1: \documentclass{emulateapj}
2:
3: \shorttitle{The Stellar Populations of NGC 4522}
4: \shortauthors{Crowl \& Kenney}
5:
6: \begin{document}
7:
8: \title{The Stellar Population of Stripped Cluster Spiral NGC 4522: A
9: Local Analog to K+A Galaxies?}
10: \author{Hugh H. Crowl}
11: \author{Jeffrey D.P. Kenney}
12: \affil{Department of Astronomy, Yale University, New Haven, CT 06520}
13: \email{hugh@astro.yale.edu}
14:
15: \begin{abstract}
16:
17: We present observations of the stripped Virgo Cluster spiral
18: NGC~4522, a clear, nearby example of a galaxy currently undergoing
19: ISM-ICM stripping. Utilizing SparsePak integral field spectroscopy
20: on the WIYN 3.5m telescope and GALEX UV photometry, we present an
21: analysis of the outer disk ($r > 3$ kpc) stellar population of this
22: galaxy, beyond the HI and H$\alpha$ truncation radius. We find that
23: the star formation in the gas-stripped outer disk ceased very
24: recently, $\sim 100$ Myr ago, in agreement with previous claims that
25: this galaxy is currently being stripped. At the time of this
26: stripping, data and models suggest that the galaxy experienced a
27: modest starburst. The stripping is occurring in a region of the
28: cluster well outside the cluster core, likely because this galaxy is
29: experiencing extreme conditions from a dynamic ICM due to an ongoing
30: sub-cluster merger. The outer disk has a spectrum of a K+A galaxy,
31: traditionally observed in high-redshift cluster galaxies. In the
32: case of NGC~4522, a K+A spectrum is formed by simple stripping of
33: the interstellar gas by the hot intracluster medium. These data
34: show K+A spectra can be created by cluster processes and that these
35: processes likely extend beyond the cluster core.
36: \end{abstract}
37: \keywords{galaxies: spiral, galaxies: clusters: individual (Virgo),
38: galaxies: individual (NGC 4522), galaxies: evolution}
39:
40: \section{Introduction}
41:
42: K+A galaxies \citep{dg83}, originally discovered in clusters at high
43: redshift, differ from typical elliptical, spiral or irregular galaxies
44: in an important way. These galaxies have spectra characterized by
45: strong Balmer absorption lines and no significant emission from
46: ongoing star formation. While the formation mechanisms of K+A galaxies
47: are still unknown \citep{tran03,poggianti04a,bekki05,goto05}, these
48: spectra are generally interpreted as the signature of recent cessation
49: of star formation (however, c.f. \citealp{burstein05}).
50: \citet{shioya02} show through star formation models that such
51: cessation and passive evolution of the already-existing population can
52: form a K+A spectrum. While the interpretation of these spectra as the
53: result of temporally truncated star formation is largely accepted, the
54: cause of the termination of star formation is not settled.
55:
56: Such catastrophic interruption of star formation seems likely to be caused
57: by external processes. Events such as mergers, galaxy harassment, and
58: ram pressure stripping all have the potential to terminate star
59: formation. While K+A galaxies have been chiefly observed in clusters,
60: observations of K+A spectra in lower density environments (i.e.
61: \citealp{zabludoff96,tran04,goto05}) show that they are not unique to
62: the cluster environment. What role does environment play in the
63: creation of these galaxies?
64:
65: In the Virgo cluster, many spirals have truncated H$\alpha$ disks
66: \citep{koopmann04}, suggesting that ram pressure stripping has cut off
67: the star formation in the outer disk. These galaxies have normal star
68: formation inside a truncation radius, with no detected star formation
69: at larger radii. The truncation location varies from galaxy to galaxy,
70: with some galaxies moderately truncated inside $0.8 R_{25}$ and others
71: more severely truncated inside $0.4 R_{25}$. In these galaxies, it is
72: possible to measure age-sensitive absorption lines from the stellar
73: populations beyond the truncation radius without contamination of
74: emission lines from star-forming regions. The ages of the youngest
75: stellar population in the outer disk tells us how much time has
76: elapsed since the last epoch of star formation and, therefore, the
77: time elapsed since the gas was removed. This, combined with the
78: galaxy's location in the cluster, can tell us where in the cluster
79: galaxies are stripped of their star-forming gas.
80:
81: NGC~4522 is one of the clearest examples of ongoing ISM-ICM stripping
82: observed in the nearby Virgo cluster. It is a highly inclined,
83: 0.5L$_*$ Sc galaxy $3\fdg3$ ($1.3 r_{200}$) south of M87, in a region
84: of the cluster with modest ICM density (as measured from X-ray
85: emission from \citealp{bohringer94}). There are several lines of
86: evidence that NGC~4522 is experiencing ongoing ICM pressure. Despite
87: an undisturbed stellar disk, the HI and H$\alpha$ emission are
88: spatially truncated in the disk at $r$=3 kpc=0.4 $R_{25}$ and
89: significant amounts of extraplanar HI and H$\alpha$ emission exist on
90: one side of the disk close to the truncation radius
91: \citep{kk99,kenney04}. A ridge of highly polarized radio continuum
92: emission is observed in the disk opposite the extraplanar gas,
93: suggesting compression of the magnetic field lines at the leading edge
94: of the ISM-ICM interaction \citep{vollmer04}. These data are
95: consistent with a galaxy currently undergoing ram pressure; comparison
96: of simulations and observations \citep{vollmer06} suggest that we are
97: observing this galaxy at a time close to peak pressure.
98:
99: \begin{figure*}
100: \includegraphics[width=7in]{f1.eps}
101: \caption{Left: $R$ band image of NGC~4522, with the position of the
102: SparsePak fibers superimposed over the image. Middle: H$\alpha$
103: image of NGC~4522, with HI contours \citep{kenney04} and the
104: SparsePak footprint overlayed. In both figures, the outlines of the
105: fibers used to form the composite spectrum are bold. Right: Optical
106: spectrum, resulting from the combination of several fibers just
107: beyond the gas truncation radius.}
108: \label{images}
109: \end{figure*}
110:
111: \newpage
112:
113: \section{Observations}
114:
115: \subsection{Optical Spectroscopy}
116:
117: NGC~4522 was observed with the SparsePak integrated field spectrograph
118: \citep{bershady04} on the WIYN 3.5m telescope during the night of
119: March 29-30, 2003. SparsePak has 75 object fibers, each with a
120: diameter of $4\farcs7$, spread over an $80\arcsec$ x $80\arcsec$
121: region of the sky. The fiber spacing for most of the array is
122: $10\farcs3$ , except for a denser central core where the fibers have a
123: center-to-center spacing of $5\farcs6$ The SparsePak bundle was
124: positioned with the majority of the fibers located beyond the
125: H$\alpha$ truncation radius (Figure \ref{images}). We followed
126: standard IRAF image reduction techniques and used the {\tt dohydra}
127: task to extract the spectra, determine the wavelength solution, and
128: subtract sky light. The data were then flux calibrated using
129: spectrophotometric standards. The resulting data have a resolution of
130: 5.5 \AA (FWHM). Following the complete reduction of the spectral
131: data, the spectrum from each fiber was examined to determine the
132: quality of the data. The H$\alpha$ truncation radius determined by
133: spectroscopy matches the imaging data. We combine six fibers with no
134: H$\alpha$ emission and high signal-to-noise just beyond the gas
135: truncation radius to create a higher signal-to-noise composite
136: spectrum (Figure \ref{images}). These spectra sample a region along
137: the major axis to the SW between 45\arcsec (3.5 kpc\footnote{We assume
138: a distance to Virgo of 16 Mpc.}) and 60\arcsec (4.5 kpc), within
139: 5\arcsec (380 pc) of the major axis. These fibers all show H$\alpha$
140: absorption from the underlying stellar population and no apparent
141: H$\alpha$ emission. All stellar population results presented here are
142: based on this averaged spectrum. From these data, we extract
143: absorption line indices based on the Lick/IDS
144: definitions\footnote{Note that the data are \emph{not} smoothed to the
145: Lick resolution, but that we simply use the Lick feature wavelength
146: definitions.} \citep{faber85}.
147:
148:
149: \subsection{GALEX Photometry}
150:
151: Observations of NGC~4522 were conducted with GALEX as part of that
152: instrument's All Sky Survey \citep{martin05}. NGC~4522 was observed
153: for $0.1\textrm{ksec}$ in both the FUV and NUV channels. Using a
154: series of circular apertures chosen to match SparsePak fiber
155: positions, we have extracted FUV and NUV fluxes for the same 1~kpc
156: region along the major axis where we extract optical spectra. We
157: combine these data with F606-band ($\sim V$) HST optical imaging
158: \citep{kenney06} to determine FUV-to-optical and FUV-to-NUV flux
159: ratios. Note that, because this region has been stripped of its gas
160: and dust, the UV fluxes and colors are minimally affected by internal
161: dust absorption.
162:
163: \section{Results}
164:
165: \subsection{Properties of the Stripped Region}
166:
167: The optical spectrum of NGC 4522 is very blue, with the continuum
168: continuing to rise all the way to the blue spectral limit, unlike the
169: red spectrum of a typical elliptical galaxy. This implies that the
170: light from the galaxy is dominated by young stars. This observation is
171: in agreement with the strong Balmer absorption in the outer disk of
172: NGC~4522. We see strong absorption not only in H$\beta$, but also in
173: the higher order lines: H$\gamma$ and H$\delta$, lines characteristic
174: of a young, A-star-dominated population. If we compare this spectrum
175: with those observed at higher redshift, we find a striking similarity
176: to K+A galaxies. \citet{dressler99} quantify K+A galaxies as those
177: galaxies with no significant star-forming
178: emission\footnote{\citet{dressler99} and many others use [OII]3727 as
179: their indicator of star formation activity. As our spectral
180: coverage does not extend blueward of 4000 \AA, we use H$\alpha$ as
181: our star formation diagnostic.} and strong H$\delta$ absorption
182: (EW(H$\delta$)$ > 3$ \AA). The outer disk of NGC~4522 clearly fits
183: this description, with no significant H$\alpha$ emission and
184: EW(H$\delta$)=5.2 \AA.
185:
186: The UV colors of this region show a similar signature: a nearly flat
187: spectrum extends from the FUV band ($\lambda_{\textrm{eff}} \sim 1500$
188: \AA) to the F606 band ($\lambda_{\textrm{eff}} \sim 5900$ \AA). In
189: particular, NGC~4522 is brighter in the GALEX FUV bandpass than in the
190: NUV passband by a modest but significant amount
191: ($F_{\textrm{\scriptsize FUV}} = 3.9\times 10^{-16}$
192: erg/s/cm$^2$/\AA~vs.~$F_{\textrm{\scriptsize NUV}} = 3.0\times
193: 10^{-16}$ erg/s/cm$^2$/\AA). Because the FUV flux fades quickly in an
194: evolving stellar population (e.g. \citealp{bc03}), the bright FUV
195: emission from the outer disk implies a very young stellar population.
196:
197: \subsection{Stellar Population Models}
198:
199: \begin{figure}[t]
200: \includegraphics[width=3.4in]{f2.eps}
201: % \plotone{f2.eps}
202: \caption{Index diagrams for Starburst99 models for an SSP (left),
203: temporally truncated star formation (middle), and temporally
204: truncated star formation with a 2\% starburst at the time of
205: truncation (right). Shown are the data (star) and models (lines)
206: for the age-sensitive H$\beta$ line plotted against the
207: metallicity-sensitive [MgFe]$^\prime$ index. The Starburst99
208: model plotted in the SSP figure (left) is for solar metallicity,
209: with ages in Gyr marked along the line. In each of the other two
210: figures (middle and right), the solid lines correspond to lines
211: of constant truncation age, while the broken lines correspond to
212: lines of constant metallicity (Z=0.008 (short dash), Z=0.02
213: (dot-dash), Z=0.04 (long dash)).}
214: \label{index-diag}
215: \end{figure}
216:
217:
218: Spiral galaxies are typically characterized by ongoing star formation.
219: In the case of the stripped spiral galaxies in Virgo, it's natural to
220: assume a temporally truncated star formation history: roughly constant
221: star formation, followed by abrupt cessation of star formation and
222: passive evolution of the already-existing population. If one assumes a
223: (temporally) truncated star formation history, the youngest (and
224: brightest) stellar population will dominate the light, but there will
225: be some ``pollution'' from the older stellar population. While the
226: Balmer lines from the integrated light will be strong in this
227: population, their strength will be somewhat diluted by the older
228: population. In contrast, models assuming a ``single burst'' star
229: formation history (a so-called ``Simple Stellar Population'' or
230: ``SSP''), will have stronger Balmer lines for $\sim 1$ Gyr, the
231: approximate lifetime of the Balmer-line-dominating A~stars. This
232: means that, for a given set of Balmer line indices, the last epoch of
233: star formation in the temporally truncated model will always be
234: \emph{younger}\footnote{Any ``age'' referred to with respect to the
235: temporally truncated star formation model is the age of the youngest
236: generation of stars; \emph{not} the luminosity weighted mean age of
237: the population.} than the age of the SSP model. The GALEX UV data are,
238: therefore, critical for constraining the age of the stellar
239: population. The detection of bright, FUV flux in the outer disk of
240: NGC~4522 demonstrates that the stellar population is very young.
241:
242: Due to the presence of very young stars in the outer disk of NGC~4522
243: and the lack of young, metal-poor stars in empirical libraries, we
244: have chosen to use Starburst99 models \citep{leitherer99}, with
245: theoretical stellar atmospheres \citep{martins05}. Starburst99
246: generates SSP models with a range of ages and the sum of several
247: theoretical spectra are used to model a truncated star formation
248: history. The spectra are then smoothed to the resolution of our data
249: (5.5\AA FWHM) and spectral line indices and broadband photometric
250: colors are extracted. In Figure \ref{index-diag}, we plot the
251: strength of the age-sensitive Balmer features against the strength of
252: the metallicity-sensitive [MgFe]$^\prime$ index for a SSP, a set of
253: temporally truncated star formation models, and a set of temporally
254: truncated models in which 2\% of the stellar mass formed in a burst at
255: the end of the star forming episode. In Figure \ref{fig-galexcolors},
256: we plot broadband UV and optical flux ratios for same three star
257: formation histories.
258:
259: \subsection{Time Since Outer Disk Truncation}
260:
261: \begin{figure}[t]
262: \plotone{f3.eps}
263: \caption{UV-optical colors derived for the models shown in Figure 2:
264: SSP (dashed line), temporally truncated (dash-dot line), and
265: truncated with 2\% burst (dotted line). The solid lines correspond
266: to ages of (from top to bottom) 100 Myr, 500 Myr, 1 Gyr and 5
267: Gyr. The value of the flux ratios for NGC~4522 is plotted as a
268: asterisk. A representative reddening vector (determined from the
269: results of \citealp{calzetti97} and using a value of E(B-V)=0.02
270: from \citealp{schlegel98}) is shown in the upper left section of
271: the plot.}
272: \label{fig-galexcolors}
273: \end{figure}
274:
275:
276: The GALEX UV photometry (Figure \ref{fig-galexcolors}) and optical
277: spectroscopy (Figure \ref{index-diag}) are best fit by an temporally
278: truncated Starburst99 model, with a cessation age of $\sim 100$ Myr.
279: This is based on the combination of strong Balmer absorption and
280: bright FUV emission in the outer disk. Either the optical
281: spectroscopy or the GALEX photometry, when taken independently, can be
282: fit by various models. The Balmer absorption lines indicate that
283: either the stellar population is younger than a 50 Myr truncated star
284: formation model \emph{or} as old as a 1 Gyr SSP. Truncation is clearly
285: the more logical star formation history, and we find the GALEX colors
286: are far too blue for a 1 Gyr SSP; in fact, the GALEX data is most
287: consistent with being truncated $\sim 100$ Myr ago, an estimate that
288: is relatively unaffected by reddening due to the fact that the
289: reddening vector is nearly parallel to the line of constant age
290: (Figure \ref{fig-galexcolors}). While the Balmer lines are stronger
291: than any simple truncated stellar population model, a modest 2\% (by
292: mass) starburst is enough to bring the absorption line data and GALEX
293: data into agreement (Figure \ref{index-diag}). Taken together, the
294: data exclude a SSP star formation history and show roughly constant
295: star formation in the outer part of NGC~4522 until $\sim 100$ Myr ago,
296: when there was a modest starburst, followed by the current period of
297: no star formation and passive stellar evolution. This scenario is
298: consistent with the observations that this is a spiral galaxy
299: currently being stripped of its neutral gas \citep{kenney04,vollmer06}
300: and with H$\alpha$ observations suggesting that the inner disk has a
301: moderately enhanced star formation rate \citep{kenney04}.
302:
303:
304:
305: \section{Discussion}
306:
307: Our results clearly show that the cessation of star formation was
308: recent and, therefore, that the outer disk of NGC~4522 was recently
309: stripped of its star-forming gas. We would like to understand how
310: rapidly star formation stops during an interaction with the ICM and
311: how long dense molecular star-forming clouds might survive after the
312: low-density atomic gas is stripped. Simulations \citep{vollmer06}
313: compared with the HI observations \citep{kenney04} suggest that we are
314: observing the galaxy $\sim 50$ Myrs after peak pressure. This
315: timescale is consistent with that determined from the stellar
316: population, suggesting that neutral gas stripping and termination of
317: star formation nearly coincide.
318:
319: \subsection{Stripping Outside the Cluster Core}
320:
321: From a simple timescale argument, the stellar populations of NGC~4522
322: tell us that this galaxy must be stripped locally, 0.9~Mpc ($1.3
323: r_{200}$) from M87, as opposed to closer to the cluster core. If we
324: assume that the galaxy's motion in the plane of the sky is a factor of
325: 2 higher than the average radial velocity of a Virgo Cluster galaxy
326: (i.e. we assume $v_{\textrm{\scriptsize t}} = 1400$ km/s), it would
327: still take the galaxy $\sim 500$ Myr to travel the $\sim 0.75$ Mpc
328: that separates its current location from the region where the ICM
329: density is high enough to strip the outer disk. \emph{The observation
330: that star formation ceased $\sim 100$ Myr ago therefore rules out
331: stripping in the core and argues that stripping must be occurring
332: locally.} This result is in agreement with simulations of the HI
333: morphology and kinematics \citep{vollmer06} which also argue for local
334: stripping. Stripping so far out in the cluster shows either that gas
335: stripping of galaxies in clusters is easier than the simple estimates
336: of ram pressure imply \citep{kenney04}, or that some galaxies
337: experience extreme conditions due to a dynamic ICM
338: \citep{dupke06,heinz06}. In particular, NGC~4522 appears to be located
339: between M87 and the M49 subcluster, which is likely in the process of
340: merging with Virgo \citep{schindler99,shibata01}. Moreover,
341: \citet{shibata01} observe a shock front in the ICM near the location
342: of NGC~4522, suggesting that the local ICM has a high velocity. This
343: makes it plausible that gas stripping is important for transforming
344: galaxies well outside the cluster core, including at the interface
345: between merging subclusters.
346:
347: \subsection{Similarity to Higher-z K+A Galaxies}
348:
349: The outer disk of NGC~4522 has strong Balmer lines and no emission,
350: similar to K+A galaxies observed at higher redshift. While the outer
351: disk of NGC~4522 shows a K+A spectrum, a global spectrum of this
352: galaxy does \emph{not} show the same signature, since it contains
353: emission lines from the star-forming inner disk and absorption lines
354: mostly from the stripped outer stellar disk \citep{gavazzi04}. This
355: makes it different from the K+A galaxies observed at higher redshift
356: and more similar to the ``e(a)'' class \citep{dressler99}. Spectral
357: modeling \citep{shioya02} has suggested that the e(a) class may evolve
358: into K+A galaxies, but some observations (i.e. \citealp{balogh05})
359: indicate that e(a) galaxies and K+A galaxies are morphologically
360: distinct. There may be more than one way to make a K+A and e(a)
361: galaxies, but the present observations show that ram pressure
362: stripping \emph{can} create a K+A or e(a) spectrum, either from total
363: stripping (K+A) or partial stripping (e(a)). This shows that ram
364: pressure is effective in rapidly terminating star formation and is
365: capable of creating the post star formation spectra observed at higher
366: redshift. It may be that, in the case of Virgo, the ICM is not dense
367: enough to completely strip the gas from a massive galaxy in a single
368: passage so that a central star-forming gas disk will remain after this
369: stripping event ends. In the largest clusters, the ICM density is
370: higher by a factor of $\sim 10$ and galaxy velocities are higher by a
371: factor of $\sim 2$; this means that ram pressure is $\sim 40$ times
372: higher in these clusters. In such clusters, it is likely that a
373: galaxy like NGC~4522 would be completely stripped of its gas. In
374: fact, \citet{poggianti04} have observed a striking correlation between
375: K+A galaxies and the locations of X-Ray structure in the Coma Cluster,
376: suggesting that we are observing such transformations there. Both
377: these observations and our observation of NGC~4522 indicate that
378: cluster processes are important for the creation of some K+A spectra.
379: While K+A/e(a) galaxies may not all be created in clusters, it appears
380: that at least one route for their formation is the termination of star
381: formation by ISM-ICM stripping. Moreover, it appears that stripping
382: can drive this transformation well outside the cluster core.
383:
384: \vspace{2mm}
385:
386: We gratefully acknowledge the indispensable discussions and advice of
387: Jim Rose, which were critical at the early stages of this paper. We
388: also would like to thank Pieter van Dokkum, Jacqueline van Gorkom, and
389: Bernd Vollmer for comments that served to clarify the discussion.
390: Finally, we wish to thank the referee for insightful comments and
391: suggestions. This research is supported by NSF grant AST 00-71251.
392:
393: \begin{thebibliography}
394:
395: \bibitem[Balogh et al.(2005)]{balogh05} Balogh, M.~L., Miller,
396: C., Nichol, R., Zabludoff, A., \& Goto, T.\ 2005, \mnras, 360, 587
397:
398: \bibitem[Bekki et al.(2005)]{bekki05} Bekki, K., Couch, W.~J.,
399: Shioya, Y., \& Vazdekis, A.\ 2005, \mnras, 359, 949
400:
401: \bibitem[Bershady et al(2004)]{bershady04} Bershady, M.~A.,
402: Andersen, D.~R., Harker, J., Ramsey, L.~W., \& Verheijen, M.~A.~W.\ 2004,
403: \pasp, 116, 565
404:
405: \bibitem[B\"{o}hringer et al.(1994)]{bohringer94} B\"{o}hringer, H., Briel,
406: U.~G., Schwarz, R.~A., Voges, W., Hartner, G., \& Trumper, J.\ 1994, \nat,
407: 368, 828
408:
409: \bibitem[Bruzual \& Charlot(2003)]{bc03} Bruzual, G., \&
410: Charlot, S.\ 2003, \mnras, 344, 1000
411:
412:
413: \bibitem[Burstein et al.(2005)]{burstein05} Burstein, D., Ho,
414: L.~C., Huchra, J.~P., \& Macri, L.~M.\ 2005, \apj, 621, 246
415:
416: \bibitem[Calzetti(1997)]{calzetti97} Calzetti, D.\ 1997, American
417: Institute of Physics Conference Series, 408, 403
418:
419: \bibitem[Christlein \& Zabludoff(2004)]{christlein04} Christlein, D.,
420: \& Zabludoff, A.~I.\ 2004, \apj, 616, 192
421:
422: \bibitem[Dressler et al.(1997)]{dressler97} Dressler, A., et al.\
423: 1997, \apj, 490, 577
424:
425: \bibitem[Dressler et al.(1999)]{dressler99} Dressler, A., Smail,
426: I., Poggianti, B.~M., Butcher, H., Couch, W.~J., Ellis, R.~S., \& Oemler,
427: A.~J.\ 1999, \apjs, 122, 51
428:
429: \bibitem[Dressler \& Gunn(1983)]{dg83} Dressler, A., \&
430: Gunn, J.~E.\ 1983, \apj, 270, 7
431:
432: \bibitem[Dupke \& Bregman(2006)]{dupke06} Dupke, R.~A., \&
433: Bregman, J.~N.\ 2006, \apj, 639, 781
434:
435: \bibitem[Faber et al.(1985)]{faber85} Faber, S.~M., Friel,
436: E.~D., Burstein, D., \& Gaskell, C.~M.\ 1985, \apjs, 57, 711
437:
438: \bibitem[Gavazzi et al (2004)]{gavazzi04} Gavazzi G., Zaccardo A.,
439: Sanvito G., Bonfanti C., Boselli A. 2004, A\&A, 417, 499
440:
441: \bibitem[Goto(2005)]{goto05} Goto, T.\ 2005, \mnras, 357, 937
442:
443: \bibitem[Heinz et al(2006)]{heinz06} Heinz, S., Brueggen, M., Young,
444: A., \& Levesque, E. 2006, \mnras, submitted, astro-ph/0606664
445:
446: \bibitem[Kenney et al(2004)]{kenney04}Kenney, J.D.P., van Gorkom,
447: J.H., \& Vollmer, B. 2004, \aj, 127, 3361
448:
449: \bibitem[Kenney et al(2006)]{kenney06}Kenney, J.D.P., et al. 2006, in
450: prep.
451:
452: \bibitem[Kenney \& Koopmann(1999)]{kk99} Kenney, J.~D.~P.,
453: \& Koopmann, R.~A.\ 1999, \aj, 117, 181
454:
455: \bibitem[Koopmann \& Kenney(2004)]{koopmann04} Koopmann, R.~A., \&
456: Kenney, J.~D.~P.\ 2004, \apj, 613, 851
457:
458: \bibitem[Leitherer et al.(1999)]{leitherer99} Leitherer, C., et
459: al.\ 1999, \apjs, 123, 3
460:
461: \bibitem[Martin et al.(2005)]{martin05} Martin, D.~C., et
462: al.\ 2005, \apjl, 619, L1
463:
464: \bibitem[Martins et al.(2005)]{martins05} Martins, L.~P.,
465: Delgado, R.~M.~G., Leitherer, C., Cervi{\~n}o, M., \& Hauschildt, P.\ 2005,
466: \mnras, 358, 49
467:
468: \bibitem[Oemler (1974)]{oemler74} Oemler, A. 1974, \apj, 194, 1
469:
470: \bibitem[Poggianti(2004)]{poggianti04a} Poggianti, B.~M.\ 2004,
471: Clusters of Galaxies: Probes of Cosmological Structure and Galaxy
472: Evolution, 245
473:
474: \bibitem[Poggianti et al.(2004)]{poggianti04} Poggianti, B.~M.,
475: Bridges, T.~J., Komiyama, Y., Yagi, M., Carter, D., Mobasher, B., Okamura,
476: S., \& Kashikawa, N.\ 2004, \apj, 601, 197
477:
478: \bibitem[Schindler et al.(1999)]{schindler99} Schindler, S.,
479: Binggeli, B., B\"{o}hringer, H.\ 1999, \aap, 343, 420
480:
481: \bibitem[Schlegel et al.(1998)]{schlegel98} Schlegel, D.~J.,
482: Finkbeiner, D.~P., \& Davis, M.\ 1998, \apj, 500, 525
483:
484: \bibitem[Shibata et al.(2001)]{shibata01} Shibata, R.,
485: Matsushita, K., Yamasaki, N.~Y., Ohashi, T., Ishida, M., Kikuchi, K.,
486: B{\"o}hringer, H., \& Matsumoto, H.\ 2001, \apj, 549, 228
487:
488: \bibitem[Shioya et al.(2002)]{shioya02} Shioya, Y., Bekki, K., Couch,
489: W.~J., \& De Propris, R.\ 2002, \apj, 565, 223
490:
491: \bibitem[Tran et al.(2003)]{tran03} Tran, K.-V.~H., Franx, M.,
492: Illingworth, G., Kelson, D.~D., \& van Dokkum, P.\ 2003, \apj, 599, 865
493:
494: \bibitem[Tran et al.(2004)]{tran04} Tran, K.-V.~H., Franx, M.,
495: Illingworth, G.~D., van Dokkum, P., Kelson, D.~D., \& Magee, D.\ 2004,
496: \apj, 609, 683
497:
498: \bibitem[Vollmer et al(2004)]{vollmer04} Vollmer, B., Beck, R.,
499: Kenney, J.~D.~P., \& van Gorkom, J.~H.\ 2004, \aj, 127, 3375
500:
501: \bibitem[Vollmer et al.(2006)]{vollmer06} Vollmer, B., Soida, M.,
502: Otmianowska-Mazur, K., Kenney, J.D.P., van Gorkom, J.H., \&
503: Beck, R. 2006, A\&A, in press, astro-ph/0603854
504:
505: \bibitem[Zabludoff et al.(1996)]{zabludoff96} Zabludoff, A.~I.,
506: Zaritsky, D., Lin, H., Tucker, D., Hashimoto, Y., Shectman, S.~A., Oemler,
507: A., \& Kirshner, R.~P.\ 1996, \apj, 466, 104
508:
509: \end{thebibliography}
510:
511: \end{document}
512: