} and~\ref{rad_age}.  The \hb, \mgb, and Fe 5270

877: indices have radial gradients of 2.8, -1.5, and -1.3 \AA/$h_R$

878: respectively.  These gradients are fairly large for pure disk

879: systems. \citet{MacArthur06} finds similar gradients with similar

880: magnitudes in four late-type face-on systems.  \citet{Moorthy06} also

881: find strong radial gradients, but attribute them to the transition

882: between bulge and disk dominated regions of their galaxies.

883: 

884: 

885: The gradients in the Lick indices are consistent with a large age

886: gradient, with the central region of FGC~1440 having an SSP age of

887: $~\sim12$ Gyr dropping to 2 Gyr after one scale length.  The

888: metallicity gradients, however, are slightly ambiguous.  The

889: metallicity measured from the \mgb\ index shows a mostly flat radial

890: gradient, with a large drop on only one side of the galaxy.  On the

891: other hand the Fe~5270 index shows a decreasing metallicity on both

892: sides of the galaxy.  Fitting a line to the radial data, we find that

893: the \mgb\ index has a metallicity gradient of $-0.40 \pm 0.1$

894: dex/$h_R$.  If we restrict the fit to the region inside one

895: scale length, the \mgb\ metallicity gradient is consistent with

896: zero. The Fe~5270 index reveals a much steeper gradient of $-0.70 \pm

897: 0.1$ dex/$h_R$.  This seems to imply that there is a radially changing

898: level of $\alpha$-enhancement throughout the galaxy, with the central

899: regions being close to solar composition and the outer regions

900: becoming more $\alpha$-enhanced, thereby inflating the metallicity

901: measured from \mgb.  In Figure~\ref{rad_age}, we plot the best fitting

902: $\alpha$-element enhancement and it does appear that the outer regions

903: are $\alpha$-enhanced compared to the central region which is best fit

904: with a nearly solar chemical composition.  These $\alpha$-element

905: measurements should be regarded with caution, as we have fit a model

906: with three free parameters (age, metallicity, and $\alpha$ composition)

907: using only three measurements (\hb, \mgb, and Fe 5270).  There is also

908: the possibility that we should adopt a radially varying emission line

909: correction.  This seems likely, as FGC 1440 hosts a dust lane which

910: becomes less prominent with radius.

911: 

912: While we cannot draw broad conclusions based on a single galaxy, it is

913: clear we detect stronger radial gradients than have been found in

914: other disk systems.  This is even more surprising given that edge-on

915: projection effects should act to smooth any radial population

916: gradients we observe.  We discuss these results further in \S\ref{rcg}.

917: 

918: %\clearpage

919: 

920: \begin{figure}

921: \plotone{f8.eps}

922: \caption{ Radial gradients of the Lick equivalent-widths measured in FGC 1440.  Top left shows the \hb\ index after correcting for emission line fill-in, top right shows \mgb, lower left shows the Fe5270 index, and the lower right shows the uncorrected \hb\ EW as well as the \ha\ EW.  Dotted lines show the uncertainties calculated from the extracted spectra SNR. \label{rad_ind}}

923: \end{figure}

924: 

925: 

926: \begin{figure}

927: \plotone{f9.eps}

928: \caption{ The interpolated age and metallicity measured along the midplane of FGC 1440.  The solid line shows the best fit when the \hb, \mgb, and Fe~5270 indices are simultaneously fit to the best matching \citet{Thomas03} model.  Dashed and dotted lines show the best fit if a solar composition is assumed.  There is a very strong age gradient present.  The \mgb\ index shows a fairly flat metallicity gradient, while the Fe~5270 index shows a radially decreasing metallicity.  The third panel shows the best fitting $\alpha$-enhancement when we simultaneously fit all three indices.  \label{rad_age}}

929: \end{figure}

930: 

931: %\clearpage

932: 

933: 

934: 

935: 

936: \subsection{Possible Uncertainties}

937: \subsubsection{Emission Line Correction}\label{elcorr}

938: 

939: Both the midplane and offplane have prominent emission lines.  For the

940: midplane spectra, we applied \hb\ corrections that had an average of 4.9 \AA,

941: while the offplane corrections had an average of 2.2 \AA.  If we had

942: not take reddening or continuum shape into account and naively used

943: just the case B recombination correction for both the midplane and

944: offplane, we would have derive slightly younger ages for the midplane.

945: This simple correction would place many of the observations near the

946: upper edges of the model grids, suggesting we were under correcting.

947: If instead we assume an even higher amount of dust extinction for the

948: midplane, several of the thin and thick disks would then have similar

949: SSP ages.  However, it would take fairly extreme levels of dust

950: extinction to drive all of the thin disks to have similar ages as the

951: thick disks.  The blue broad-band colors for these galaxies are

952: inconsistent with such high levels of dust.

953: 

954: Propagating an uncertainty of $\pm10\%$ in the \hb\ emission

955: correction results in a median $\mp$0.2 dex shift in metallicity and

956: $\pm$1.9 Gyr in age for the midplane and $\mp$0.1 dex and $\pm$2.8 Gyr

957: for the offplane.

958: 

959: \subsubsection{Cross Contamination}

960: 

961: While we have placed our longslits at regions that should be dominated

962: by the thin or thick components, we expect some thick disk stars to be

963: present in the midplane and vice-versa.  Using the photometric fits in

964: \citet{Yoachim06}, we find that our midplane observations typically

965: contain $\sim$20\% thick disk flux, while the offplane observations

966: have a flux contribution of $\sim$25\% from the thin disk.  As can be

967: seen in Figure~\ref{slit_place}, we did not have large gaps between

968: our slit positions.  Observations made in poor seeing conditions might

969: therefore experience extra cross-contamination as light from the

970: midplane could be scattered into the offplane slit position. This should

971: not be a major problem as we avoided making offplane observations

972: during the worst seeing conditions, but could slightly increase the

973: amount of expected cross-contamination.

974: 

975: \citet{Serra07} study how Lick indices and their derived SSP ages and

976: metallicities are affected when there are multiple stellar populations

977: present.  They find that the derived ages are very sensitive to the

978: youngest stars present, while the metallicity measures are

979: predominantly influenced by any older population.  Contamination could

980: therefore explain the similar metallicities we measure for the thin

981: and thick disk stars.  However, resolved stellar population studies

982: have found small vertical metallicity gradients in low mass galaxies

983: as well \citep{Seth05b}.  If enough thick disk stars contaminate the

984: midplane, our SSP derived metallicities will be slightly biased

985: towards those of the older population, even in the young midplane.  Of

986: course, it is also possible that these low mass galaxies have simply

987: not undergone substantial star formation episodes and thus have not

988: chemically enriched the thin disks above the level of the thick disk.

989: Overall, the effects of cross-contamination would lead us to

990: underestimate the true metallicity differences between the thin and

991: thick disks, but to overestimate the flux-weighted age differences.

992: 

993: 

994: \subsubsection{Complex Stellar Populations}

995: 

996: While we are measuring SSP-equivalent ages and metallicities, it is

997: fairly obvious that the midplanes of disk galaxies have undergone

998: multiple epochs of star formation and are not well described by a

999: single age and metallicity.  Our ages and metallicities are thus best

1000: interpreted as flux-weighted averages across the extracted radial

1001: region, particularly for our midplane spectra where we have evidence

1002: for very steep radial age gradients in some systems.  Because we are

1003: forced to bin our spectra over a large spatial region to reach

1004: adequate SNR, we include flux from the younger outer regions of the

1005: galaxy.  Our midplane ages should probably be interpreted as minimums,

1006: as the central region is undoubtedly older than the flux averaged

1007: measure we report.

1008: 

1009: 

1010: \subsubsection{$\alpha$-element Enhancement}

1011: 

1012: Many spectroscopic observations of elliptical galaxies and spiral

1013: bulges have found stellar populations that have systematic 

1014: differences between the metallicity calculated from the \mgb\ index

1015: compared to the Fe indices.  This systematic shift is usually

1016: interpreted as being caused by a stellar population that is

1017: significantly enhanced with $\alpha$-elements compared to the spectra

1018: that were used in building the Lick model grids.  Such an enhancement

1019: is expected for stellar populations that form rapidly ($<1$ Gyr) and

1020: that are primarily enriched by Type II supernovae \citep{Matteucci94}.

1021: Enhancements in $\alpha$-elements are often seen in elliptical

1022: galaxies \citep{Worthey92,Fisher96,Thomas03}, as well as in local MW

1023: thick disk stars \citep{Bensby05}.  

1024: 

1025: The galaxies in our sample are all

1026: fairly low mass and therefore also low metallicity.  In the low

1027: metallicity regime, the signature of $\alpha$-element enhancement

1028: becomes weaker in the Lick indices.  Unlike massive elliptical

1029: galaxies where the metallicity indicators can show systematic offsets

1030: of $\sim$0.5 dex for an $\alpha$-enhanced population, our galaxies are

1031: all sub-solar metallicity and thus would show little bias even if they

1032: are $\alpha$-enhanced.  If we used model SSP grids with

1033: [$\alpha$/Fe]=0.3, our derived metallicities would change by only

1034: $\sim0.1$ dex.

1035: 

1036: We are hesitant to use our data to fit the $\alpha$-element

1037: enhancement level.  If we forge ahead and do so, we find considerable

1038: spread between the metallicities returned, but neither the thin or

1039: thick disk have systematically larger metallicities returned from the

1040: \mgb\ index, as we would expect if the stars were $\alpha$-enhanced.

1041: However, given the small expected offset and lower SNR than available

1042: for elliptical galaxy spectra, we do not consider this a significant

1043: result, and include it here only for completeness.

1044: 

1045: 

1046: 

1047: \section{Discussion}

1048:   

1049: 

1050: \subsection{Are Low Mass Thick Disks Old?}

1051: 

1052: Measuring accurate ages for thick disk stars has been done in

1053: relatively few systems.  In the MW, stars that are kinematically

1054: identified as thick disk stars typically have ages greater than 8 Gyr

1055: \citep{Fuhrmann98,Bensby04b}.  HST studies of resolved stellar

1056: populations in nearby galaxies show the offplane regions are dominated

1057: by old stars.  \citet{Seth05b} find that in 8 galaxies the offplane

1058: RGB stars have ages in the range of 2-6 Gyr.  Similarly,

1059: \citet{Mould05} uses the ratio of RGB and AGB stars to find ages of

1060: thick disks in a sample of 4 galaxies to be older than 3 Gyr.  Our

1061: measured thick disk SSP ages fall between 3.8 and 10.9 Gyr with a

1062: median age of 7.1 Gyr, consistent with these other studies that show

1063: thick disks to be dominated by ancient stars.

1064: 

1065: 

1066: \subsection{Are Low Mass Thick Disks Metal Poor?}

1067: 

1068: Many of our thick disks appear to be more metal rich than the embedded

1069: thin disks.  This counter-intuitive result is probably due to the flux

1070: weighted nature of our measurement.  The young, metal-poor outer

1071: regions dilute the true central ages and metallicities of the thin

1072: disks.

1073: 

1074: Thick disk metallicities have only been measured for a handful of

1075: systems.  MW thick disk stars typically have metallicities in the

1076: range [Fe/H]$\sim-0.7$ to -0.2, with the highest metallicity thick

1077: disk stars possibly reaching solar values \citep{Bensby06}.  One

1078: difficulty with comparing to the MW thick disk is that the observed

1079: MW properties are of thick disk stars near the solar radius, while we

1080: have only been able to measure thick disk properties near the central

1081: regions of the galaxies.  Fortunately there are signs that the MW

1082: thick disk has relatively small age and metallicity gradients

1083: \citep{Bensby05}.

1084: 

1085: Constraints of thick disk metallicities have also been derived from

1086: HST studies imaging resolved stars.  \citet{Seth05b} used ACS images

1087: of 6 nearby edge-on disks to constrain vertical gradients in the

1088: stellar populations using the color and distribution of AGB and RGB

1089: stars.  The older RGB stars have a systematically larger scale height

1090: compared to the younger AGB and main sequence stars.  They find little

1091: to no metallicity gradients in the thick disk stars in their systems,

1092: with the metallicities of the thick disks peaked around

1093: [Fe/H]$\sim-1$.  This is slightly more metal poor than the

1094: measurements we have for our thick disks.  However, the offset is

1095: unlikely to be significant, given that the \citet{Seth05b} study is

1096: able to study a cleaner sample of thick disk stars by reaching higher

1097: vertical heights which reducing the contamination of thin disk stars.

1098: Like the data presented here, \citet{Seth05b} only studies lower mass

1099: systems, limiting the amplitude of any possible metallicity gradient

1100: due to the low metallicity of the midplane.  Using similar HST

1101: observations, \citet{Mould05} finds that thick disk stars in 4 edge-on

1102: galaxies have [Fe/H] between -1.0 and -0.78, again very similar to the

1103: metallicities we find.

1104: 

1105: In Figure~\ref{comp}, we compare our thin disk metallicities to the

1106: low mass sample in \citet{Lee06b} and the large SDSS sample of

1107: \citet{Tremonti04}.  We also compare our thick disk values to thick

1108: disk and halo samples presented in

1109: \citet{Seth05b,Mouhcine05,Tikhonov05}, and \citet{Reddy06}.  In cases

1110: where the authors presented values of log(O/H)+12, we converted to

1111: [Fe/H] assuming log(O/H)$_\odot$+12=8.69 and [Fe/O]=0.  We also plot

1112: the midplane nebular abundances for our galaxies calculated from the

1113: S2N2 calibrator \citep{Viironen07}.  With the exception of a few

1114: outliers, our measured metallicities are consistent with metallicities

1115: measured in similar systems.

1116: 

1117: 

1118: Recently, \citet{Ivezic08} have questioned if the Milky Way thick disk

1119: is a unique component or simply an extension of the thin disk that has

1120: non-Gaussian metallicity and velocity distributions \citep{Norris87}.

1121: They cite a lack of correlation between velocity and metallicity in

1122: large SDSS samples as a major problem for traditional disk

1123: decompositions and rule out a ``traditional" two-disk model at the

1124: 8$\sigma$ level.  Unfortunately, the model \citet{Ivezic08} rule out

1125: is not applicable to their observations.  In particular, they model

1126: 1,142 stars observed in the region $1.0 < z/kpc < 1.2$.

1127: \citet{Ivezic08} correctly assume the observations will contain a

1128: similar number of thin and thick disk stars, however they do not use a

1129: realistic thin disk component.  The thin disk stars are modeled as

1130: having an asymmetric drift of 9 \kms and [Fe/H]=-0.50 with a spread of

1131: 0.04 dex (their Figure~16).  These parameters would be appropriate for

1132: modeling nearby thin disk stars, but they are observing stars 3-4 thin

1133: disk scale heights above the Galactic plane.  At this large height,

1134: only the kinematically hottest thin disk stars will be present in the

1135: sample, and one should expect a much larger velocity lag than observed

1136: in local cooler thin disk stars.  \citet{Holmberg07} show that the

1137: hottest local disk stars are the oldest and that the age-metallicity

1138: relation (AMR) for local stars has large intrinsic dispersion at large

1139: ages ($\sigma\sim0.2$ dex).  This large metallicity dispersion is

1140: probably a result of radial migration of the older high velocity

1141: dispersion stars \citep{Haywood08}.

1142: 

1143: If the offplane thin disk stars are assumed to be similar to older

1144: local thin disk stars (i.e., kinematically hot with a broad

1145: metallicity distribution and larger asymmetric drift), their properties

1146: should well match SDSS observations of regions at large scale height.

1147: Specifically, there should be little correlation between kinematics

1148: and metallicity despite the mixture of thin and thick disk stars.

1149: This revised model is consistent with the observations of thick disks

1150: presented here and elsewhere \citep{Seth05b} that find stars at large

1151: scale heights are a significantly older population than those found

1152: near the midplane.

1153: 

1154: 

1155: 

1156: %\clearpage

1157: 

1158: \begin{figure}

1159: \plotone{f10.eps}

1160: \caption{Comparison of our thin and thick disk metallicities to

1161: similar studies.  The panel on the left shows our thin disk

1162: metallicities along with the low mass galaxies in \citet{Lee06b} and

1163: the fits to larger samples in \citet{Tremonti04} and

1164: \citet{Yoachim07b}. On the right, we compare our thick disk

1165: metallicities to the systems measured in \citet{Seth05b},

1166: \citet{Mouhcine05}, \citet{Reddy06}, and

1167: \citet{Tikhonov05}. \label{comp} }

1168: \end{figure}

1169: 

1170: %\clearpage

1171: 

1172: 

1173: \subsection{Are Thick Disks $\alpha$-enhanced?}

1174: 

1175: The level of $\alpha$-enhancement can be a major clue to the formation

1176: process of a stellar population.  If $\alpha$-enhanced, it is a sign

1177: that a stellar population has been enriched mostly over a short time

1178: period by Type II SNe, whereas stars with solar composition formed

1179: over an extended period and have been enriched by both Type II and

1180: Type Ia SNe.  Numerous papers have found that MW thick disk stars are

1181: enhanced in $\alpha$-elements compared to thin disk stars at similar

1182: total metallicities \citep[e.g.,][]{Bensby03, Bensby04a, Bensby05,

1183: Feltzing03, Reddy06, Taut01,mashonkina03, Prochaska00, Fuhrmann98,

1184: Fuhrmann04}.  Unfortunately, we do not have the SNR to definitively

1185: say if our thick disks are $\alpha$-enhanced, due to the overall low

1186: metallicities of our target galaxies.

1187: 

1188: 

1189: 

1190: \subsection{Are Thick Disks ``Normal'' Stellar Populations?}

1191: 

1192: 

1193: Using a sample of $>$1000 SDSS images of edge-on disk galaxies,

1194: \citet{Zibetti04} examine the faint halo that appears when the images

1195: are stacked.  This extended halo has anomalous colors, requiring stars

1196: that are either metal rich or have a bottom heavy IMF

1197: \citep{Zibetti04,Zackrisson06}.  We find no such anomalies with the

1198: thick disks we observe spectroscopically, as most of them fall on

1199: stellar synthesis model grids using standard IMFs and metallicities.

1200: The few galaxies where we measure Lick indices that are inconsistent

1201: with the SSP models can easily be explained as spurious measurements

1202: caused by low SNR, and do not require exotic stellar populations.

1203: 

1204: \subsection{Radial Color Gradients in the Thin Disks}\label{rcg}

1205: 

1206: \citet{Bell00} observed broadband colors for a large sample of

1207: galaxies and found that the radial gradients were predominantly caused

1208: by age gradients in the stellar populations.  By averaging their

1209: sample together, They found a metallicity gradient, but broadband

1210: sensitivity to dust makes this a measurement in individual galaxies.

1211: Their work has been expanded on by \citet{MacArthur04}, who found both

1212: metallicity and age gradients are stronger in the inner regions of

1213: galaxies, and that galaxies with strong age gradients had smaller

1214: metallicity gradients.  The Lick indices ability to lift the

1215: age-metallicity degeneracy, and relative insensitivity to dust makes

1216: it much easier to quantify how much of the radial color gradients in

1217: disk galaxies are due to age versus metallicity changes.

1218: 

1219: 

1220: In one of the only study that has explicitly targeted Lick indices in the

1221: disk-dominated regions of galaxies, \citet{MacArthur06} detect

1222: age gradients in only 2 of their 8 galaxies, and find strong negative

1223: metallicity gradients in 4.  The age gradients in \citet{MacArthur06}

1224: are also rather small (-0.5 and -1.3 Gyr/$h_R$).  Our measurements of

1225: FGC 1440 show a much steeper age gradient, with the SSP age dropping

1226: by $\sim9$ Gyr over one scale length.

1227: 

1228: \citet{Ganda07} use the SAURON integral field unit spectrograph to

1229: measure Lick index strengths across the face of 18 late-type disk

1230: galaxies.  They find in general \hb\ increasing with radius and the

1231: metal sensitive lines decreasing with radius.  The SAURON observations

1232: show a very wide range of galaxy-to-galaxy radial behavior, the most

1233: extreme are consistent with the strong gradients we find in FGC 1440.

1234: 

1235: Our finding that the thin disk of FGC 1440 might have stellar

1236: populations of near solar composition and be $\alpha$-enhanced at

1237: larger radii is puzzling.  Looking at the transition between bulge and

1238: disk dominated regions, \citet{Moorthy06} finds that the central

1239: bulges are either solar-composition or $\alpha$-enhanced, with little

1240: to no $\alpha$-enhancement in the disks.  With the presence of old

1241: thick disks in all of the galaxies, we would expect the galaxies to

1242: have experienced plenty of chemical evolution and enrichment from SNe

1243: Ia.  Instead, the outer regions of FGC 1440 are $\alpha$-enhanced,

1244: suggesting that the central region of the galaxy has undergone

1245: extended chemical enrichment, while the outer regions have not, despite

1246: being surrounded by old thick disk stars.

1247: 

1248: This could be a sign that the thick disk stars in FGC 1440 have been

1249: recently accreted, and thus have not contributed to the chemical

1250: enrichment the galaxy.  Another possibility is that the central region

1251: of the galaxy is the only place where the gravitational potential is

1252: deep enough to retain SN ejecta, and the outer disk has historically

1253: suffered from SN blow-out and failed to retain metal enriched gas.

1254: Another possibility is that the luminosity-weighted metallicity in the

1255: outer disk is dominated by enrichment from the latest burst of

1256: star-formation which has $\alpha$-enhanced the region.

1257: 

1258: 

1259: 

1260: 

1261: 

1262: 

1263: 

1264: \section{Conclusions}

1265: 

1266: We have spectroscopically confirmed that the thick disks observed in

1267: edge-on late type galaxies are old, metal-poor stellar populations,

1268: analogous to the thick disk stars seen in the MW and nearby edge-on

1269: systems.  This is the first time ages and metallicities of thick disks

1270: have been measured in unresolved stellar populations.  Because all of

1271: our targets are fairly low mass, we are unable to detect any

1272: significant differences between thin disk and thick disk

1273: metallicities.  After correcting for emission line contamination, the

1274: thin disks in our sample are found to be quite young, with strong

1275: radial age gradients.

1276: 

1277: We fail to detect any significant trend for thick disk stars to be

1278: enhanced in $\alpha$-elements compared to their thin disks which is a

1279: defining characteristic of the MW thick disk.  Our failure to observe

1280: $\alpha$-enhancements is most likely a result of our sample being

1281: dominated by low-mass and therefore low metallicity galaxies, for

1282: which solar and $\alpha$-enhanced models are similar.

1283: 

1284: For one galaxy in our sample we have measured the radial gradients of

1285: the Lick indices in the thin disk and find the large gradients that

1286: are dominated by changes in the average stellar age with a small

1287: contribution from a changing average metallicity.

1288: 

1289: 

1290: 

1291: 

1292: 

1293: %--------------------------------------

1294: \acknowledgments

1295: 

1296: We thank Connie Rockosi and the UW machine shop for helping in the

1297: design and manufacture of our slit.  We also thank the APO observing

1298: specialists for their help executing the observations.  We thank

1299: Suzanne Hawley for reading an early version of this paper and making

1300: helpful comments.  JJD and PY were partially supported through NSF

1301: grant CAREER AST-0238683 and the Alfred P.\ Sloan Foundation.  Based

1302: on observations obtained with the Apache Point Observatory 3.5-meter

1303: telescope, which is owned and operated by the Astrophysical Research

1304: Consortium.

1305: 

1306: 

1307: 

1308: 

1309: %\bibliography{../../../Papers/Bib_files/big_jabref}

1310: 

1311: \begin{thebibliography}{108}

1312: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi

1313: 

1314: \bibitem[{Abadi {et~al.}(2003)Abadi, Navarro, Steinmetz, \& Eke}]{Abadi203}

1315: Abadi, M.~G., Navarro, J.~F., Steinmetz, M., \& Eke, V.~R. 2003, \apj, 597, 21

1316: 

1317: \bibitem[{Abe {et~al.}(1999)Abe, Bond, Carter, Dodd, Fujimoto, Hearnshaw,

1318:   Honda, Jugaku, Kabe, Kilmartin, Koribalski, Kobayashi, Masuda, Matsubara,

1319:   Miyamoto, Muraki, Nakamura, Nankivell, Noda, Pennycook, Pipe, Rattenbury,

1320:   Reid, Rumsey, Saito, Sato, Sato, Sekiguchi, Sullivan, Sumi, Watase,

1321:   Yanagisawa, Yock, \& Yoshizawa}]{Abe99}

1322: Abe, F., Bond, I.~A., Carter, B.~S., Dodd, R.~J., Fujimoto, M., Hearnshaw,

1323:   J.~B., Honda, M., Jugaku, J., Kabe, S., Kilmartin, P.~M., Koribalski, B.~S.,

1324:   Kobayashi, M., Masuda, K., Matsubara, Y., Miyamoto, M., Muraki, Y., Nakamura,

1325:   T., Nankivell, G.~R., Noda, S., Pennycook, G.~S., Pipe, L.~Z., Rattenbury,

1326:   N.~J., Reid, M., Rumsey, N.~J., Saito, T., Sato, H., Sato, S., Sekiguchi, M.,

1327:   Sullivan, D.~J., Sumi, T., Watase, Y., Yanagisawa, T., Yock, P.~C.~M., \&

1328:   Yoshizawa, M. 1999, \aj, 118, 261

1329: 

1330: \bibitem[{Beauchamp \& Hardy(1997)}]{Beauchamp97}

1331: Beauchamp, D., \& Hardy, E. 1997, \aj, 113, 1666

1332: 

1333: \bibitem[{Bekki \& Chiba(2001)}]{Bekki01}

1334: Bekki, K., \& Chiba, M. 2001, \apj, 558, 666

1335: 

1336: \bibitem[{Bell \& de~Jong(2000)}]{Bell00}

1337: Bell, E.~F., \& de~Jong, R.~S. 2000, \mnras, 312, 497

1338: 

1339: \bibitem[{Bensby {et~al.}(2003)Bensby, Feltzing, \& Lundstr\"om}]{Bensby03}

1340: Bensby, T., Feltzing, S., \& Lundstr\"om, I. 2003, \aap, 410, 527

1341: 

1342: \bibitem[{Bensby {et~al.}(2004{\natexlab{a}})Bensby, Feltzing, \&

1343:   Lundstr\"om}]{Bensby04a}

1344: ---. 2004{\natexlab{a}}, \aap, 415, 155

1345: 

1346: \bibitem[{Bensby {et~al.}(2004{\natexlab{b}})Bensby, Feltzing, \&

1347:   Lundstr\"om}]{Bensby04b}

1348: ---. 2004{\natexlab{b}}, \aap, 421, 969

1349: 

1350: \bibitem[{Bensby {et~al.}(2005)Bensby, Feltzing, Lundstr\"om, \&

1351:   Ilyin}]{Bensby05}

1352: Bensby, T., Feltzing, S., Lundstr\"om, I., \& Ilyin, I. 2005, \aap, 433, 185

1353: 

1354: \bibitem[{{Bensby} {et~al.}(2007){Bensby}, {Zenn}, {Oey}, \&

1355:   {Feltzing}}]{Bensby06}

1356: {Bensby}, T., {Zenn}, A.~R., {Oey}, M.~S., \& {Feltzing}, S. 2007, Astronomical

1357:   Society of the Pacific Conference Series, 374, 181

1358: 

1359: \bibitem[{Benson {et~al.}(2004)Benson, Lacey, Frenk, Baugh, \& Cole}]{Benson04}

1360: Benson, A.~J., Lacey, C.~G., Frenk, C.~S., Baugh, C.~M., \& Cole, S. 2004,

1361:   \mnras, 351, 1215

1362: 

1363: \bibitem[{Brewer \& Carney(2006)}]{Brewer06}

1364: Brewer, M.-M., \& Carney, B.~W. 2006, \aj, 131, 431

1365: 

1366: \bibitem[{Brook {et~al.}(2004)Brook, Kawata, Gibson, \& Freeman}]{Brook04}

1367: Brook, C.~B., Kawata, D., Gibson, B.~K., \& Freeman, K.~C. 2004, \apj, 612, 894

1368: 

1369: \bibitem[{Bruzual \& Charlot(2003)}]{Bruzual03}

1370: Bruzual, G., \& Charlot, S. 2003, \mnras, 344, 1000

1371: 

1372: \bibitem[{Burstein(1979)}]{Burstein79}

1373: Burstein, D. 1979, \apj, 234, 829

1374: 

1375: \bibitem[{Burstein {et~al.}(1984)Burstein, Faber, Gaskell, \&

1376:   Krumm}]{Burstein84}

1377: Burstein, D., Faber, S.~M., Gaskell, C.~M., \& Krumm, N. 1984, \apj, 287, 586

1378: 

1379: \bibitem[{{Calzetti}(2001)}]{Calzetti01}

1380: {Calzetti}, D. 2001, \pasp, 113, 1449

1381: 

1382: \bibitem[{Chen {et~al.}(2001)Chen, Stoughton, Smith, Uomoto, Pier, Yanny,

1383:   Ivezi\'c, York, Anderson, Annis, Brinkmann, Csabai, Fukugita, Hindsley,

1384:   Lupton, Munn, \& the SDSS~Collaboration}]{Chen01}

1385: Chen, B., Stoughton, C., Smith, J.~A., Uomoto, A., Pier, J.~R., Yanny, B.,

1386:   Ivezi\'c, v.~Z., York, D.~G., Anderson, J.~E., Annis, J., Brinkmann, J.,

1387:   Csabai, I., Fukugita, M., Hindsley, R., Lupton, R., Munn, J.~A., \& the

1388:   SDSS~Collaboration. 2001, \apj, 553, 184

1389: 

1390: \bibitem[{Chiba \& Beers(2000)}]{Chiba00}

1391: Chiba, M., \& Beers, T.~C. 2000, \aj, 119, 2843

1392: 

1393: \bibitem[{Dalcanton \& Bernstein(2000)}]{Dalcanton00}

1394: Dalcanton, J.~J., \& Bernstein, R.~A. 2000, \aj, 120, 203

1395: 

1396: \bibitem[{Dalcanton \& Bernstein(2002)}]{Dalcanton02}

1397: ---. 2002, \aj, 124, 1328

1398: 

1399: \bibitem[{de~Grijs \& Peletier(1997)}]{deGrijs97b}

1400: de~Grijs, R., \& Peletier, R.~F. 1997, \aap, 320, L21

1401: 

1402: \bibitem[{de~Grijs \& van~der Kruit(1996)}]{deGrijs96}

1403: de~Grijs, R., \& van~der Kruit, P.~C. 1996, \aaps, 117, 19

1404: 

1405: \bibitem[{{Denicol{\'o}} {et~al.}(2005){Denicol{\'o}}, {Terlevich},

1406:   {Terlevich}, {Forbes}, {Terlevich}, \& {Carrasco}}]{Denicol05}

1407: {Denicol{\'o}}, G., {Terlevich}, R., {Terlevich}, E., {Forbes}, D.~A.,

1408:   {Terlevich}, A., \& {Carrasco}, L. 2005, \mnras, 356, 1440

1409: 

1410: \bibitem[{{Faber} {et~al.}(1985){Faber}, {Friel}, {Burstein}, \&

1411:   {Gaskell}}]{Faber85}

1412: {Faber}, S.~M., {Friel}, E.~D., {Burstein}, D., \& {Gaskell}, C.~M. 1985,

1413:   \apjs, 57, 711

1414: 

1415: \bibitem[{Feltzing {et~al.}(2003)Feltzing, Bensby, \& Lundstr\"om}]{Feltzing03}

1416: Feltzing, S., Bensby, T., \& Lundstr\"om, I. 2003, \aap, 397, L1

1417: 

1418: \bibitem[{Fisher {et~al.}(1996)Fisher, Franx, \& Illingworth}]{Fisher96}

1419: Fisher, D., Franx, M., \& Illingworth, G. 1996, \apj, 459, 110

1420: 

1421: \bibitem[{Fuhrmann(1998)}]{Fuhrmann98}

1422: Fuhrmann, K. 1998, \aap, 338, 161

1423: 

1424: \bibitem[{Fuhrmann(2004)}]{Fuhrmann04}

1425: ---. 2004, Astronomische Nachrichten, 325, 3

1426: 

1427: \bibitem[{{Ganda} {et~al.}(2007){Ganda}, {Peletier}, {McDermid},

1428:   {Falc{\'o}n-Barroso}, {de Zeeuw}, {Bacon}, {Cappellari}, {Davies},

1429:   {Emsellem}, {Krajnovi{\'c}}, {Kuntschner}, {Sarzi}, \& {van de

1430:   Ven}}]{Ganda07}

1431: {Ganda}, K., {Peletier}, R.~F., {McDermid}, R.~M., {Falc{\'o}n-Barroso}, J.,

1432:   {de Zeeuw}, P.~T., {Bacon}, R., {Cappellari}, M., {Davies}, R.~L.,

1433:   {Emsellem}, E., {Krajnovi{\'c}}, D., {Kuntschner}, H., {Sarzi}, M., \& {van

1434:   de Ven}, G. 2007, \mnras, 380, 506

1435: 

1436: \bibitem[{Gilmore {et~al.}(2002)Gilmore, Wyse, \& Norris}]{Gilmore02}

1437: Gilmore, G., Wyse, R.~F.~G., \& Norris, J.~E. 2002, \apjl, 574, L39

1438: 

1439: \bibitem[{Gonz{\'a}lez(1993)}]{Gonzalez93}

1440: Gonz{\'a}lez, J.~J. 1993, Ph.D.~Thesis

1441: 

1442: \bibitem[{H{\"a}nninen \& Flynn(2002)}]{Hann02}

1443: H{\"a}nninen, J., \& Flynn, C. 2002, \mnras, 337, 731

1444: 

1445: \bibitem[{{Hayashi} \& {Chiba}(2006)}]{Hayashi06}

1446: {Hayashi}, H., \& {Chiba}, M. 2006, \pasj, 58, 835

1447: 

1448: \bibitem[{{Haywood}(2008)}]{Haywood08}

1449: {Haywood}, M. 2008, ArXiv e-prints, 0805.1822

1450: 

1451: \bibitem[{{Holmberg} {et~al.}(2007){Holmberg}, {Nordstr{\"o}m}, \&

1452:   {Andersen}}]{Holmberg07}

1453: {Holmberg}, J., {Nordstr{\"o}m}, B., \& {Andersen}, J. 2007, \aap, 475, 519

1454: 

1455: \bibitem[{{Ivezic} {et~al.}(2008){Ivezic}, {Sesar}, {Juric}, {Bond},

1456:   {Dalcanton}, {Rockosi}, {Yanny}, {Newberg}, {Beers}, {Allende Prieto},

1457:   {Wilhelm}, {Lee}, {Sivarani}, {Norris}, {Bailer-Jones}, {Re Fiorentin},

1458:   {Schlegel}, {Uomoto}, {Lupton}, {Knapp}, {Gunn}, {Covey}, {Allyn Smith},

1459:   {Miknaitis}, {Doi}, {Tanaka}, {Fukugita}, {Kent}, {Finkbeiner}, {Munn},

1460:   {Pier}, {Quinn}, {Hawley}, {Anderson}, {Kiuchi}, {Chen}, {Bushong}, {Sohi},

1461:   {Haggard}, {Kimball}, {Barentine}, {Brewington}, {Harvanek}, {Kleinman},

1462:   {Krzesinski}, {Long}, {Nitta}, {Snedden}, {Lee}, {Harris}, {Brinkmann},

1463:   {Schneider}, \& {York}}]{Ivezic08}

1464: {Ivezic}, Z., {Sesar}, B., {Juric}, M., {Bond}, N., {Dalcanton}, J., {Rockosi},

1465:   C.~M., {Yanny}, B., {Newberg}, H.~J., {Beers}, T.~C., {Allende Prieto}, C.,

1466:   {Wilhelm}, R., {Lee}, Y.~S., {Sivarani}, T., {Norris}, J.~E., {Bailer-Jones},

1467:   C.~A.~L., {Re Fiorentin}, P., {Schlegel}, D., {Uomoto}, A., {Lupton}, R.~H.,

1468:   {Knapp}, G.~R., {Gunn}, J.~E., {Covey}, K.~R., {Allyn Smith}, J.,

1469:   {Miknaitis}, G., {Doi}, M., {Tanaka}, M., {Fukugita}, M., {Kent}, S.,

1470:   {Finkbeiner}, D., {Munn}, J.~A., {Pier}, J.~R., {Quinn}, T., {Hawley}, S.,

1471:   {Anderson}, S., {Kiuchi}, F., {Chen}, A., {Bushong}, J., {Sohi}, H.,

1472:   {Haggard}, D., {Kimball}, A., {Barentine}, J., {Brewington}, H., {Harvanek},

1473:   M., {Kleinman}, S., {Krzesinski}, J., {Long}, D., {Nitta}, A., {Snedden}, S.,

1474:   {Lee}, B., {Harris}, H., {Brinkmann}, J., {Schneider}, D.~P., \& {York},

1475:   D.~G. 2008, ArXiv e-prints, 0804.3850

1476: 

1477: \bibitem[{Karachentsev {et~al.}(2000)Karachentsev, Karachentseva, Kudrya,

1478:   Makarov, \& Parnovsky}]{Kara00}

1479: Karachentsev, I.~D., Karachentseva, V.~E., Kudrya, Y.~N., Makarov, D.~I., \&

1480:   Parnovsky, S.~L. 2000, Bull.~Special Astrophys.~Obs., 50, 5

1481: 

1482: \bibitem[{Karachentsev {et~al.}(1993)Karachentsev, Karachentseva, \&

1483:   Parnovskij}]{Karachentsev93}

1484: Karachentsev, I.~D., Karachentseva, V.~E., \& Parnovskij, S.~L. 1993,

1485:   Astronomische Nachrichten, 314, 97

1486: 

1487: \bibitem[{{Kazantzidis} {et~al.}(2007){Kazantzidis}, {Bullock}, {Zentner},

1488:   {Kravtsov}, \& {Moustakas}}]{kaz07}

1489: {Kazantzidis}, S., {Bullock}, J.~S., {Zentner}, A.~R., {Kravtsov}, A.~V., \&

1490:   {Moustakas}, L.~A. 2007, ArXiv e-prints, 708

1491: 

1492: \bibitem[{Kewley \& Dopita(2002)}]{Kewley02}

1493: Kewley, L.~J., \& Dopita, M.~A. 2002, \apjs, 142, 35

1494: 

1495: \bibitem[{Kroupa(2002)}]{Kroupa02}

1496: Kroupa, P. 2002, \mnras, 330, 707

1497: 

1498: \bibitem[{{Lee} {et~al.}(2006){Lee}, {Skillman}, {Cannon}, {Jackson}, {Gehrz},

1499:   {Polomski}, \& {Woodward}}]{Lee06b}

1500: {Lee}, H., {Skillman}, E.~D., {Cannon}, J.~M., {Jackson}, D.~C., {Gehrz},

1501:   R.~D., {Polomski}, E.~F., \& {Woodward}, C.~E. 2006, \apj, 647, 970

1502: 

1503: \bibitem[{MacArthur(2005)}]{MacArthur05}

1504: MacArthur, L.~A. 2005, \apj, 623, 795

1505: 

1506: \bibitem[{MacArthur(2006)}]{MacArthur06}

1507: ---. 2006, Ph.D.~Thesis

1508: 

1509: \bibitem[{MacArthur {et~al.}(2004)MacArthur, Courteau, Bell, \&

1510:   Holtzman}]{MacArthur04}

1511: MacArthur, L.~A., Courteau, S., Bell, E., \& Holtzman, J.~A. 2004, \apjs, 152,

1512:   175

1513: 

1514: \bibitem[{Martin {et~al.}(2004)Martin, Ibata, Bellazzini, Irwin, Lewis, \&

1515:   Dehnen}]{Martin04}

1516: Martin, N.~F., Ibata, R.~A., Bellazzini, M., Irwin, M.~J., Lewis, G.~F., \&

1517:   Dehnen, W. 2004, \mnras, 348, 12

1518: 

1519: \bibitem[{Mashonkina {et~al.}(2003)Mashonkina, Gehren, Travaglio, \&

1520:   Borkova}]{mashonkina03}

1521: Mashonkina, L., Gehren, T., Travaglio, C., \& Borkova, T. 2003, \aap, 397, 275

1522: 

1523: \bibitem[{{Matteucci}(1994)}]{Matteucci94}

1524: {Matteucci}, F. 1994, \aap, 288, 57

1525: 

1526: \bibitem[{Matthews {et~al.}(1999)Matthews, Gallagher, \& van

1527:   Driel}]{Matthews99}

1528: Matthews, L.~D., Gallagher, J.~S., \& van Driel, W. 1999, \aj, 118, 2751

1529: 

1530: \bibitem[{Moorthy \& Holtzman(2006)}]{Moorthy06}

1531: Moorthy, B.~K., \& Holtzman, J.~A. 2006, \mnras, 371, 583

1532: 

1533: \bibitem[{Morrison {et~al.}(1997)Morrison, Miller, Harding, Stinebring, \&

1534:   Boroson}]{Morrison97}

1535: Morrison, H.~L., Miller, E.~D., Harding, P., Stinebring, D.~R., \& Boroson,

1536:   T.~A. 1997, \aj, 113, 2061

1537: 

1538: \bibitem[{{Mouhcine} {et~al.}(2005){Mouhcine}, {Rich}, {Ferguson}, {Brown}, \&

1539:   {Smith}}]{Mouhcine05}

1540: {Mouhcine}, M., {Rich}, R.~M., {Ferguson}, H.~C., {Brown}, T.~M., \& {Smith},

1541:   T.~E. 2005, \apj, 633, 828

1542: 

1543: \bibitem[{Mould(2005)}]{Mould05}

1544: Mould, J. 2005, \aj, 129, 698

1545: 

1546: \bibitem[{Navarro {et~al.}(2004)Navarro, Helmi, \& Freeman}]{Navarro04}

1547: Navarro, J.~F., Helmi, A., \& Freeman, K.~C. 2004, \apjl, 601, L43

1548: 

1549: \bibitem[{Neeser {et~al.}(2002)Neeser, Sackett, De~Marchi, \&

1550:   Paresce}]{Neeser02}

1551: Neeser, M.~J., Sackett, P.~D., De~Marchi, G., \& Paresce, F. 2002, \aap, 383,

1552:   472

1553: 

1554: \bibitem[{{Norris}(1987)}]{Norris87}

1555: {Norris}, J. 1987, \apjl, 314, L39

1556: 

1557: \bibitem[{Osterbrock(1989)}]{Osterbrock89}

1558: Osterbrock, D.~E. 1989

1559: 

1560: \bibitem[{Osterbrock {et~al.}(1996)Osterbrock, Fulbright, Martel, Keane,

1561:   Trager, \& Basri}]{Oster96}

1562: Osterbrock, D.~E., Fulbright, J.~P., Martel, A.~R., Keane, M.~J., Trager,

1563:   S.~C., \& Basri, G. 1996, \pasp, 108, 277

1564: 

1565: \bibitem[{{Peletier} {et~al.}(2007){Peletier}, {Falcon-Barroso}, {Bacon},

1566:   {Cappellari}, {Davies}, {de Zeeuw}, {Emsellem}, {Ganda}, {Krajnovic},

1567:   {Kuntschner}, {McDermid}, {Sarzi}, {van de Ven}, {Emsellem}, {Ganda},

1568:   {Krajnovic}, {Kuntschner}, {McDermid}, {Sarzi}, \& {van de Ven}}]{Peletier07}

1569: {Peletier}, R.~F., {Falcon-Barroso}, J., {Bacon}, R., {Cappellari}, M.,

1570:   {Davies}, R.~L., {de Zeeuw}, P.~T., {Emsellem}, E., {Ganda}, K., {Krajnovic},

1571:   D., {Kuntschner}, H., {McDermid}, R.~M., {Sarzi}, M., {van de Ven}, G.,

1572:   {Emsellem}, {Ganda}, K., {Krajnovic}, D., {Kuntschner}, H., {McDermid},

1573:   R.~M., {Sarzi}, M., \& {van de Ven}, G. 2007, ArXiv e-prints, 704

1574: 

1575: \bibitem[{Pohlen {et~al.}(2004)Pohlen, Balcells, L\"utticke, \&

1576:   Dettmar}]{Pohlen04}

1577: Pohlen, M., Balcells, M., L\"utticke, R., \& Dettmar, R.-J. 2004, \aap, 422,

1578:   465

1579: 

1580: \bibitem[{Prochaska {et~al.}(2000)Prochaska, Naumov, Carney, McWilliam, \&

1581:   Wolfe}]{Prochaska00}

1582: Prochaska, J.~X., Naumov, S.~O., Carney, B.~W., McWilliam, A., \& Wolfe, A.~M.

1583:   2000, \aj, 120, 2513

1584: 

1585: \bibitem[{{Proctor} {et~al.}(2000){Proctor}, {Sansom}, \& {Reid}}]{Proctor00}

1586: {Proctor}, R.~N., {Sansom}, A.~E., \& {Reid}, I.~N. 2000, \mnras, 311, 37

1587: 

1588: \bibitem[{{Prugniel} {et~al.}(2001){Prugniel}, {Maubon}, \&

1589:   {Simien}}]{Prugniel01}

1590: {Prugniel}, P., {Maubon}, G., \& {Simien}, F. 2001, \aap, 366, 68

1591: 

1592: \bibitem[{Quinn {et~al.}(1993)Quinn, Hernquist, \& Fullagar}]{Quinn93}

1593: Quinn, P.~J., Hernquist, L., \& Fullagar, D.~P. 1993, \apj, 403, 74

1594: 

1595: \bibitem[{Rampazzo {et~al.}(2005)Rampazzo, Annibali, Bressan, Longhetti,

1596:   Padoan, \& Zeilinger}]{Rampazzo05}

1597: Rampazzo, R., Annibali, F., Bressan, A., Longhetti, M., Padoan, F., \&

1598:   Zeilinger, W.~W. 2005, \aap, 433, 497

1599: 

1600: \bibitem[{{Read} {et~al.}(2008){Read}, {Lake}, {Agertz}, \&

1601:   {Debattista}}]{Read08}

1602: {Read}, J.~I., {Lake}, G., {Agertz}, O., \& {Debattista}, V.~P. 2008, ArXiv

1603:   e-prints, 803

1604: 

1605: \bibitem[{Reddy {et~al.}(2006)Reddy, Lambert, \& Allende~Prieto}]{Reddy06}

1606: Reddy, B.~E., Lambert, D.~L., \& Allende~Prieto, C. 2006, \mnras, 367, 1329

1607: 

1608: \bibitem[{Reid \& Majewski(1993)}]{Reid93}

1609: Reid, N., \& Majewski, S.~R. 1993, \apj, 409, 635

1610: 

1611: \bibitem[{Robin {et~al.}(1996)Robin, Haywood, Creze, Ojha, \&

1612:   Bienayme}]{Robin96}

1613: Robin, A.~C., Haywood, M., Creze, M., Ojha, D.~K., \& Bienayme, O. 1996, \aap,

1614:   305, 125

1615: 

1616: \bibitem[{Ryder {et~al.}(2005)Ryder, Fenner, \& Gibson}]{Ryder05}

1617: Ryder, S.~D., Fenner, Y., \& Gibson, B.~K. 2005, \mnras, 358, 1337

1618: 

1619: \bibitem[{{S{\'a}nchez-Bl{\'a}zquez} {et~al.}(2007){S{\'a}nchez-Bl{\'a}zquez},

1620:   {Forbes}, {Strader}, {Brodie}, \& {Proctor}}]{SB07}

1621: {S{\'a}nchez-Bl{\'a}zquez}, P., {Forbes}, D.~A., {Strader}, J., {Brodie}, J.,

1622:   \& {Proctor}, R. 2007, \mnras, 377, 759

1623: 

1624: \bibitem[{Schiavon(2006)}]{Schiavon06}

1625: Schiavon, R.~P. 2006, ArXiv Astrophysics e-prints

1626: 

1627: \bibitem[{Serra \& Trager(2007)}]{Serra07}

1628: Serra, P., \& Trager, S.~C. 2007, \mnras, 374, 769

1629: 

1630: \bibitem[{Seth {et~al.}(2005)Seth, Dalcanton, \& de~Jong}]{Seth05b}

1631: Seth, A.~C., Dalcanton, J.~J., \& de~Jong, R.~S. 2005, \aj, 130, 1574

1632: 

1633: \bibitem[{Shaw \& Gilmore(1989)}]{Shaw89}

1634: Shaw, M.~A., \& Gilmore, G. 1989, \mnras, 237, 903

1635: 

1636: \bibitem[{Statler(1988)}]{Statler88}

1637: Statler, T.~S. 1988, \apj, 331, 71

1638: 

1639: \bibitem[{Tautvai{\v s}ien{\.e} {et~al.}(2001)Tautvai{\v s}ien{\.e},

1640:   Edvardsson, Tuominen, \& Ilyin}]{Taut01}

1641: Tautvai{\v s}ien{\.e}, G., Edvardsson, B., Tuominen, I., \& Ilyin, I. 2001,

1642:   \aap, 380, 578

1643: 

1644: \bibitem[{Thomas {et~al.}(2003)Thomas, Maraston, \& Bender}]{Thomas03}

1645: Thomas, D., Maraston, C., \& Bender, R. 2003, \mnras, 339, 897

1646: 

1647: \bibitem[{Tikhonov \& Galazutdinova(2005)}]{Tikhonovo5b}

1648: Tikhonov, N.~A., \& Galazutdinova, O.~A. 2005, Astrophysics, 48, 221

1649: 

1650: \bibitem[{{Tikhonov} \& {Galazutdinova}(2008)}]{Tik08}

1651: {Tikhonov}, N.~A., \& {Galazutdinova}, O.~A. 2008, Astronomy Reports, 52, 19

1652: 

1653: \bibitem[{Tikhonov {et~al.}(2005)Tikhonov, Galazutdinova, \&

1654:   Drozdovsky}]{Tikhonov05}

1655: Tikhonov, N.~A., Galazutdinova, O.~A., \& Drozdovsky, I.~O. 2005, \aap, 431,

1656:   127

1657: 

1658: \bibitem[{Trager {et~al.}(2000{\natexlab{a}})Trager, Faber, Worthey, \&

1659:   Gonz\'alez}]{Trager00b}

1660: Trager, S.~C., Faber, S.~M., Worthey, G., \& Gonz\'alez, J.~J.

1661:   2000{\natexlab{a}}, \aj, 120, 165

1662: 

1663: \bibitem[{Trager {et~al.}(2000{\natexlab{b}})Trager, Faber, Worthey, \&

1664:   Gonz\'alez}]{Trager00}

1665: ---. 2000{\natexlab{b}}, \aj, 119, 1645

1666: 

1667: \bibitem[{Trager {et~al.}(1998)Trager, Worthey, Faber, Burstein, \&

1668:   Gonzalez}]{Trager98}

1669: Trager, S.~C., Worthey, G., Faber, S.~M., Burstein, D., \& Gonzalez, J.~J.

1670:   1998, \apjs, 116, 1

1671: 

1672: \bibitem[{Tremonti {et~al.}(2004)Tremonti, Heckman, Kauffmann, Brinchmann,

1673:   Charlot, White, Seibert, Peng, Schlegel, Uomoto, Fukugita, \&

1674:   Brinkmann}]{Tremonti04}

1675: Tremonti, C.~A., Heckman, T.~M., Kauffmann, G., Brinchmann, J., Charlot, S.,

1676:   White, S.~D.~M., Seibert, M., Peng, E.~W., Schlegel, D.~J., Uomoto, A.,

1677:   Fukugita, M., \& Brinkmann, J. 2004, \apj, 613, 898

1678: 

1679: \bibitem[{Tsikoudi(1979)}]{Tsikoudi79}

1680: Tsikoudi, V. 1979, \apj, 234, 842

1681: 

1682: \bibitem[{van~der Kruit(1984)}]{Kruit84}

1683: van~der Kruit, P.~C. 1984, \aap, 140, 470

1684: 

1685: \bibitem[{van Dokkum {et~al.}(1994)van Dokkum, Peletier, de~Grijs, \&

1686:   Balcells}]{vanDokkum94}

1687: van Dokkum, P.~G., Peletier, R.~F., de~Grijs, R., \& Balcells, M. 1994, \aap,

1688:   286, 415

1689: 

1690: \bibitem[{van Zee {et~al.}(1998)van Zee, {Salzer}, {Haynes}, {O'Donoghue}, \&

1691:   {Balonek}}]{vanzee98}

1692: van Zee, L., {Salzer}, J.~J., {Haynes}, M.~P., {O'Donoghue}, A.~A., \&

1693:   {Balonek}, T.~J. 1998, \aj, 116, 2805

1694: 

1695: \bibitem[{{Vazdekis} \& {Arimoto}(1999)}]{Vazdekis99}

1696: {Vazdekis}, A., \& {Arimoto}, N. 1999, \apj, 525, 144

1697: 

1698: \bibitem[{Velazquez \& White(1999)}]{Velazquez99}

1699: Velazquez, H., \& White, S.~D.~M. 1999, \mnras, 304, 254

1700: 

1701: \bibitem[{{Viironen} {et~al.}(2007){Viironen}, {Delgado-Inglada}, {Mampaso},

1702:   {Magrini}, \& {Corradi}}]{Viironen07}

1703: {Viironen}, K., {Delgado-Inglada}, G., {Mampaso}, A., {Magrini}, L., \&

1704:   {Corradi}, R.~L.~M. 2007, \mnras, 381, 1719

1705: 

1706: \bibitem[{{Villalobos} \& {Helmi}(2008)}]{Villalobos08}

1707: {Villalobos}, {\'A}., \& {Helmi}, A. 2008, ArXiv e-prints, 803

1708: 

1709: \bibitem[{Villumsen(1985)}]{Villumsen85}

1710: Villumsen, J.~V. 1985, \apj, 290, 75

1711: 

1712: \bibitem[{Walker {et~al.}(1996)Walker, Mihos, \& Hernquist}]{Walker96}

1713: Walker, I.~R., Mihos, J.~C., \& Hernquist, L. 1996, \apj, 460, 121

1714: 

1715: \bibitem[{Worthey(1994)}]{Worthey94b}

1716: Worthey, G. 1994, \apjs, 95, 107

1717: 

1718: \bibitem[{{Worthey} {et~al.}(1992){Worthey}, {Faber}, \&

1719:   {Gonzalez}}]{Worthey92}

1720: {Worthey}, G., {Faber}, S.~M., \& {Gonzalez}, J.~J. 1992, \apj, 398, 69

1721: 

1722: \bibitem[{Worthey {et~al.}(1994)Worthey, Faber, Gonzalez, \&

1723:   Burstein}]{Worthey94}

1724: Worthey, G., Faber, S.~M., Gonzalez, J.~J., \& Burstein, D. 1994, \apjs, 94,

1725:   687

1726: 

1727: \bibitem[{Worthey \& Ottaviani(1997)}]{Worthey97}

1728: Worthey, G., \& Ottaviani, D.~L. 1997, \apjs, 111, 377

1729: 

1730: \bibitem[{Wu {et~al.}(2002)Wu, Burstein, Deng, Zhou, Shang, Zheng, Chen, Su,

1731:   Windhorst, Chen, Zou, Xia, Jiang, Ma, Xue, Zhu, Cheng, Byun, Chen, Deng, Fan,

1732:   Fang, Kong, Li, Lin, Lu, Sun, Tsay, Xu, Yan, Zhao, \& Zheng}]{Wu02}

1733: Wu, H., Burstein, D., Deng, Z., Zhou, X., Shang, Z., Zheng, Z., Chen, J., Su,

1734:   H., Windhorst, R.~A., Chen, W., Zou, Z., Xia, X., Jiang, Z., Ma, J., Xue, S.,

1735:   Zhu, J., Cheng, F., Byun, Y., Chen, R., Deng, L., Fan, X., Fang, L., Kong,

1736:   X., Li, Y., Lin, W., Lu, P., Sun, W., Tsay, W., Xu, W., Yan, H., Zhao, B., \&

1737:   Zheng, Z. 2002, \aj, 123, 1364

1738: 

1739: \bibitem[{Yoachim \& Dalcanton(2005)}]{Yoachim05}

1740: Yoachim, P., \& Dalcanton, J.~J. 2005, \apj, 624, 701

1741: 

1742: \bibitem[{Yoachim \& Dalcanton(2006)}]{Yoachim06}

1743: ---. 2006, \aj, 131, 226

1744: 

1745: \bibitem[{{Yoachim} \& {Dalcanton}(2007)}]{Yoachim07b}

1746: {Yoachim}, P., \& {Dalcanton}, J.~J. 2007, AAS Meeting Abstracts, 211, 95.18

1747: 

1748: \bibitem[{{Yoachim} \& {Dalcanton}(2008)}]{Yoachim07inp2}

1749: ---. 2008, astro-ph 0804.3966

1750: 

1751: \bibitem[{Zackrisson {et~al.}(2006)Zackrisson, Bergvall, \"Ostlin, Micheva, \&

1752:   Leksell}]{Zackrisson06}

1753: Zackrisson, E., Bergvall, N., \"Ostlin, G., Micheva, G., \& Leksell, M. 2006,

1754:   \apj, 650, 812

1755: 

1756: \bibitem[{{Zaritsky} {et~al.}(1994){Zaritsky}, {Kennicutt}, \&

1757:   {Huchra}}]{Zaritsky94}

1758: {Zaritsky}, D., {Kennicutt}, Jr., R.~C., \& {Huchra}, J.~P. 1994, \apj, 420, 87

1759: 

1760: \bibitem[{Zibetti {et~al.}(2004)Zibetti, White, \& Brinkmann}]{Zibetti04}

1761: Zibetti, S., White, S.~D.~M., \& Brinkmann, J. 2004, \mnras, 347, 556

1762: 

1763: \end{thebibliography}

1764: 

1765: 

1766: 

1767: \end{document}

1768: 

1769: % LocalWords:  Tremonti Peletier Worthey passbands VESPA Tojeiro readnoise APG

1770: % LocalWords:  overlayed IC PyRAF overscan LSR arcsec FHWM EWs HD Schiavon EW

1771: % LocalWords:  Trager dichroic Rampazzo Annibali Denicol Calzetti starbursts ph

1772: % LocalWords:  Osterbrock Puzia MgFe SNe Zackrisson AST Zaritsky Beauchamp CCDs

1773: % LocalWords:  LOSVD BCGs Tikhonovo IMFs Koeppen Rockosi Villumsen Hann Hayashi

1774: % LocalWords:  Prugniel inp APO's Wyckoff dex ACS Vazdekis Blazquez isochrone

1775: % LocalWords:  vanzee centroiding Kewley longslits jd's Reddy mashonkina RMS

1776: % LocalWords:  versa Tik Villalobos kaz passband Astronomische Nachrichten AAS

1777: % LocalWords:  Agertz astro Sohi Doi Kiuchi Bushong

1778: