1: \documentstyle[emulateapj,psfig,apjfonts]{article}
2: %\documentstyle[11pt,aasms4,psfig]{article}
3: %
4: \def\gs{\mathrel{\raise0.35ex\hbox{$\scriptstyle >$}\kern-0.6em
5: \lower0.40ex\hbox{{$\scriptstyle \sim$}}}}
6: \def\ls{\mathrel{\raise0.35ex\hbox{$\scriptstyle <$}\kern-0.6em
7: \lower0.40ex\hbox{{$\scriptstyle \sim$}}}}
8: %
9: %\addtolength{\topmargin}{0.3in}
10: \addtolength{\textheight}{0.1in}
11: \submitted{Received: 2000 March 26; accepted 2000 May 9}
12: %\submitted{\hbox{\sc DRAFT: \today}}
13: \lefthead{Ivison et al.}
14: \righthead{A candidate submm proto-cluster at $z$ = 3.8}
15:
16: \begin{document}
17: %\doublespace
18:
19: \title{An excess of submm sources near 4C\,41.17: a
20: candidate proto-cluster at $z$=3.8?}
21:
22: \author{R.\,J.\ Ivison,$\!$\altaffilmark{1,2}
23: J.\,S.\ Dunlop,$\!$\altaffilmark{3} Ian Smail,$\!$\altaffilmark{4,5}
24: Arjun Dey\altaffilmark{6},
25: Michael C. Liu\altaffilmark{7} \& J.\,R.\ Graham\altaffilmark{7}
26: }
27: \affil{\small 1) Department of Physics \& Astronomy, University College
28: London, Gower Street, London WC1E 6BT, UK}
29: \affil{\small 3) Institute for Astronomy, University of Edinburgh, Blackford
30: Hill, Edinburgh EH9 3HJ, UK}
31: \affil{\small 4) Department of Physics, University of Durham, South Road,
32: Durham DH1 3LE, UK}
33: \affil{\small 6) NOAO, 950 N.\ Cherry Ave, Tucson AZ 85719}
34: \affil{\small 7) Astronomy Department, University of California at Berkeley,
35: CA 94720}
36:
37: \setcounter{footnote}{5}
38:
39: \altaffiltext{2}{PPARC Advanced Fellow.}
40: \altaffiltext{5}{Royal Society University Research Fellow.}
41:
42: \begin{abstract}
43: Biased galaxy-formation theories predict that massive galaxies at high
44: redshifts should act as signposts to high-density environments in the
45: early universe, which subsequently evolve into the cores of the
46: richest clusters seen at the present day. These regions are
47: characterised by over-densities of young galaxies, perhaps including a
48: population of dusty, interaction-driven starbursts --- the progenitors
49: of massive cluster ellipticals. By searching for this population at
50: submillimeter (submm) wavelengths we can therefore test both galaxy-
51: and structure-formation models. We have undertaken such a search in
52: the field of a $z=3.8$ radio galaxy, 4C\,41.17, with the SCUBA submm
53: camera. Our extremely deep 450- and 850-$\mu$m maps reveal an
54: order-of-magnitude over-density of luminous submm galaxies compared to
55: typical fields (the likelihood of finding such an over-density in a
56: random field is $<2 \times 10^{-3}$). The SCUBA galaxies have
57: bolometric luminosities, $>10^{13}\,$L$_\odot$, which imply
58: star-formation rates (SFRs) consistent with those required to form a
59: massive galaxy in only a few $10^8$ years. We also note that this
60: field exhibits an over-density of extremely red objects (EROs), some
61: of which may be associated with the submm sources, and Lyman-break
62: galaxies. We propose that the over-densities of both submm and ERO
63: sources in this field represent young dusty, starburst galaxies
64: forming within a proto-cluster centered on the radio galaxy at
65: $z=3.8$, which is also traced by a less-obscured population of
66: Lyman-break galaxies.
67: \end{abstract}
68:
69: \keywords{cosmology: observations --- galaxies: evolution ---
70: galaxies: formation --- infrared: galaxies --- radio: galaxies}
71:
72: %\newpage
73:
74: %
75: %
76: %
77: \section{Introduction}
78:
79: Galaxy-formation theories are developing rapidly and have claimed some
80: success at reproducing the properties of galaxies in the local
81: universe (Cole et al.\ 1994). However, at the moment their predictions
82: of the early evolution of galaxies are comparatively
83: untested. Moreover, the areas where the theoretical predictions are
84: most reliable tend to be those where the observational tests are most
85: difficult at high redshift, e.g., the evolution in the mass function
86: of galaxies (White \& Frenk 1991). One testable prediction of
87: hierarchical galaxy-formation models at high redshifts deals with the
88: clustering behaviour of massive galaxies at early epochs: such
89: galaxies are expected to cluster strongly in regions of highest
90: density, areas which should subsequently form the cores of massive
91: clusters of galaxies in the local universe (Kaiser 1984; Baron \&
92: White 1987; Efstathiou \& Rees 1988; Kauffmann et al.\ 1999). Thus by
93: searching for over-densities of massive galaxies in the distant
94: universe we can test the predictions of hierarchical models {\it as
95: well as} investigating the formation and evolution of the luminous
96: galaxies seen in present-day clusters. However, to achieve this we
97: must find a method of selecting the densest regions in the early
98: universe.
99:
100: One possible technique is to use massive galaxies as signposts for
101: high-density environments at high redshifts. Observations of distant,
102: luminous radio galaxies have led to the suggestion that they are
103: massive ellipticals (Matthews, Morgan \& Schmidt 1964; Lilly \&
104: Longair 1984), a belief confirmed recently by {\it Hubble Space
105: Telescope} ({\it HST}) studies which show they possess $r^{1/4}$-law
106: profiles characteristic of elliptical galaxies (McLure et al.\ 1999;
107: Zirm et al.\ 2000). Locally these massive elliptical galaxies typical
108: reside in galaxy clusters and hence luminous radio galaxies should
109: make good markers to search for the progenitors of rich galaxy
110: clusters at high redshifts.
111:
112: The picture of hierarchical formation of massive galaxies and their
113: environments is supported on small scales by recent rest-frame optical
114: imaging of radio galaxies (van Breugel et al.\ 1998; Pentericci et
115: al.\ 1998): at the earliest epochs ($z\ge 4$) there is evidence of
116: diffuse emission on large scales (5--10$\,$$''$) and sub-clumps
117: similar in scale ($\sim10$\,kpc) to radio-quiet star-forming galaxies;
118: these then appear to evolve into more compact structures by $z\sim
119: 2$. Interestingly, on somewhat larger scales there is also an apparent
120: 10--100-fold increase in the surface density of EROs --- some of which
121: are dusty, star-forming galaxies akin to local ULIRGs (Dey et al.\
122: 1999; Smail et al.\ 1999) --- around high-redshift radio galaxies
123: (HzRGs) and quasars as compared with the field (Hu \& Ridgway 1994;
124: Elston, Rieke \& Rieke 1988; Arag\'on-Salamanca et al.\ 1994; Dey,
125: Spinrad \& Dickinson 1995; Yamada et al.\ 1997). On similar scales,
126: the distorted radio morphologies and high rotation measures of some
127: HzRGs suggest they reside in high-density environments (Carilli et
128: al.\ 1997) with extended X-ray emission detected around one example,
129: 1138$-$262 at $z=2.2$ (Carilli et al.\ 1998).
130:
131: %
132: % Figure 1
133: %
134: \begin{figure*}
135: \centerline{\psfig{file=f1.eps,width=7.3in,angle=0}}
136:
137: \noindent{\scriptsize
138: {\sc Fig.~1.}---a) Full 850-$\mu$m image of the $z=3.8$ radio galaxy,
139: 4C\,41.17, at the original $\sim$14$''$ resolution. Negative features
140: are due to the chopped/nodded observing procedure. Contours are plotted
141: at $3, 4, 5, 6, 7 \times 1.5\,$mJy\,beam$^{-1}$. b) 450-$\mu$m image of
142: 4C\,41.17, smoothed to a resolution of 10$''$ FWHM. Contours are
143: plotted at $3, 4 \times 8.6\,$mJy per 10$''$ beam$^{-1}$. c) Central,
144: cleaned portion of the 850-$\mu$m image. Contours are plotted at $3, 4
145: \ldots 15 \times 0.8\,$mJy\,beam$^{-1}$, where the beam has been smoothed
146: to 20$''$ in the background regions.
147:
148: \vspace*{-1mm}
149: }
150: \end{figure*}
151:
152: Confirming that HzRGs reside in high-density environments would also
153: provide substantial insight into galaxy evolution in these regions,
154: especially the formation of giant ellipticals in the hierarchical
155: models already discussed. Optical/IR studies have targeted the
156: evolution of ellipticals in clusters at $z < 1$ (Arag\'on-Salamanca et
157: al.\ 1993; Stanford, Eisenhardt \& Dickinson 1998; Bower, Lucey \&
158: Ellis 1992; Ellis et al.\ 1997); theoretical work has investigated
159: their evolution in the context of hierarchical formation in
160: high-density regions (Kauffmann \& Charlot 1998; Baugh et al.\
161: 1999). Pinpointing the progenitors of ellipticals at very high
162: redshifts would allow straightforward tests of the latter models and
163: would enable the time-line of the optical/IR cluster studies to be
164: extended beyond $z\sim 1.3$ (Stanford et al.\ 1997, 1998; Rosati et
165: al.\ 1999; Liu et al.\ 2000).
166:
167: In the hierarchical picture, lower-mass structures will accrete onto
168: the high-mass peak of the proto-cluster identified with the radio
169: galaxy, the final assembly of this structure achieved through a series
170: of mergers which will strongly affect the galaxies residing within the
171: dark matter halo, triggering intense starbursts analogous to mergers
172: in the local universe (Sanders \& Mirabel 1996). The dust created in
173: these starbursts will absorb a considerable fraction of the UV/optical
174: light emitted by young stars, re-emitting it in the rest-frame
175: far-IR. An extreme and highly obscured starburst with a SFR of $\sim
176: 1000\,$M$_\odot$yr$^{-1}$ would have a bolometric luminosity, $L_{\rm
177: bol}$, of $\sim 10^{13}\,$L$_\odot$, and the majority of this would
178: appear at rest-frame far-IR wavelengths (Ivison et al.\ 1998).
179:
180: The negative $K$-correction for dust emission in the submm passband
181: (Blain \& Longair 1993) means that a luminous, dusty starburst with
182: $L_{\rm bol} \sim 10^{13}\,$L$_\odot$ would have an 850-$\mu$m flux of
183: $\sim 10\,$mJy if observed at {\it any} redshift between 1 and 10. The
184: advent of sensitive arrays working at submm wavelengths, in particular
185: the SCUBA camera (Holland et al.\ 1999) on the James Clerk Maxwell
186: Telescope (JCMT) have enabled efficient and sensitive surveys of dusty
187: starbursts, probing out to high redshifts (Smail, Ivison \& Blain
188: 1997; Barger et al.\ 1998; Hughes et al.\ 1998; Eales et al.\ 1999).
189:
190: Thus by using SCUBA to undertake targeted observations of the
191: environments of known HzRGs, searching for over-densities of bright
192: submm sources, we can test whether HzRGs are located in the cores of
193: rich proto-clusters {\it and} constrain the formation epoch of massive
194: cluster galaxies. For targets at $z\gs 2$ the resolution and field of
195: view of SCUBA at 850$\,\mu$m means we are sensitive to star-forming
196: galaxies distributed on scales from 100 to 1000\,kpc --- well-matched
197: to the predicted virial radii of the most massive clusters at these
198: epochs (Jenkins et al.\ 2000).
199:
200: In this paper we present deep 450- and 850-$\mu$m maps of one of the
201: most distant and powerful known radio galaxies, 4C\,41.17 at $z=3.8$
202: (for which 1$''$ = 6.6\,$h_{50}^{-1}$\,kpc, with $h_{50} =
203: H_0/50$\,km\,s$^{-1}$\,Mpc$^{-1}$ and $q_0=0.5$). There is already
204: some evidence for a massive structure associated with 4C\,41.17.
205: Based upon a Lyman-break search using $UVR$ imaging, Lacy \& Rawlings
206: (1996) have suggested that there is a modest excess of high-redshift
207: galaxies in this region, having found six candidate $z\gs 3.4$
208: galaxies in a 1.5-arcmin$^2$ field, roughly centered on 4C\,41.17.
209: They show that this density is slightly higher than that expected from
210: the field density found by Steidel, Pettini \& Hamilton (1995), but
211: the complications introduced by comparing their $UVR$ study with
212: Steidel et al.'s $\mathcal{UGR}$ survey, as well as issues of cosmic
213: variance (Steidel et al.\ 1999), make it difficult to assess the
214: over-density. Other evidence for a deep potential centered on the
215: radio galaxy includes the presence of an extended Ly$\alpha$ halo
216: reaching to $>$100\,kpc (Chambers, Miley \& van Breugel 1990; Dey
217: 1999).
218:
219: In the next section we present our submm, near-IR and optical
220: observations of this field and their reduction. We discuss our analysis
221: and results, in the context of supporting archival data at other
222: wavelengths, in \S3 and give our conclusions in \S4.
223:
224: %
225: %
226: %
227: \section{Observations and Data Reduction}
228:
229: \subsection{Submillimeter Observations}
230:
231: During 1998 October we used SCUBA to map a $\sim 2.5'$-diameter field
232: at 450- and 850-$\mu$m, centred on 4C\,41.17 (Fig.~1). The secondary
233: mirror followed a jiggle pattern designed to fully sample the image
234: plane, chopping by 45$''$ between the `on' and `off' positions at
235: 7\,Hz whilst the telescope nodded between the same positions every
236: 16\,s in an on--off--off--on pattern. The chop throw was chosen to
237: optimise calibration accuracy, minimise noise, maximise regions for
238: which both `on' and `off' beam information would be available, and
239: minimise the possibility of bright sources close to the field of view
240: contaminating the map. Two directions were used for the chopping:
241: north--south (N--S) as the target rose and set; east-west (E--W) when
242: the target was near transit. This method ensured the chop direction
243: was close to azimuthal whilst retaining the advantages of on-array
244: chopping. It also provided independent maps which could be cleaned
245: separately and coadded, reducing the risk of spurious detections and
246: of chopping consistently from one source onto another.
247:
248: %
249: % Figure 2
250: %
251: \begin{figure*}
252: %\centerline{\psfig{file=f2.eps,width=7.3in,angle=0}}
253: \centerline{\psfig{file=f2_compressed.eps,width=7.3in,angle=270}}
254:
255: \noindent{\scriptsize {\sc Fig.~2.}--- KPNO 2.1-m $K'$-band imaging
256: and {\it HST} F702W images of $25'' \times 25''$ regions around the
257: submm sources, smoothed with 0.5$''$- and 0.15$''$-FWHM Gaussians at
258: $K'$ and $F702W$ respectively. For HzRG850.1, the optical data are
259: from Keck\,{\sc ii} (\S2.2). EROs are marked with squares; 850-$\mu$m
260: positions are marked with 6$''$- or 8$''$-diameter black circles (an
261: ellipse for HzRG850.2); the 450-$\mu$m position of HzRG850.1 is marked
262: with a white circle.
263:
264: \vspace*{-1mm}
265: }
266: \end{figure*}
267:
268: The sky was exceptionally stable and transparent for the five nights
269: during which data were obtained. The total exposure time was 36\,ks
270: and the opacity at 850\,$\mu$m, measured every hour, was typically
271: 0.15--0.20, never rising above 0.25. Flux calibration was accomplished
272: using beam maps of Mars. Nightly calibration factors were stable:
273: consistent at the 5\% level (r.m.s.) at 850\,$\mu$m. Even at
274: 450\,$\mu$m, factors were stable at the 10\% level. Absolute flux
275: calibration should be accurate to $\sim 10$\%.
276:
277: Maps were created using {\sc surf} (Jenness \& Lightfoot 1997). In
278: order to extract reliable source positions and flux densities, the
279: images (Fig.~1) and the symmetric $-1, +2, -1$ zero-flux beam (which
280: arises from chopping and nodding) were deconvolved using a modified
281: version of the {\sc clean} algorithm (Hogbom 1974) as follows. The
282: E--W- and N--S-chopped maps were treated separately for the purpose of
283: deconvolution. First, both the image and the appropriate beam (from
284: contemporaneous maps of blazars) were convolved with a Gaussian (FWHM
285: 14$''$, similar to the core of the JCMT beam at 850\,$\mu$m). The
286: smoothed image was then cleaned using the smoothed beam, down to a
287: level equivalent to twice the r.m.s.\ noise of the smoothed data,
288: using a loop-gain of 0.1. This has the effect of gradually removing
289: the positive core of a real source from the data, while at the same
290: time filling in its negative sidelobes. When the process is complete,
291: the original image is reduced to a map of background noise (still
292: containing real sources which are too faint to have been identified as
293: significant peaks by the {\sc clean} algorithm) while the absolute
294: values of removed flux are stored as a set of delta functions (in the
295: case of unresolved sources) located at the positions of the original
296: source peaks.
297:
298: The restored images were produced by halving the size of the delta
299: functions then convolving them with a Gaussian (FWHM 14$''$) and
300: adding them back into the residual noise image left by the cleaning
301: process. The result is a restored image with the angular resolution of
302: the original but free from negative sidelobes associated with
303: significant sources. One advantage of this process is that it should
304: resurrect any significant sources whose flux has been depressed (or
305: annihilated) because they happen to lie in the negative sidelobe of
306: another source, though only sources within 70$''$ of the field centre
307: are treated properly by the cleaning procedure. Sources which were
308: produced in both the E--W- and N--S-chopped, restored sub-images, with
309: positions differing by less than 3$''$, can be regarded as robust
310: detections.
311:
312: \subsection{Near-Infrared and Optical Observations}
313:
314: We used the Ohio State/NOAO IR Spectrometer (ONIS) on the 2.1-m
315: telescope of the Kitt Peak National Observatory to obtain $K$-band
316: images of the 4C\,41.17 field covering a $5.6'\times8.0'$ field at
317: 0.34\arcsec/pixel sampling. The observations were taken in 0.9\arcsec\
318: seeing on the nights of 1998 November 14, 15 and 1999 March 20.
319: Observations were made using a random dither pattern, with multiple
320: 60\,s integrations at each pointing and a combined exposure time of
321: 16.3\,ks. The data were dark subtracted, linearized, flat fielded,
322: and the final mosaic (Fig.~2) was constructed using a modified version
323: of {\sc dimsum} in {\sc iraf}\footnote{{\sc iraf} is distributed by
324: the National Optical Astronomy Observatories.}. The photometric
325: calibration was accomplished using observations of several {\it HST}
326: IR standard stars (Persson et al.\ 1998) and checked against the
327: photometry of Graham et al.\ (1994) for the sources in common. The
328: images were astrometrically calibrated using the U.S.\ Naval
329: Observatory A2.0 Catalog. The final $K$-band mosaic has a seeing FWHM
330: of 1.0\arcsec\ and reaches a $5\sigma$ limit of $K=19.2$ in a
331: 2\arcsec\ diameter aperture.
332:
333: In addition to these wide-field observations, we also obtained deep
334: $JHK$ images of the fields of two of the brighter SCUBA sources
335: (HzRG850.1 and HzRG850.2; see below) using the Near-IR Camera (NIRC,
336: Matthews \& Soifer 1994) at the W.\,M.\ Keck Observatory in
337: photometric conditions on the nights of 1999 March 24 and 25. NIRC
338: has a 38\arcsec\ field of view with 0.15\arcsec/pixel sampling and the
339: observations were taken with a random dither pattern, with a 120-s
340: exposure per pointing. Over the two nights, we obtained total
341: integrations on HzRG850.1 of 1.2\,ks, 2.4\,ks and 1.2\,ks in $JHK$
342: respectively, with 2.4\,ks, 1.2\,ks and 2.4\,ks in $JHK$ on the
343: HzRG850.2 field. The frames were reduced and combined in a standard
344: manner and we show the resulting composite color images in Fig.~3.
345: The seeing measured off the final stacked frames is 0.5--0.6\arcsec
346: and the frames were calibrated using the faint standards from Persson
347: et al.\ (1998) yielding detection limits of $J\sim 24$, $H\sim 22.5$
348: and $K\sim 21$.
349:
350: $V$-band images of the 4C\,41.17 field were obtained using the Low
351: Resolution Imaging Spectrometer (LRIS; Oke et al. 1995) on Keck\,{\sc
352: ii} on 1998 September 19. The total exposure time was 2.7\,ks, broken
353: into 9 dithered sub-exposures. Conditions were non-photometric, but
354: the data were calibrated using magnitudes of objects in the field from
355: Lacy \& Rawlings (1996). The final image reaches a $5\sigma$ limiting
356: magnitude of $V=26.3$ in a 2\arcsec\ diameter aperture.
357:
358: Finally, we have also made use of the archival {\it HST} WFPC2 image
359: of this field (GO\#5511). This comprises a total of 21.6\,ks
360: integration through the F702W filter and reaches a 3-$\sigma$ limiting
361: magnitude of $R=26.0$ (2\arcsec\ aperture, converted to Cousins $R$
362: using photometry of sources with $R-K\sim 4$ in Graham et al.\ 1994).
363:
364: These combined optical and near-IR observations identify several new
365: ERO sources to add to the two originally found by Graham et al.\
366: (1994). In total we identify at least five objects with extreme
367: colors, $R-K\geq 6$ or $V-K\geq 7$, brighter than $K=20$ within
368: roughly a 2 arcmin$^2$ field. This surface density is nearly an order
369: of magnitude higher than that seen in blank fields (e.g.\ Thompson et
370: al.\ 1999) and as we discuss below several of these objects maybe
371: associated with the submm sources in this field.
372:
373: %
374: % Figure 3
375: %
376: \begin{figure*}
377:
378: \centerline{\psfig{file=f3.eps,width=7.4in,angle=0}}
379:
380: \noindent{\scriptsize {\sc Fig.~3.}--- Composite color images of the
381: fields of HzRG850.1 and HzRG850.2, constructed from the Keck $JHK$
382: imaging. We label the possible candidate counterparts to the submm
383: sources as well as several other sources in the close vicinity. The
384: field of HzRG850.1 is $54\arcsec \times 47\arcsec$, while HzRG850.2 is
385: 47\arcsec\ square. Orientations are shown.
386:
387: \vspace*{-1mm}
388: }
389: \end{figure*}
390:
391: %
392: %
393: %
394: \section{Results and Discussion}
395:
396: \subsection{Submillimeter sources}
397:
398: At 850\,$\mu$m, the cleaned portion of the map (4.3\,arcmin$^2$)
399: reaches a 4-$\sigma$ sensitivity of 3.2\,mJy\,beam$^{-1}$
400: (6.0\,mJy\,beam$^{-1}$ for a further 1.6\,arcmin$^2$), sufficient to
401: detect galaxies with SFRs of a few 100\,M$_\odot$ yr$^{-1}$ at the
402: redshift of 4C\,41.17.
403:
404: In the cleaned image we detect three sources well above the 4-$\sigma$
405: flux limit with positions differing by $<3''$ in the sub-images
406: (Table~1); of these, one is associated with the radio galaxy, as
407: determined previously by Dunlop et al.\ (1994). The others are new
408: detections and one of them, HzRG850.1, is the brightest 850-$\mu$m
409: source yet reported. As we discuss in \S3.2, this represents a large
410: over-density compared to the expected blank-field count of 0.1--0.3
411: per SCUBA field.
412: %\newpage
413:
414: %
415: % TABLE 1
416: %
417: {\scriptsize
418: \begin{center}
419: {\sc Table 1}
420:
421: \vspace{0.1cm}
422: {\sc Positions and Flux Densities of Submm Sources}
423:
424: \vspace{0.3cm}
425: \begin{tabular}{lllrc}
426: \hline\hline
427: \noalign{\smallskip}
428: {IAU} &{~~~~~~~~RA$^{\rm a}$} &{~~~~~~Dec.$^{\rm a}$} &{$S_{850\mu{\rm m}}^{\rm b}$}~~ &{$S_{450\mu{\rm m}}^{\rm b}$}~~ \cr
429: {name}&{~~~(h min s $\pm$ s)} &{~~($^{\circ}$ $'$ $''$ $\pm$ $''$)}&{(mJy)}~ & {(mJy)}~ \cr
430: \hline
431: \noalign{\smallskip}
432: HzRG850.1$^{\rm c}$ &
433: $06\ 50\ 51.2\ \pm 0.3$ &
434: $+41\ 30\ 05\ \pm 3$ &
435: $15.6 \pm 1.8$ &
436: $34.1 \pm 9.3$ \cr
437: &
438: $06\ 50\ 51.0\ \pm 0.3^{\rm d}$ &
439: $+41\ 30\ 08\ \pm 3^{\rm d}$&
440: &
441: \cr
442: HzRG850.2$^{\rm c}$ &
443: $06\ 50\ 49.3\ \pm 0.4$ &
444: $+41\ 31\ 27\ \pm 3$ &
445: $8.7 \pm 1.2$ &
446: ---$^{\rm e}$ \cr
447: HzRG850.3$^{\rm f}$ &
448: $06\ 50\ 59.3\ \pm 0.4$ &
449: $+41\ 30\ 45\ \pm 3$ &
450: $6.5 \pm 1.6$ &
451: ---$^{\rm e}$ \cr
452: 4C\,41.17$^{\rm c}$ &
453: $06\ 50\ 52.14$ &
454: $+41\ 30\ 30.8$ &
455: $11.0 \pm 1.4$ &
456: $35.3 \pm 9.3$ \cr
457: HzRG850.4 &
458: $06\ 50\ 54.0\ \pm 0.4$ &
459: $+41\ 29\ 39\ \pm 4$ &
460: $2.8 \pm 0.8$ &
461: $3\sigma < 26$ \cr
462: HzRG850.5 &
463: $06\ 50\ 55.3\ \pm 0.4$ &
464: $+41\ 30\ 20\ \pm 4$ &
465: $2.4 \pm 0.8$ &
466: $3\sigma < 26$ \cr
467: \noalign{\smallskip}
468: \noalign{\hrule}
469: \noalign{\smallskip}
470: \end{tabular}
471: \end{center}
472:
473: \vspace*{-2mm}
474: \noindent
475: {\sc Notes:} $a$) 850-$\mu$m position (J2000); $b$) Photometry measurements,
476: taking the 15$''$-diameter aperture-corrected value at 850\,$\mu$m and
477: the 7.5$''$-diameter aperture-corrected value at 450\,$\mu$m; errors
478: include a 10\% contribution from the uncertainty in absolute flux
479: calibration; $c$) Positions and fluxes were measured from the cleaned
480: image. The 850-$\mu$m maps were shifted 0.3$''$ west, 2.4$''$ north
481: (1.0$''$ west, 0.4$''$ south for the 450-$\mu$m map) such that the
482: submm emission from 4C\,41.17 coincided with the radio core; $d$)
483: 450-$\mu$m position; $e$) Source lies off the 450-$\mu$m map; $f$)
484: Lies outside of 70$''$-radius cleaned zone.
485:
486: }
487: \vspace*{2mm}
488:
489: The flux measurements of HzRG850.1 (Table~1) suggest that it is
490: resolved. Aperture-corrected fluxes (based on Mars, with a 4.5$''$
491: diameter) increase with aperture size, showing a 50\% increase in flux
492: when going from a 5$''$- to a 15$''$-diameter aperture. This indicates
493: that HzRG850.1 subtends $>4.5''$, consistent with the result of
494: deconvolving it from the best-fit Gaussian profile for the
495: instrumental beam which suggests a FWHM of $\sim$10$''$. The
496: 850-$\mu$m emission from 4C\,41.17 and HzRG850.2 may also be resolved;
497: their beam-corrected FWHM are $\sim 7''$ and $\sim 2''$ respectively.
498:
499: Two other 850-$\mu$m sources are also {\it marginally} detected
500: ($\ge3\sigma$) in the cleaned map (one of these is also seen in the
501: uncleaned map) and one further source is detected towards the eastern
502: edge of the larger, uncleaned map: it reaches 6$\sigma$ at its peak
503: and it is seen in the E--W- and N--S-chopped sub-images so we consider
504: it a robust detection, though its proximity to the map edge means that
505: caution should be exercised regarding its precise flux and position
506: (again, see Table~1).
507:
508: Of the four robust 850-$\mu$m sources, two are also detected at
509: 450$\,\mu$m: HzRG850.1 and 4C\,41.17. The positional coincidence with
510: the 850-$\mu$m sources leaves little doubt of their reality, whilst
511: the low flux levels indicate high redshifts. HzRG850.2 is not
512: detected, although it lies on the extreme edge of the 450-$\mu$m
513: map. Neither of the two marginal 850-$\mu$m sources are detected at
514: 450$\,\mu$m, and the bright eastern source does not lie on the
515: 450-$\mu$m map. One further faint ($3\sigma$) source appears in the
516: 450-$\mu$m map; it has no counterpart at 850\,$\mu$m.
517:
518: %
519: %
520: %
521: \subsection{Source counts and clustering}
522:
523: Clearly, this field shows startling evidence for extremely luminous
524: submm galaxies, possibly in a structure associated with the radio
525: galaxy, 4C\,41.17 at $z=3.8$. We first discuss the strength of this
526: over-density and the evidence for excesses of other classes of
527: proposed high-redshift galaxies in this field, before going on to
528: examine the available information on the redshifts of the submm
529: galaxies to determine if it is consistent with them lying at the same
530: redshift as 4C\,41.17 and hence whether they can be placed in the
531: heart of the proposed proto-cluster.
532:
533: How does the submm galaxy density towards 4C\,41.17 compare with that
534: expected in blank-fields (Smail et al.\ 1997; Hughes et al.\ 1998;
535: Barger et al.\ 1999; Blain et al.\ 1999; Eales et al.\ 1999)? {\it
536: Excluding} 4C\,41.17, the density of $S(850\,\mu{\rm m}) \ge 8\,$-mJy
537: sources is $1220\pm 860$\,deg$^{-2}$; the weighted mean 8-mJy
538: blank-field count is an order of magnitude lower at $134\pm
539: 57$\,deg$^{-2}$. Delving slightly deeper, $S(850\,\mu{\rm m}) \ge
540: 6.5\,$-mJy is $1830\pm 860$\,deg$^{-2}$ versus $303\pm108$\,deg$^{-2}$
541: in blank fields. The observed source density at $S(850\,\mu{\rm m})
542: \ge 8$-mJy in the 4C\,41.17 is equal to that seen in blank fields at a
543: flux limit of only $S(850\,\mu{\rm m}) \sim 2.5\,$mJy, below our
544: 4$\sigma$ detection threshold.\footnote{This allows us to rule out the
545: alternative suggestion that the high density of submm galaxies results
546: from amplification bias due to a foreground mass structure, which also
547: amplifies the radio source. To achieve the necessary amplification
548: across the SCUBA field ($A\sim 3$), the lens would have to have a mass
549: of $\sim 10^{15}M_\odot$ within the field -- such a massive foreground
550: cluster would be easily visible.}
551:
552: Adopting the standard probabilistic methodology used in studies of
553: clustering and confusion (e.g.\ Lilly et al.\ 1999), it is a simple
554: exercise to determine the likelihood, $P$, of detecting so many bright
555: sources within the 4C\,41.17 field: $P<0.00166$, meaning that on
556: average we would have to observe more than 600 typical blank fields
557: before finding a configuration of the type seen towards
558: 4C\,41.17. Note that $P$ is given as an upper limit because HzRG850.1
559: is the brightest known 850-$\mu$m source and $S(850\,\mu{\rm m}) \ge
560: 15\,$mJy is thus unknown; we conservatively adopt [$S(850\,\mu{\rm m})
561: \ge 15\,$mJy] = [$S(850\,\mu{\rm m}) \ge 8\,$mJy].
562:
563: We note that the integrated 850-$\mu$m flux density in sources
564: brighter than 2\,mJy in this field is $\sim 3.3\times
565: 10^{-10}$\,W\,m$^{-2}$\,sr$^{-1}$, so we have resolved $\gs 90$\% of
566: the expected extragalactic background in the submm, based on its
567: detection by {\it COBE} (Fixsen et al.\ 1998), and would far exceed it
568: by $\sim 1$\,mJy.
569:
570: %
571: %
572: %
573: \subsection{Counterparts}
574:
575: Before discussing the optical and near-IR counterparts to the submm
576: sources around 4C\,41.17, we introduce a nomenclature for their
577: classification, analogous to that used for proto-stars, and building
578: upon the evolutionary scheme for ULIRGs proposed by Sanders et al.\
579: (1988). For operational purposes we base this scheme on the typical
580: depths achieved in follow-up observations ($I\sim 26$, $K\sim 21$) and
581: the properties of the submm galaxies discussed in Ivison et al.\
582: (1998, 2000), Hughes et al.\ (1998), Smail et al.\ (1999) and Soucail
583: et al.\ (1999). We define the following classes: Class~0, very-highly
584: obscured sources, where there is no plausible optical or near-IR
585: counterpart; Class~I, highly obscured sources, where only a near-IR
586: counterpart exists (often EROs --- Elston, Rieke \& Rieke 1988; Dey et
587: al.\ 1999; Smail et al.\ 1999); and Class~II, where an obvious optical
588: counterpart is seen (IIa: pure starburst; IIb: type-II AGN; IIc:
589: type-I AGN). The latter class may overlap with the most massive
590: examples of Lyman-break-selected objects (Adelberger \& Steidel 2000).
591: We note that Classes 0 and I may also include sources which are
592: fainter in the optical/near-IR by virtue of being at higher redshifts;
593: however, for the purposes of the discussion here we assume that the
594: different classifications arise solely from differences in the
595: obscuration of the sources.
596:
597: Turning to the optical and near-IR imaging of this field, we start by
598: discussing HzRG850.1. There are three possible optical or near-IR
599: counterparts to this source lying within or near the 450- and
600: 850-$\mu$m errors circles (Figs.~2 \& 3; see also Table~2). The
601: bluest of these sources is 850.1.K1 (Class~II), which falls within the
602: 450-$\mu$m error circle. Lacy \& Rawlings (1996) detect this source in
603: the $U$-band ($U-V \sim -0.9$; LR3 in their paper) and we therefore
604: conclude that 850.1.K1 is probably a low-redshift, blue galaxy and
605: unlikely to be associated with the SCUBA source. The remaining two
606: galaxies, 850.1.K2 and 850.1.K3 (both Class~I) have similar very red
607: colors, $V-K \sim 7$, with 850.1.K2 falling within both the 850-$\mu$m
608: and 450-$\mu$m error circles. The similar, very red colors of
609: 850.1.K2 and 850.1.K3 (Fig.~3; Table~2), combined with the fact that
610: the 850-$\mu$m source HzRG850.1 is spatially extended, suggests that
611: both these galaxies may be jointly responsible for the submm
612: emission. We also note some {\it very} faint emission in $V$ and $K$
613: to the west of 850.1.K2, and in $V$ alone to the south.
614:
615: For HzRG850.2, three faint galaxies are visible in the $K$ image
616: (Figs.~2 \& 3): the bluer of these, 850.2.K1, lies just within the
617: error ellipse, while the others, 850.2.K2 and 850.2.K3, are much
618: redder, $R-K \ge 6.4$, and lie on the edge of the nominal error
619: ellipse. In the deep {\it HST} WFPC2 F702W image, both 850.2.K1 and
620: 850.2.K2 are each resolved into two components, while 850.2.K3 is
621: undetected ($R>26$, $3\sigma$). For 850.2.K1 the sub-components both
622: appear to be galaxies, while in 850.2.K2 they may either be two
623: galaxies or a single system crossed by a dust lane. As with
624: HzRG850.1, it is plausible that both of these counterparts may be
625: contributing to the submm emission from HzRG850.2, given its
626: morphology (Fig.~1).
627:
628: %
629: % TABLE 2
630: %
631: %\begin{table}
632: {\scriptsize
633: \begin{center}
634: {\sc Table 2}
635:
636: \vspace{0.1cm}
637: {\sc Candidate Near-IR Identifications for Submm Sources}
638:
639: \vspace{0.3cm}
640: \begin{tabular}{lccccc}
641: \hline\hline
642: \noalign{\smallskip}
643: ID & R.A.\ (J2000) & Dec.\ (J2000) & $K$ & $V-K$ & $R-K$ \cr
644: \hline
645: \noalign{\smallskip}
646: 850.1.K1 & 06 50 51.12 & +41 30 09.2 & 20.0$\pm$0.4 & 4.5$\pm$0.4 & ... \cr
647: 850.1.K2 & 06 50 51.05 & +41 30 06.8 & 19.0$\pm$0.2 & 7.0$\pm$0.3 & ... \cr
648: 850.1.K3 & 06 50 51.29 & +41 29 58.6 & 19.4$\pm$0.3 & 6.9$\pm$0.3 & ... \cr
649: 850.2.K1 & 06 50 49.14 & +41 31 25.6 & 19.6$\pm$0.3 & 4.6$\pm$0.3 & 4.4$\pm$0.3 \cr
650: 850.2.K2 & 06 50 49.07 & +41 31 23.7 & 19.2$\pm$0.2 & 7.4$\pm$0.4 & 6.4$\pm$0.2 \cr
651: 850.2.K3 & 06 50 49.71 & +41 31 28.2 & 20.2$\pm$0.4 & $5\sigma>$6.1 & $5\sigma>5.3$ \cr
652: 850.3 & ... & ... & $5\sigma>20.0$ & ... & ... \cr
653: 850.4.K1 & 06 50 53.95 & +41 29 40.4 & 20.5$\pm$0.5 & ... & $5\sigma>$5.6 \cr
654: 850.5 & ... & ... & $5\sigma>20.0$ & ... & ... \cr
655: \noalign{\smallskip}
656: \noalign{\hrule}
657: \noalign{\smallskip}
658: \end{tabular}
659: \end{center}
660: }
661: %\end{table}
662:
663: HzRG850.3 unfortunately lies in a region where the $K$-band imaging is
664: relatively shallow (Fig.~2). No obvious counterpart is visible in the
665: KPNO $K$-band image down to $K=19.8$ ($3\sigma$). There is, however,
666: a very faint $V\gs 27$ source visible within the 850-$\mu$m error
667: circle. In addition, there are a pair of faint ($V\sim 25.6$, 26.0)
668: galaxies that lie $\sim 5.6\arcsec$ northeast of the SCUBA source
669: centroid. However, the faintness of these galaxies and the lack of any
670: useful color information renders it difficult to place much confidence
671: in these galaxies as likely candidates for the submm emission. We
672: conclude that HzRG850.3 must either be a Class~0 or I submm galaxy.
673:
674: The remaining SCUBA sources lie in comparatively shallow regions of
675: the near-IR and optical images and do not show unambiguous
676: identifications. HzRG850.4 has a very faint ($K\sim 20.5$) and
677: extended counterpart (850.4.K1) visible in the $K$-band image within
678: the 850-$\mu$m error circle (Fig.~2). The position of 850.4.K1 in the
679: $V$-band image is contaminated by the diffraction spike from a bright
680: star, but there is no obvious counterpart to this source visible in
681: the deep {\it HST} F702W image and we conclude that it is a Class~I
682: source. For HzRG850.5 there are no obvious sources visible in the $K$
683: or F702W image (Class 0 or I), and the $V$-band image is again
684: contaminated by the nearby star.
685:
686: In summary, HzRG850.1 and HzRG850.2 are likely to be associated with
687: EROs (i.e.\ Class I sources). HzRG850.3 and HzRG850.4 may be
688: associated with very faint galaxies, and HzRG850.5 has no obvious
689: counterpart (Class 0 or I). It is interesting that our success in
690: obtaining candidate identifications, in particular the association
691: with EROs, correlates with the observed flux density of the submm
692: emission. Is it possible that the brightest submm sources are
693: typically associated with Class~I or II counterparts (e.g.\ HR10, Dey
694: et al.\ 1999; SMM\,J09429+4658, Smail et al.\ 1999;
695: SMM\,J02399$-$0136, Ivison et al.\ 1998; SMM\,J14011+0252, Ivison et
696: al.\ 2000) whereas fainter submm sources typically have Class~0
697: counterparts (see also Hughes et al.\ 1998; Smail et al.\ 2000).
698:
699: %
700: %
701: %
702: \subsection{Redshift constraints}
703:
704: Finally, we discuss whether the submm sources can be unambiguously
705: placed at high redshifts, $z\gg 1$, and hence whether we can
706: strengthen the case for their association with 4C\,41.17. In the
707: absence of deep spectroscopy of the counterparts discussed in \S3.3,
708: we are reliant on other redshift-sensitive parameters. Two examples
709: are: $a$) the 450- to 850-$\mu$m flux ratio, which falls steadily as
710: the peak of the dust SED shifts through the 450-$\mu$m filter at
711: increasing redshift; and $b$) the 850-$\mu$m to 1.4-GHz flux ratio,
712: based upon the local far-IR/radio correlation for star-forming
713: galaxies, is a robust redshift indicator where $F_{\nu} \propto
714: \nu^{+3.5}$ at submm wavelengths, while $F_{\nu} \propto \nu^{-0.7}$
715: in the radio, yielding a flux ratio that rises initially as
716: $(1+z)^{+4.2}$ (Carilli \& Yun 1999, 2000; Blain 1999; see also Smail
717: et al.\ 2000).
718:
719: Dealing with the 450- to 850-$\mu$m flux ratio first, the observed
720: ratios for 4C\,41.17 and HzRG850.1 are $3.2\pm 0.8$ and $2.2\pm 0.6$,
721: consistent with their being at the same redshift, with best-fit values
722: and ranges of $z \sim 3.8$ ($2.8 < z < 5.1$) and $\sim 4.8$ ($3.5 < z
723: < 6.1$), respectively, based on extreme models of starbursts (Hughes
724: et al.\ 1998). These limits are supported by constraints from the
725: 850-$\mu$m to 1.4-GHz flux ratio. For HzRG850.1, where $S_{850\mu{\rm
726: m}}/S_{1.4{\rm GHz}} \ge 180$ (taking radio limits of $3\sigma <
727: 90$\,$\mu$Jy from the re-reduced radio image published originally by
728: Carilli, Owen \& Harris 1994), we arrive at a robust limit of $z\ge
729: 2$. The limits for HzRG850.2 and HzRG850.3 are slightly weaker, but
730: they are certainly not local galaxies.
731:
732: The extreme colors, $V-K \sim 7$, of the probable galaxy counterparts
733: to HzRG850.1 and HzRG850.2, as well as the faintness of the
734: counterparts to the remaining sources, suggests that these systems
735: also lie at high redshifts. The $K$-band magnitudes for the candidate
736: identifications quoted are fairly bright: if these galaxies do lie at
737: $z=3.8$ then they are very luminous, in excess of 50$L^*$, although we
738: note that luminosities close to this are found for the confirmed Class
739: II counterparts SMM\,J02399$-$0136 ($z=2.8$, Ivison et al.\ 1998) and
740: SMM\,J14011+0252 ($z=2.6$, Ivison et al.\ 2000).
741:
742: We conclude that the available constraints suggest that the
743: over-density of bright submm galaxies identified in the 4C\,41.17
744: field is likely to lie at $z>2.8$ and hence is consistent with being
745: associated with the radio galaxy at $z=3.8$ (though spectroscopic
746: confirmation is clearly a top priority). This suggests that in
747: addition to the proposed over-density of Lyman-break galaxies
748: identified around 4C\,41.17 by Lacy \& Rawlings (1996), we should also
749: add a comparably numerous population of highly-obscured and very
750: luminous submm galaxies.
751:
752: %
753: %
754: %
755: \section{Conclusions}
756:
757: We report the discovery of five new luminous submm sources within a
758: $2.5'$-diameter (1\,Mpc at $z=3.8$) area centered on the submm-bright
759: $z=3.8$ radio galaxy, 4C\,41.17. Three of these sources are as bright
760: or brighter than any other known blank-field submm galaxies. This
761: surface density of submm galaxies is an order of magnitude greater
762: than that expected from blank-field counts: we would need to map at
763: least 600 blank fields to find a chance configuration of such bright
764: sources. Using the available constraints, we suggest that this
765: over-density lies at $z>2.8$ and is therefore consistent with a
766: structure associated with 4C\,41.17.
767:
768: We note that this field exhibits an over-density above that expected
769: in blank fields, not only at submm wavelengths, but also in samples of
770: galaxies selected via the Lyman-break technique (Lacy \& Rawlings
771: 1996) and for their extremely red colors (EROs). As many as five of
772: the latter class of galaxies may be directly associated with the
773: luminous submm sources seen in this field.
774:
775: We introduce a classification scheme for the counterparts of submm
776: sources, based upon observations of well-studied SCUBA galaxies.
777: Within the framework of this scheme we propose that submm galaxies and
778: ERO starbursts represent different aspects of a single evolutionary
779: cycle. Following on from this, we suggest that the over-densities of
780: both the submm and ERO populations in this field represent young,
781: dusty starbursts forming within a proto-cluster centered on the radio
782: galaxy at $z=3.8$, which also hosts a population of less-obscured
783: Lyman-break galaxies.
784:
785: More observations are clearly needed to confirm the nature of this
786: over-density, to test our suggestion that it is associated with the
787: radio source at $z=3.8$ and, most interestingly, to investigate if it
788: represents a virialised structure. The most feasible test of the
789: latter issue will be to image this field using the {\it Chandra} or
790: {\it Newton} X-ray observatories to identify emission from the hot
791: intracluster medium confined within the potential well of the proposed
792: proto-cluster.
793:
794: We are currently expanding our survey to cover 15 fields centered on
795: HzRGs to determine more reliable limits on the prevalence of similar
796: over-densities of submm sources around massive galaxies at high
797: redshifts, $z=3$--5.
798:
799: \section*{Acknowledgements}
800:
801: We thank Chris Carilli and Wil van Breugel for providing their Very
802: Large Array and {\it HST} images of the 4C\,41.17 field, Dr F.\
803: Chaffee for assistance with obtaining the $V$-band image, Gary Punawai
804: and Greg Wirth for their expert assistance at the W.\,M.\ Keck
805: Observatory, as well as H.\ Halbedel, G.\ MacDougal, C.\ Snedden, H.\
806: Schweiker and G.\ Tiede for their expert assistance at the KPNO
807: 2.1m. We are grateful to the referee, Chris Carilli, for suggestions
808: that improved this paper markedly. The JCMT is operated by the Joint
809: Astronomy Centre on behalf of the United Kingdom Particle Physics and
810: Astronomy Research Council (PPARC), the Netherlands Organisation for
811: Scientific Research, and the National Research Council of Canada.
812: KPNO is a division of the National Optical Astronomy Observatory,
813: which is operated by the AURA, under cooperative agreement with the
814: NSF. The W.\,M.\ Keck Observatory is operated as a scientific
815: partnership among the California Institute of Technology, the
816: University of California and NASA.
817:
818: \begin{references}
819:
820: \reference{} Adelberger, K.\ L., \& Steidel, C.\ C.\ 2000, ApJ, submitted
821: (astro-ph/0001126)
822: \reference{} Arag\'on-Salamanca, A., Ellis, R.\ S., Couch, W.\ J., \& Carter,
823: D.\ 1993, MNRAS, 262, 764
824: \reference{} Arag\'on-Salamanca, A., Ellis, R.\ S., Schwartzenberg, J.-M., \&
825: Bergeron, J.\ A.\ 1994, ApJ, 421, 21
826: \reference{} Barger, A.\ J., Cowie, L.\ L., Sanders, D.\ B., Fulton, E.,
827: Taniguchi, Y., Sato, Y., Kawara, K., \& Okuda, H.\ 1998,
828: Nature, 394, 248
829: \reference{} Barger, A.\ J., Cowie, L.\ L., \& Sanders, D.\ B.\ 1999,
830: ApJ, 518, L5
831: \reference{} Baron, E., \& White, S.\ D.\ M.\ 1987, ApJ, 322, 585
832: \reference{} Baugh, C.\ M., Benson, A.\ J., Cole, S., Frenk, C.\ S., \&
833: Lacey, C.\ G.\ 1999, MNRAS, 305, L21
834: \reference{} Blain, A.\ W.\ 1999, MNRAS, 309, 955
835: \reference{} Blain, A.\ W., \& Longair, M.\ S.\ 1993, MNRAS, 264, 509
836: \reference{} Blain, A.\ W., Kneib, J.-P., Ivison, R.\ J., \& Smail, I.\
837: 1999, ApJ, 512, L87
838: \reference{} Bower, R.\ G., Lucey, J.\ R., \& Ellis, R.\ S.\ 1992, MNRAS,
839: 254, 601
840: \reference{} Carilli, C.\ L., Owen, F.\ N., \& Harris, D.\ E.\ 1994,
841: AJ, 107, 480
842: \reference{} Carilli, C.\ L., R\"ottgering, H., van Ojik, R., Miley, G.,
843: \& van Breugel, W.\ J.\ M.\ 1997, ApJS, 109, 1
844: \reference{} Carilli, C.\ L., Harris, D.\ E., Pentericci, L., R\"ottgering,
845: H., Miley, G., \& Bremer, M.\ N.\ 1998, ApJ, 496, L57
846: \reference{} Carilli, C.\ L., \& Yun, M.\ S.\ 1999, ApJ, 513, L13
847: \reference{} Carilli, C.\ L., \& Yun, M.\ S.\ 2000, ApJ, 530, 618
848: \reference{} Chambers, K.\ C., Miley, G.\ K.\ \& van Breugel, W.\ J.\ M.\ 1990,
849: ApJ, 363, 21
850: \reference{} Cole, S.\ M., Aragon-Salamanca, A., Frenk, C.\ S., Navarro, J.\
851: F., \& Zepf, S.\ E., 1994, MNRAS, 271, 781
852: \reference{} Dey, A., Spinrad, H., \& Dickinson, M.\ E.\ 1995, ApJ, 440, 515
853: \reference{} Dey, A.\ 1999, in The Most Distant Radio Galaxies, Proc.\ of the
854: Royal Netherland Academy of Arts \& Sciences, ed.\ H.\ J.\ A.\
855: Rottgering, P.\ A.\ Best, \& M.\ D.\ Lehnert, p.\ 12
856: \reference{} Dey, A., Graham, J.\ R., Ivison, R.\ J., Smail, I., Wright,
857: G.\ S., \& Liu, M.\ C.\ 1999, ApJ, 519, 610
858: \reference{} Dunlop, J.\ S., Hughes, D.\ H., Rawlings, S., Eales, S.\ A.,
859: \& Ward, M.\ J.\ 1994, Nature, 370, 347
860: %\reference{} Dunne, L., Clements, D.\ L., \& Eales, S.\ A.\ 2000, MNRAS,
861: % in press (astro-ph/0002436)
862: \reference{} Eales, S., Lilly, S., Gear, W., Dunne, L., Bond, J.\ R., Hammer,
863: F., Le F\`evre, O., \& Crampton, D.\ 1999, ApJ, 515, 518
864: \reference{} Efstathiou, G., \& Rees, M.\ 1988, MNRAS, 230, P5
865: \reference{} Ellis, R.\ S., Smail, I., Dressler, A., Couch, W.\ J., Oemler,
866: A.\ Jr., Butcher, H., \& Sharples, R.\ M.\ 1997, ApJ, 483, 582
867: \reference{} Elston, R., Rieke, G.\ H., \& Rieke, M.\ J.\ 1988, ApJ, 331, L77
868: \reference{} Fixsen, D.\ J., Dwek, E., Mather, J.\ C., Bennett, C.\ L., \&
869: Shafer, R.\ A.\ 1998, ApJ, 508 123
870: \reference{} Graham, J.\ R., Matthews, K., Soifer, B.\ T., Nelson, J.\ E.,
871: Harrison, W., Jernigan, J.\ G., Lin, S., Neugebauer, G., Smith,
872: G., \& Ziomkowski, C.\ 1994, ApJ, 420, L5
873: \reference{} Hogbom, J.\ A.\ 1974, A\&AS, 15, 417
874: \reference{} Holland, W.\ S., et al.\ 1999, MNRAS, 303, 659
875: \reference{} Hughes, D.\ H., et al.\ 1998, Nature, 394, 241
876: \reference{} Ivison, R.\ J., Smail, I., Le Borgne, J.-F., Blain, A.\ W., Kneib,
877: J.-P., B\'ezecourt, J., Kerr, T.\ H., \& Davies, J.\ K.\ 1998,
878: MNRAS, 298, 583
879: \reference{} Ivison, R.\ J., Smail, I., Barger, A.\ W., Kneib, J.-P., Blain,
880: A.\ W., Owen, F.\ N., Kerr, T.\ H., \& Cowie L.\ L.\ 2000, MNRAS,
881: in press (astro-ph/9911069)
882: \reference{} Jenkins, A., Colerg, J.\ M., Frenk, C.\ S., Evrard, A.\ E.,
883: White, S.\ D.\ M., Yoshida, N., \& Cole, S.\ M.\ 2000,
884: MNRAS, in press
885: \reference{} Jenness, T., \& Lightfoot, J.\ F.\ 1998, in ASP Conf.\ Ser.\ 145,
886: Astronomical Data Analysis Software and Systems VII, ed.\ R.\
887: Albrecht, et al.\ (San Francisco: ASP), 216
888: \reference{} Kaiser, N.\ 1984, ApJ, 284, L9
889: \reference{} Kauffmann, G., \& Charlot, S.\ 1998, MNRAS, 294, 705
890: \reference{} Kauffmann, G., Colberg, J.\ M., Diaferio, A., \& White,
891: S.\ D.\ M.\ 1999, MNRAS, 303, 188
892: \reference{} Lacy, M.\ \& Rawlings, S.\ 1996, MNRAS, 280, 888
893: \reference{} Lilly, S.\ J., \& Longair, M.\ S.\ 1984, MNRAS, 211, 833
894: \reference{} Lilly, S.\ J., Eales, S.\ A., Gear, W.\ K.\ P., Hammer, F.,
895: Le F\`evre, O., Crampton, D., Bond, J.\ R., \& Dunne, L.\ 1999,
896: ApJ, 518, 641
897: \reference{} Liu, M.\ C., Dey, A., Graham, J.\ R., Bundy, K.\ A., Steidel,
898: C.\ C., Adelberger, K.\ L., \& Dickinson, M.\ E.\ 2000, ApJ,
899: in press (astro-ph/0002443)
900: \reference{} Matthews, T.\ A., Morgan, W.\ W., \& Schmidt, M.\ 1964, ApJ,
901: 140, 35
902: \reference{} Matthews, K., \& Soifer, B.\ T.\ 1994, in: Infrared Astronomy
903: with Arrays: the Next Generation, ed.\ I.\ McLean, (Dordrecht:
904: Kluwer), 239
905: \reference{} McLure, R.\ J., Dunlop, J.\ S., Kukula, M.\ J., Baum, S.\ A.,
906: O'Dea, C.\ P., \& Hughes, D.\ H.\ 1999, MNRAS, 308, 377
907: \reference{} Oke, J.\ B., et al.\ 1995, PASP, 107, 375
908: \reference{} Pentericci, L., R\"ottgering, H.\ J.\ A., Miley, G.\ K., Spinrad,
909: H., McCarthy, P.\ J., van Breugel, W.\ J.\ M., \& Macchetto, F.\
910: 1998, ApJ, 504, 139
911: \reference{} Persson, S.\ E., Murphy, D.\ C., Krzeminski, W., Roth, M., \&
912: Rieke, M.\ J.\ 1998, AJ, 116, 2475
913: \reference{} Rosati, P., Stanford, S.\ A., Eisenhardt, P.\ R., Elston, R.,
914: Spinrad, H., Stern, D., \& Dey, A.\ 1999, AJ, 118, 76
915: \reference{} Sanders, D.\ B., \& Mirabel, I.\ F.\ 1996, ARAA, 34, 749
916: \reference{} Sanders, D.\ B., Soifer, B.\ T., Elias, J.\ H., Madore, B.\ F.,
917: Matthews, K., Neugebauer, G., \& Scoville, N.\ Z.\
918: 1988, ApJ, 325, 74
919: \reference{} Smail, I., Ivison, R.\ J., \& Blain, A.\ W.\ 1997, ApJ, 490, L5
920: \reference{} Smail, I., Ivison, R.\ J., Kneib, J.-P., Cowie, L.\ L., Blain,
921: A.\ W., Barger, A.\ J., Owen, F.\ N., \& Morrison, G.\ 1999,
922: MNRAS, 308, 1061
923: \reference{} Smail, I., Ivison, R.\ J., Owen, F.\ N., Blain, A.\ W., \& Kneib,
924: J.-P.\ 2000, ApJ, 528, 612
925: \reference{} Soucail, G., Kneib, J.-P., B\'ezecourt, J., Metcalfe, L.,
926: Altieri, B., \& Le Borgne, J.-F.\ 1999, A\&A, 343, L70
927: \reference{} Stanford, S.\ A., Elston, R., Eisenhardt, P.\ R., Spinrad, H.,
928: Stern, D., \& Dey, A.\ 1997, AJ, 114, 2232
929: \reference{} Stanford, S.\ A., Eisenhardt, P.\ R., \& Dickinson, M.\ 1998,
930: ApJ, 492, 461
931: \reference{} Steidel, C.\ C., Pettini, M.\ \& Hamilton, D.\ 1995, AJ, 110, 2519
932: \reference{} Steidel, C.\ C., Adelberger, K.\ L., Giavalisco, M., Dickinson,
933: M., \& Pettini, M.\ 1999, ApJ, 519, 1
934: \reference{} Thompson, D., Beckwith, S.\ V.\ W., Fockenbrock, R., Fried, J.,
935: Hippelein, H., Huang, J.-S., von Kuhlmann, B., Leinert, C.,
936: Meisenheimer, K., Phleps, S., Roser, H.-J., Thommes, E., \&
937: Wolf, C.\ 1999, ApJ, 523, 100
938: \reference{} van Breugel, W.\ J.\ M., Stanford, S.\ A., Spinrad, H.,
939: Stern, D., \& Graham, J.\ R.\ 1998, ApJ, 502, 614
940: \reference{} White, S.\ D.\ M., \& Frenk, C.\ S.\ 1991, ApJ, 379, 52
941: \reference{} Yamada, T., Tanaka, I., Arag\'on-Salamanca, A., Kodama, T., Ohta,
942: K., \& Arimoto, N.\ 1997, ApJ, 487, L125
943: \reference{} Zirm, A., Dey, A., Dickinson, M., McCarthy, P.\ J., Eisenhardt,
944: P., Djorgovski, S.\ G., Spinrad, H., Stanford, S.\ A., \& van
945: Breugel, W.\ 2000, in ASP Conf.\ Ser.\ 193, The Hy-Redshift
946: Universe: Galaxy Formation and Evolution at High Redshift, ed.\
947: A.\ J.\ Bunker, \& W.\ J.\ M.\ van Breugel (San Francisco: ASP),
948: 114
949: \end{references}
950:
951: \end{document}
952: