9512.net
甜梦文库
当前位置:首页 >> >>

The evolution and star formation of dwarf galaxies in the Fornax Cluster


Mon. Not. R. Astron. Soc. 000, 000–000 (0000)

Printed 1 February 2008

A (MN L TEX style ?le v1.4)

The evolution and star formation of dwarf galaxies in the Fornax Cluster
M.J. Drinkwater1,5, M.D. Gregg2,3, B.A. Holman1, M.J.I. Brown1,4
of Physics, University of Melbourne, Victoria 3010, Australia of Physics, University of California at Davis, Davis, CA 95616, USA 3 Institute for Geophysics and Planetary Physics, Lawrence Livermore National Laboratory, L-413, Livermore, CA 94550, USA 4 National Optical Astronomy Observatory, 950 North Cherry Avenue, P.O. Box 26732, Tucson, Arizona 85726, USA 5 mjdrin@unimelb.edu.au
2 Department 1 School

arXiv:astro-ph/0106376v1 21 Jun 2001

Version 2; revised.

ABSTRACT

We present the results of a spectroscopic survey of 675 bright (16.5 < bJ < 18) galaxies in a 6 degree ?eld centred on the Fornax cluster with the FLAIR-II spectrograph on the UK Schmidt Telescope. Three galaxy samples were observed: compact galaxies to search for new blue compact dwarfs, candidate M 32-like compact dwarf ellipticals, and a subset of the brightest known cluster members in order to study the cluster dynamics. We measured redshifts for 516 galaxies of which 108 were members of the Fornax Cluster. De?ning dwarf galaxies to be those with bJ ≥ 15 (MB ≥ ?16.5), there are a total of 62 dwarf cluster galaxies in our sample. Nine of these are new cluster members previously misidenti?ed as background galaxies. The cluster dynamics show that the dwarf galaxies are still falling into the cluster whereas the giants are virialised. We classi?ed the observed galaxies as late-type if we detected Hα emission at an equivalent width greater than 1 ?. The spectra were obtained through ?xed apertures, A so they re?ect activity in the galaxy cores, but this does not signi?cantly bias the classi?cations of the compact dwarfs in our sample. The new classi?cations reveal a higher rate of star formation among the dwarf galaxies than suggested by morphological classi?cation: 35 per cent have signi?cant Hα emission indicative of star formation but only 19 per cent were morphologically classi?ed as late-types. The star-forming dwarf galaxies span the full range of physical sizes and we ?nd no evidence in our data for a distinct class of star-forming blue compact dwarf (BCD) galaxy. The distribution of scale sizes is consistent with evolutionary processes which transform late-type dwarfs to early-type dwarfs. The fraction of dwarfs with active star formation drops rapidly towards the cluster centre: this is the usual densitymorphology relation con?rmed here for dwarf galaxies. The star-forming dwarfs are concentrated in the outer regions of the cluster, the most extreme in an infalling subcluster. We estimate gas depletion time scales for 5 dwarfs with detected H I emission: these are long (of order 1010 yr), indicating that an active gas removal process must be involved if they are transformed into gas-poor dwarfs as they fall further into the cluster. Finally, in agreement with our previous results, we ?nd no compact dwarf elliptical (M 32-like) galaxies in the Fornax Cluster.
1 INTRODUCTION ter members from the large contamination of background objects. When the Fornax Cluster Catalog (Ferguson, 1989b) was compiled, only 85 of the 340 likely cluster members had measured redshifts; of the whole FCC catalogue of 2678 galaxies, only 112 (4 per cent) had redshifts. Since then, a number of small-scale spectroscopic surveys have been made of the cluster (Held & Mould, 1994; Bureau, Mould, & Staveley-Smith, 1996; Hilker, Infante, Vieira, Kissler-Patig, & Richtler, 1999), but even as of late 1999, NED? still
? The NASA/IPAC Extragalactic Database (NED) is operated

Our understanding of the role of dwarf galaxies in clusters has long been dominated by photographic studies of the nearest clusters. The major surveys of the Virgo cluster in the North (Binggeli, Sandage, & Tarenghi, 1984; Binggeli, Sandage, & Tammann, 1985; Binggeli, Popescu, & Tammann, 1993) and the Fornax cluster in the South (Ferguson & Sandage, 1988; Ferguson, 1989b) have established the presence of large populations of dwarf galaxies in these clusters. Lacking redshift information, these works have had to rely on image morphology for statistical separation of clusc 0000 RAS

2

Drinkwater et al.
Binggeli, 1994). Our photographic imaging data is not able to resolve di?erences at this level however, hence the overlap in the two samples. In Paper 1 we reported that only one of the M 32 candidates we observed was a cluster member, but that it had an emission line spectrum inconsistent with a cdE identi?cation. We concluded that tidal stripping had not produced large numbers of cdEs in Fornax (Holman et al., 1995; Drinkwater et al., 1997). In this paper we expand our discussion of the M 32 candidate galaxies in the context of complete samples. Our measurements of candidate compact cluster galaxies detected a majority of background galaxies, interesting in their own right for large scale structure information. In addition to the new compact cluster members discovered (Paper I), we also observed some 100 known cluster galaxies in order to investigate star formation rates and the dynamics of the cluster, particularly subclustering (Drinkwater et al., 2001, Paper III). In this paper we adopt a cluster distance of 20 Mpc (Mould et al., 2000, but see discussion in Paper III) corresponding to a distance modulus of 31.5 mag. In Section 2 we describe how the galaxies were selected and in Section 3 we describe the observations and redshift measurements. We discuss the structure of the cluster in Section 4 and in Section 5 we discuss the star formation process in the dwarf cluster galaxies. The compact dwarf elliptical population is discussed in Section 6 and we summarise all our results in Section 7. We list the background galaxies we observed in the Appendix.

lists only about 120 objects within a 5? radius of NGC1399 with velocities that would place them in the cluster. In this paper we describe a large new spectroscopic investigation of the Fornax cluster designed to add radial velocity membership, spectral classi?cation, and dynamical information to the existing knowledge of the cluster. Our study is based on data from the FLAIR-II spectrograph on the UK Schmidt Telescope. We measured the redshifts of over 500 galaxies brighter than bJ = 18 in the direction of Fornax, of which 108 are members of the cluster. This study is complementary to the Fornax Cluster Spectroscopic Survey (FCSS) Drinkwater et al. (2000b) which is using the Two degree Field spectrograph on the Anglo-Australian Telescope to make a deeper survey of a smaller region towards the centre of the cluster. Unlike the current study, the FCSS will measure all objects in its target magnitude range (16.5 < bJ < 19.7), irrespective of morphology (i.e. both ‘stars’ and ‘galaxies’) in order to sample the largest possible range of surface brightness. A number of issues regarding the Fornax dwarf population can be addressed de?nitively only through spectroscopic information to determine cluster membership. One interesting question is the possible existence of a population of very compact dwarfs that may have been misclassi?ed as background galaxies in the FCC. Drinkwater et al. (1996) searched unsuccessfully for additional blue compact dwarf (BCD) galaxies in the Virgo Cluster; any compact Virgo cluster galaxies must be fainter than their limit of bJ = 17.6. The Fornax cluster has a much lower fraction of latetype dwarfs than Virgo (Ferguson & Sandage, 1988), and its core galaxy density is about twice that of Virgo (Ferguson, 1989a), o?ering the chance for an interesting comparison of two dwarf galaxy populations in rather di?erent environments. The initial results of our new spectroscopic survey have already revealed a number of new compact dwarf galaxy members, both early and late-type (Drinkwater & Gregg, 1998, Paper I hereafter). In this paper we present the rest of our observations with the analysis of a well-de?ned sample of compact galaxies. The FCC lists a total of 131 objects as candidate “M 32like” compact dwarf ellipticals (cdE). Faber (1973) suggested that high surface-brightness low-luminosity elliptical galaxies (like M 32 ) are formed by tidal stripping of brighter ellipticals in rich environments, but more recent models ?nd that the amount of mass lost by this process in a Hubble time is too small to produce compact dwarf ellipticals from larger galaxies (Nieto & Prugniel, 1987). To test the tidal stripping hypothesis, we observed 97 of the possible cdEs in Fornax to determine if they were cluster members. Whilst there was considerable overlap between these and our compact sample described above, the M 32-like sample also included a number of galaxies with larger scale sizes not included in the compact sample. It should be possible to distinguish galaxies like M 32 from normal dwarf galaxies on the basis of their pro?les. We would expect M 32-like bulges to have de Vaucouleurs pro?les (Michard & Nieto, 1991), whereas most dwarf galaxies have exponential pro?les (Ferguson &

2

SAMPLE SELECTION

The selection criteria for the objects we observed were set by our three overlapping science goals, subject to the capacity of the FLAIR-II spectrograph on the UK Schmidt Telescope of the Anglo-Australian Observatory (Parker & Watson, 1995b). This is a multi-object ?bre spectrograph with a square ?eld approximately 6 degrees across. The limiting magnitude of the system is about bJ =17.5 and each observation can take 70 or 90 spectra depending on the plate holder used. It is ideally suited to the measurement of the brighter galaxies in the Fornax cluster which has a core radius of 0.7? (FCC). Most of out observations were made centred on the standard UK Schmidt sky survey ?eld termed “F358” s (03h 37m 55. 9, ?34? 50′ 14′′ J2000). As seen in Fig. 1, the cluster is o?set from ?eld F358, so some observations were centred on a second ?eld displaced about half a degree souths west (03h 36m 55. 1 ?35? 30′ 11′′ J2000). The overlapping samples we observed are described in the rest of this section. 2.1 Compact galaxies

by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.

The primary goal of the project was to search for new cluster compact galaxies, both red and blue, so we concentrated e?orts on galaxies with a compact morphology that were not classi?ed as cluster members in the FCC. These were selected on the basis of their image parameters measured from the two Schmidt plates described above. Although the targets were initially selected on the basis of compact image morphologies measured from the ?rst Schmidt plate (as described below), many additional targets were later added as
c 0000 RAS, MNRAS 000, 000–000

Dwarf galaxies in Fornax

3

Figure 1. Distribution on the sky of all FCC galaxies (small points), and galaxies observed with FLAIR-II (squares). The region de?ned by the APM sample is also shown.

the second plate was used and additional “M 32-like” candidates were included. We can therefore de?ne three di?erent samples of compact galaxies: a statistical sample with de?ned limits and completeness from a single plate; the list of all M 32 candidates observed; and a heterogeneous list of both of these plus any other compact galaxies observed. The ?rst sample was de?ned by image parameters measured by the APM (Automated Plate Measuring facility, Cambridge) from the UK Schmidt bJ survey plate of ?eld F358. See Irwin et al. (1994) for more details of the APM image catalogue. The APM image catalogue lists image positions, magnitudes and morphological classi?cations (as ‘star’, ‘galaxy’, ‘noise’, or ‘merged’) measured from both the blue (bJ ) and red survey plates. The ‘merged’ image classi?cation indicates two overlapping images: at the magnitudes of interest for this project, the merged objects nearly always consisted of a star overlapping a much fainter galaxy. The APM magnitudes are calibrated for unresolved (stellar) objects, so for most purposes we use magnitudes estimated by ?tting exponential disk pro?les to the APM image parameters calculated by Davies et al. (1988) and Irwin et al. (1990) (see Morshidi-Esslinger et al., 1999). We refer to these as Davies magnitudes. This process also measures the exponential scale lengths of the images. For galaxies brighter than bJ =13 we have used magnitudes from Ferguson (1989b). There is no evidence for any signi?cant o?set between the Davies and FCC magnitudes (Drinkwater et al., 2000b). In one speci?c case (FCC 39) there seems to be a problem with both the FCC and Davies magnitudes being too faint, so we have adopted the value of bJ =13.6 given by Maddox et al. (1990). We de?ned the compact galaxy sample by scale lengths and magnitudes as shown in Fig. 2. To optimise our measurement for the compact objects we used the APM (stellar) magnitudes for this selection, linearly scaled to agree with
c 0000 RAS, MNRAS 000, 000–000

the Davies magnitudes over this range. The compact galaxy sample is de?ned in the magnitude range of 16.5 < bJ < 17.5 by the parallelogram shown in the Figure. Basically it includes all galaxies and possible galaxies (“merged” objects) in that magnitude range with sizes smaller than the galaxies classi?ed as cluster members in the FCC. This compact subsample consists of 815 objects of which 25 are FCC members, 400 are FCC background galaxies and the remainder (390) are not listed in the FCC, presumably because they are smaller than the FCC limiting diameter of 17 arc seconds. We successfully observed 306 or 38 per cent of the objects in this sample. Most of the objects not observed were classi?ed as “merged” and were not measured because they were clearly dominated by a star with visible di?raction spikes. The magnitude distribution of the compact sample is compared with the other samples observed in Fig. 3. None of the new cluster members were found amongst the smallest objects in the compact sample: these were mostly merged objects dominated by a star. We therefore de?ned a second, more conservative compact sample by excluding the smallest objects below the dashed line in Fig. 2. This second sample has a higher completeness, comprising 352 compact objects (25 FCC members, 272 FCC background galaxies and 55 not listed in the FCC) of which we measured 211 (60 per cent) to ?nd 7 new members. Our second compact galaxy sample consisted of the 131 galaxies listed in the FCC as possible cdE “M 32-like” candidates. We observed 97 of these, obtaining 76 reliable redshifts. Fig. 3 shows that these objects were observed to a magnitude limit of BT = 17.8. Our total compact galaxy sample consisted of the above two samples plus additional compact objects with similar properties. From this sample we measured a total of 437 redshifts.

2.2

Cluster members

We also attempted to observe as many as possible of the cluster members from the FCC in order to have a good sample for studies of the cluster dynamics (Paper III). These were measured with a higher spectral dispersion to give more accurate velocities. The Fornax cluster extends well beyond the limits of a single Schmidt plate; the FCC includes galaxies in a slightly larger region than de?ned by our APM sample as is shown in Fig. 1. This Figure shows that we observed some galaxies beyond the APM sample. Fig. 3 shows that the member galaxies observed were as faint as BT = 18, but being of generally low surface brightness the success rate fainter than BT = 16.6 was lower than for the other samples.

3 3.1

OBSERVATIONS FLAIR-II Observations

We observed Fornax for a total of 6 observing runs. The ?rst 5 used the low-dispersion 250B grating (coverage 36707230?, resolution 13?), chosen to identify as many galaxies A A as possible at all redshifts. These observations were severely a?ected by bad weather, but by the end of the ?fth run we

4

Drinkwater et al.

Figure 2. Distribution of bJ magnitudes and scale length of all resolved objects in the APM catalogue (small points), galaxies observed with FLAIR-II (open triangles; those with redshifts are ?lled triangles and cluster members are indicated by large ?lled squares) and galaxies listed as members in the FCC (large circles). New members of the cluster discovered with FLAIR are indicated by asterisks. The parallelogram de?nes the compact galaxy sample discussed in the text. Objects in the parallelogram above the dashed line form the more conservative sample. Note that for the de?nitions of these samples, the magnitudes were taken from the APM catalogue, rescaled to approximate the calibrated Davies magnitudes used elsewhere.

Figure 3. Distributions of BT magnitudes (from the FCC) of all galaxies from each of the three samples observed: compact galaxies, M 32 candidates from the FCC and de?nite cluster members from the FCC. In each case the unshaded histogram shows the parent sample, the shaded histogram shows the galaxies observed, and the solid histogram those galaxies with measured redshifts. Note the di?erent vertical scale in the top panel.

3.2

Data Reduction and Velocity Measurement

Table 1. Details of our FLAIR-II observations. Run 1 1 2 2 2 3 4 4 5 5 6 6 6 Date 1992 1992 1993 1993 1993 1994 1995 1995 1996 1996 1997 1997 1997 October 25 October 26 November 14 November 15 November 16 November 06 November 22 November 26 December 08 December 09 October 31 November 01 November 08 vhelio ( km s?1 ) 1.4 1.1 ?5.3 ?5.6 ?5.9 ?2.7 ?7.7 ?8.9 ?12.5 ?12.7 ?0.7 ?1.0 ?3.3 grating 250B 250B 250B 250B 250B 250B 250B 250B 250B 250B 600V 600V 600V

The data were reduced using the IRAF (Tody, 1993) DOFIBERS package. We used the IRAF add-on package RVSAO (Kurtz & Mink, 1998) to measure the galaxy redshifts by cross-correlation with template spectra. All velocities were then corrected to heliocentric values using the o?sets listed in Table 1. We checked our velocity errors using measurements of 52 galaxies that were observed twice. The mean and rms values of the velocity di?erences are shown in Table 2 for the whole sample and then for two sets of subsamples split according to grating and also by magnitude. For comparison, we also give the RVSAO estimated mean velocity errors for each sample: these are based on a formal statistical calculation. The results shown in the table are very good considering the relatively low dispersion and signal-to-noise of our spectra. The velocity resolutions of the 250B and 600V gratings are 650 km s?1 and 265 km s?1 respectively. In particular, the error estimates from RVSAO are shown to be realistic, underestimating the empirically-determined errors by only around 10 per cent. The improvement with the 600V grating is less than expected, but this is based on a very small number of repeated observations of poor spectra at this dispersion. Note that the rms values discussed here are for the di?erence of two measurements and are a fac√ tor of 2 larger than the rms for a single measurement. We have therefore adopted the velocity uncertainties measured by RVSAO. As a further test of our velocity measurements we compared them to previous measurements in the literature. We
c 0000 RAS, MNRAS 000, 000–000

Notes: vhelio is the correction applied to convert our observed velocities to heliocentric values.

had measured most of the compact galaxies within our magnitude limits. For the sixth run we observed FCC-classi?ed members (most of which we had not yet observed) with the medium-dispersion 600V grating (coverage 5150–6680?, resA A olution 5.3?) to measure more accurate velocities and study the dynamics of the cluster. A log of our observing runs is shown in Table 1.

Dwarf galaxies in Fornax
Table 2. Velocity di?erences for repeated observations sample all 600V 250B BT < 16 BT > 16 N 52 9 43 22 30 ?v ( km s?1 ) 10 32 4 26 2 σ?v σv ( km s?1 ) 67 45 69 56 74 59 44 60 53 62

5

Notes: ?v and σ?v are the mean and rms velocity di?erences from repeated observations. For each sample σv is the mean statistical uncertainty based on the RVSAO errors. These rms values are for √ the di?erence of two measurements and are 2 larger than the rms for a single measurement. Table 3. Objects with spectra dominated by light from a foreground Galactic star RA (J2000) Dec 03:30:10.12 03:31:26.86 03:33:10.30 03:37:22.01 03:38:15.18 03:39:19.56 03:44:51.01 03:47:39.18 03:49:18.67 03:50:48.11 ?33:41:35.8 ?37:10:48.5 ?37:55:11.9 ?34:51:54.1 ?34:50:55.2 ?35:43:28.9 ?38:18:14.0 ?36:22:19.3 ?32:18:47.3 ?34:03:58.4 bJ (mag) 17.40 16.60 17.30 16.90 15.50 16.80 16.70 16.70 17.10 16.10 cz ?cz ( km s?1 ) 193 41 ?24 86 39 ?5 ?42 74 99 1 64 23 27 43 12 60 34 54 56 38 FCC 81 B723 B849 – B1239 223 B1781 – B2112 –

Figure 4. Distribution of the 516 galaxy velocities measured for this work (unshaded histogram) compared to the 84 measurements for the same galaxies in the FCC (shaded histogram).

Table 5. Comparison of new spectroscopic classi?cation with FCC membership classes FLAIR-II classi?cation: FCC member FCC background not in FCC total star 2 5 3 10 member 99 9 – 108 background 3 398 7 408 total 104 412 10 526

found redshift measurements of 102 galaxies in common with our sample in NED (as of 1999 November 23). The mean velocity di?erence (FLAIR?NED) was ?v = ?22 ± 11 km s?1 and σ?v = 108 km s?1 , demonstrating that there were no large systematic errors in our measurements. In the course of our 6 observing runs, we observed a total of 675 objects and obtained reliable redshifts for 526 of them. We found that 10 of these objects had absorption line spectra and velocities consistent with Galactic stars. These were merged objects consisting of a star superimposed on a background galaxy. These were included in the compact sample in case the “star” was the core of the galaxy, but in each case it was shown to be in the foreground and dominated the spectrum so the background galaxy could not be measured. These objects are listed for reference in Table 3 and are not discussed further in this paper. Our ?nal sample consists of 108 con?rmed cluster galaxies with successfully measured redshifts and is listed in Table 4. The 408 con?rmed background galaxies are listed in Table A1. Note that our position for the galaxy FCC 13 is taken from Loveday (1996) as the FCC position is in error.

3.3

Membership Status

The full impact of our new measurements is demonstrated in Fig. 4 which gives the distribution of our 516 measured redshifts compared to the 84 redshifts available in the FCC for the same galaxies. The majority of our spectra are of background galaxies as expected since our aim was to search
c 0000 RAS, MNRAS 000, 000–000

for new cluster members among galaxies classi?ed as “background” in the FCC. The membership classi?cations of the objects we observed are compared to the FCC classi?cations in Table 5. As in Paper I, we detect a clear void behind the Fornax cluster with no galaxies having redshifts between 2320 and 4180 km s?1 , so we de?ne any galaxy with a redshift less than 3000 km s?1 as a member of the cluster. We ?nd that the FCC membership classi?cations were very reliable: out of 516 galaxies measured, only 9 were members misclassi?ed as background galaxies in the FCC and only 3 were background galaxies misclassi?ed as members. These 12 objects with new membership classi?cations are listed in Table 6. Note that one additional new cluster member (FCC B1241) was reported in Paper I. Although this galaxy was in our FLAIR-II sample, it was not observed successfully with FLAIR-II (the identi?cation was from 2dF observations), so shall not be discussed in this current paper. The properties of the new cluster members are discussed in detail below. Six of these were ?rst presented in Paper I; our present analysis has produced an additional three new cluster members. We also found three background galaxies that were classi?ed as members in the FCC; these include two of the FCC candidate late-type dwarf galaxies (FCC 24 and FCC 123). These both had emission line spectra con-

6

Drinkwater et al.

Table 4. Catalogue of all Fornax Cluster members with redshifts measured with FLAIR-II
FCC 13 19 21 22 26 28 29 32 33 35 37 36 39 43 46 47 55 B 470 62 67 76 83 88 90 95 B 729 100 102 106 113 119 121 B 904 B 905 136 135 139 143 147 148 B 1005 150 153 152 161 164 167 170 176 174 179 177 B 1108 181 182 184 188 190 193 201 202 203 206 204 207 211 213 219 B 1379 235 243 245 249 252 253 255 261 263 266 267 B 1554 276 277 278 282 285 288 290 296 298 299 301 302 303 305 306 308 310 312 316 315 319 322 324 B 2144 335 336 338 RA (J2000) Dec 03:21:03.7 03:22:22.9 03:22:42.2 03:22:44.6 03:23:37.3 03:23:54.3 03:23:56.3 03:24:52.5 03:24:58.5 03:25:04.1 03:25:09.1 03:25:12.2 03:25:19.8 03:26:02.5 03:26:25.1 03:26:32.1 03:27:18.1 03:27:33.8 03:27:58.3 03:28:48.8 03:29:43.4 03:30:35.2 03:31:08.1 03:31:08.2 03:31:24.8 03:31:32.5 03:31:47.7 03:32:10.8 03:32:47.8 03:33:06.8 03:33:33.9 03:33:36.3 03:33:56.2 03:33:57.2 03:34:29.5 03:34:30.9 03:34:57.4 03:34:59.2 03:35:16.9 03:35:16.9 03:35:20.4 03:35:24.1 03:35:31.1 03:35:33.2 03:36:04.1 03:36:12.9 03:36:27.6 03:36:31.7 03:36:45.1 03:36:45.4 03:36:46.4 03:36:47.5 03:36:49.7 03:36:53.2 03:36:54.4 03:36:57.0 03:37:04.6 03:37:09.0 03:37:11.8 03:37:53.7 03:38:06.6 03:38:09.3 03:38:13.5 03:38:13.8 03:38:19.3 03:38:21.5 03:38:29.3 03:38:52.2 03:39:55.0 03:40:09.3 03:40:27.0 03:40:33.8 03:40:42.1 03:40:50.4 03:40:55.2 03:41:03.5 03:41:21.5 03:41:32.3 03:41:41.3 03:41:45.6 03:41:59.5 03:42:19.3 03:42:22.8 03:42:27.3 03:42:45.6 03:43:01.9 03:43:22.8 03:43:37.1 03:44:32.9 03:44:44.4 03:44:58.6 03:45:03.6 03:45:12.3 03:45:14.1 03:45:33.8 03:45:45.4 03:45:54.9 03:46:13.8 03:46:19.0 03:47:01.4 03:47:04.7 03:47:16.1 03:47:33.2 03:47:52.8 03:49:53.3 03:50:36.8 03:50:52.6 03:52:01.1 ?37:05:58 ?37:23:50 ?37:12:36 ?37:06:16 ?35:46:43 ?37:30:33 ?36:27:57 ?35:26:08 ?37:00:34 ?36:55:40 ?36:21:55 ?32:54:10 ?36:23:05 ?32:53:42 ?37:07:39 ?35:42:49 ?34:31:34 ?35:43:04 ?37:08:57 ?35:10:49 ?33:33:26 ?34:51:19 ?33:37:47 ?36:17:24 ?35:19:51 ?38:03:43 ?35:03:05 ?36:13:13 ?34:14:19 ?34:48:32 ?33:34:23 ?36:08:28 ?34:33:43 ?34:36:43 ?35:32:46 ?34:17:51 ?32:38:22 ?35:10:15 ?35:13:39 ?35:16:01 ?32:36:08 ?36:21:49 ?34:26:49 ?32:27:50 ?35:26:34 ?36:09:59 ?34:58:36 ?35:17:48 ?36:15:22 ?33:00:49 ?36:00:02 ?34:44:22 ?33:27:39 ?34:56:17 ?35:22:27 ?35:30:29 ?35:35:24 ?35:11:42 ?35:44:44 ?37:16:47 ?35:26:23 ?34:31:06 ?37:17:23 ?33:07:36 ?35:07:43 ?35:15:35 ?35:27:07 ?35:35:42 ?33:03:09 ?35:37:28 ?36:29:56 ?35:01:21 ?37:30:38 ?35:44:53 ?37:50:16 ?33:46:43 ?33:46:09 ?34:53:22 ?35:10:12 ?33:47:29 ?35:20:53 ?35:23:41 ?35:09:15 ?33:52:14 ?33:55:12 ?36:16:16 ?33:56:20 ?35:51:17 ?35:11:45 ?35:41:01 ?36:53:42 ?35:58:22 ?35:34:14 ?36:56:12 ?37:04:58 ?36:20:45 ?36:21:30 ?36:41:48 ?34:56:36 ?36:26:15 ?33:42:34 ?32:18:08 ?38:34:54 ?36:28:18 ?32:15:33 ?35:54:34 ?35:10:19 ?33:28:07 cz ( km s?1 ) σcz 1792 1497 1659 1980 1718 1381 1244 1318 1994 1797 1785 2325 991 1323 2220 1473 1279 723 1813 1345 1808 1477 1924 1813 1275 1676 1660 1723 2064 1365 1384 1446 2254 1242 1217 1381 1752 1364 1299 794 1256 2031 1589 1389 1342 1430 1953 1763 1399 1645 972 1622 1734 1113 1657 1257 1004 1740 921 1922 808 1173 1402 1369 1419 2276 1433 1897 745 1974 1404 2124 1571 1415 1677 1283 1492 1691 1551 834 1642 1382 1613 2125 1267 891 1103 1392 816 1719 2151 1005 806 1980 1228 898 1484 1373 1898 1546 1071 1445 978 1493 1184 1367 1956 1562 25 47 17 18 46 28 19 26 19 15 59 90 13 17 32 23 17 79 13 44 14 11 15 15 26 31 31 61 35 23 40 12 56 23 24 49 24 16 17 17 36 26 10 12 16 33 13 14 30 78 17 21 137 53 19 12 80 17 12 126 22 30 20 28 43 23 26 21 21 20 52 68 18 35 57 25 42 15 39 10 52 12 25 30 36 6 37 20 122 36 40 19 23 31 25 14 25 13 22 105 15 61 15 44 25 24 67 16 ? EW(Hα)(A) 11.8 ?0.7 1.0 15.4 3.5 9.3 18.6 9.6 4.0 175 3.8 1.3 23.3 ?1.7 2.1 ?0.6 ?0.8 0.3 61.6 2.8 6.1 0.4 0.1 11.2 ?0.5 6.8 ?0.4 13.0 ?1.4 10.1 ?0.1 43.5 ?0.1 18.2 ?1.1 ?0.5 13.6 ?1.1 -0.8 ?1.3 ?0.2 ?1.1 ?0.7 7.4 ?1.4 ?1.9 0.1 ?0.7 ?1.1 ?1.2 4.5 ?0.3 ?0.1 ?0.2 ?1.6 0.1 ?1.6 ?1.2 ?1.1 0.7 ?0.7 ?0.6 14.4 ?0.5 2.2 ?0.1 0.1 -0.2 17.3 23.1 ?2.8 ?1.1 ?0.6 ?1.9 ?2.0 ?1.0 4.5 32.9 ?1.3 13.4 ?0.8 ?0.8 ?0.4 ?1.2 8.2 22.8 ?1.7 ?0.4 ?0.9 ?1.1 29.7 ?0.9 20.2 ?2.9 ?1.2 34.7 16.3 ?1.0 22.5 ?0.4 ?0.1 ?0.1 54.4 ?1.1 65.1 ?0.1 0.4 32.9 T-type 5 ?5 ?1 1 ?1 9 1 ?5 7 9 5 ?5 7 ?5 ?5 ?4 ?1 0 4 5 10 ?4 3 ?4 ?1 ?5 ?5 10 ?1 5 ?1 4 ?5 10 ?5 ?5 9 ?4 ?4 ?1 ?1 ?5 ?1 0 ?4 ?5 0 ?1 1 ?5 1 ?1 ?5 ?5 ?1 ?1 ?5 ?1 ?1 ?5 ?4 ?5 ?5 ?5 ?5 ?4 ?4 ?4 ?5 10 ?5 ?5 ?4 ?5 ?5 ?1 ?5 6 ?5 9 ?4 ?4 ?4 ?5 10 7 ?5 5 ?5 ?5 7 ?4 8 ?5 ?5 9 7 ?1 5 ?5 2 ?5 7 ?5 ?4 ?4 ?5 ?1 bJ (mag) 12.0 15.2 09.4 11.9 16.4 13.5 11.8 16.1 14.2 16.1 13.8 16.2 13.6 15.1 16.5 13.3 15.5 17.5 12.6 14.2 16.3 12.3 11.8 15.6 15.2 16.5 15.9 16.8 16.3 15.5 15.7 10.2 17.4 17.7 16.3 16.5 15.3 14.5 12.0 13.9 13.7 16.2 14.2 14.9 12.6 17.0 11.3 12.5 13.7 16.8 12.4 13.2 17.8 17.7 15.0 12.3 16.1 13.5 13.2 17.3 16.1 16.8 16.9 15.8 16.1 16.8 10.6 10.0 17.3 13.4 16.7 16.6 14.1 16.1 17.0 13.9 16.7 14.9 16.0 16.4 17.7 12.6 13.9 17.4 15.0 14.2 16.8 12.4 16.7 17.1 17.9 14.6 17.1 15.5 16.8 16.1 13.8 14.0 13.5 16.9 12.6 17.1 12.5 16.7 17.1 15.2 18.0 14.3 SB (mag arcsecond ?2 ) — — — — 22.3 18.4 — 21.8 21.4 21.3 21.8 21.2 21.6 21.6 21.3 21.3 20.6 21.1 15.9 21.1 22.3 — — 21.8 20.4 — 22.4 22.2 22.3 22.2 21.2 — 21.7 22.5 21.6 22.3 20.2 20.9 — 19.4 19.9 21.4 19.2 20.7 — 22.1 — 17.5 21.4 21.8 — 20.3 22.5 22.9 21.3 — 22.1 20.1 — 23.2 21.5 22.1 22.9 21.4 21.5 21.6 — — 21.7 22.1 22.5 22.7 19.7 22.0 22.4 19.2 22.3 20.2 20.7 22.3 21.8 — 20.0 22.1 20.1 22.5 22.0 — 22.3 21.7 22.5 20.2 23.2 21.9 22.3 20.4 22.3 20.5 — 23.0 — 23.1 — 22.4 21.6 20.5 22.7 — α ( arc second) — — — — 6.1 3.9 — 5.4 6.3 4.2 7.1 4.0 6.1 7.6 3.6 8.7 4.1 2.1 3.7 5.5 6.4 — — 6.9 4.3 — 7.9 5.0 6.3 8.7 5.0 — 2.9 3.7 4.8 5.8 3.8 7.6 — 5.0 6.8 4.5 4.1 5.9 — 4.2 — 4.0 9.6 4.0 — 5.4 3.6 4.4 7.2 — 6.3 6.4 — 6.0 4.9 4.5 6.1 5.3 4.9 3.6 — — 3.0 7.3 5.7 6.6 5.2 6.0 4.7 4.5 5.3 4.7 3.4 5.9 2.7 — 5.0 3.5 4.3 9.3 4.5 — 5.0 3.4 3.3 5.2 6.5 7.4 5.1 2.8 7.5 7.9 — 6.6 — 6.6 — 5.4 3.2 4.6 3.4 — P1(1) 0.01 0.01 0.01 0.01 0.18 0.02 0.01 0.17 0.02 0.02 0.04 1.00 0.02 1.00 0.06 0.18 0.95 0.12 0.46 0.85 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00

Note: (1) P1 is the probability that the galaxy lies in the Fornax-main cluster and not the Fornax-SW subcluster.

c 0000 RAS, MNRAS 000, 000–000

Dwarf galaxies in Fornax
Table 6. Galaxies listed in the FCC with changed membership classi?cations. RA (J2000) Dec 03:23:31.7 03:27:33.8 03:31:32.5 03:33:43.4 03:33:56.2 03:33:57.2 03:35:20.4 03:36:49.7 03:39:55.0 03:41:59.5 03:46:18.2 03:49:53.3 ?34:36:34 ?35:43:04 ?38:03:43 ?35:51:33 ?34:33:43 ?34:36:43 ?32:36:08 ?33:27:39 ?33:03:09 ?35:20:53 ?33:45:48 ?32:15:33 bJ 15.2 17.5 16.5 17.9 17.4 17.7 13.7 17.8 17.3 17.7 17.9 17.1 cz ?cz ( km s?1 ) 13222 723 1676 15483 2254 1242 1256 1734 745 1642 47658 1184 14 79 31 60 56 23 36 137 21 52 60 25 FCC 24 B470 B729 123 B904 B905 B1005 B1108 B1379 B1554 311 B2144 type BCD S/Im (d)S0 ImV (d)E ? S0 (d)S0 dE E dS0 (d)E

7

Table 8. Correspondence between new galaxy “t-types” adopted and FCC classi?cations t-type -5 -4 -1 0 1 2 3 4 FCC type dS0, dE E S0, SB0 S0/a, S, SB Sa, SBa, RSa Sab, SBab Sb, SBb Sbc, SBbc t-type 5 6 7 8 9 10 15 20 FCC type Sc, SBc Scd, SBcd Sd, SBd Sdm Sm, SBm Im, BCD, ? other unclassi?ed

clearly no large population of new members still to be found in this magnitude range.

Note: the ?nal column (“type”) gives the FCC morphological classi?cation. Table 7. Background galaxies not listed in the FCC Name FCSS J032700.2?354854 FCSS J033154.8?343045 FCSS J033744.9?344834 FCSS J033757.0?372542 FCSS J034111.5?364550 APMBGC 359+114+047 FCSS J035159.5?361420 RA (J2000) Dec 03:27:00.2 03:31:54.9 03:37:45.0 03:37:57.1 03:41:11.5 03:51:17.1 03:51:59.5 ?35:48:55 ?34:30:45 ?34:48:34 ?37:25:43 ?36:45:50 ?35:45:49 ?36:14:20 cz ?cz ( km s?1 ) 25548 21972 26232 16923 31302 18102 14505 44 33 133 31 36 51 34

3.4

Spectroscopic Classi?cation

?rming the late-type classi?cation if not the membership. We inspected images of all the re-classi?ed objects by eye on the bJ survey plates and there was no apparent reason why they should not have been given the FCC classi?cations. This only goes to emphasise the limitations of the morphological classi?cation. More interesting, our sample of compact galaxies drawn from the APM data included 10 objects not in the FCC: three of these were objects dominated by the light from foreground stars (Table 3), but 7 are background galaxies listed in Table 7. These were not included in the FCC because of their very compact appearance; the FCC is diameter-limited at 17 arc second, while these objects have scale lengths of 2 arc second or less. Only one has been previously measured (APMBGC 359+114+047) as part of the Stromlo/APM galaxy redshift survey (Loveday et al., 1996); the others we give new names in Table 7 according to the IAU-approved convention for our larger Fornax Cluster Spectroscopic Survey (Drinkwater et al., 2000b). We can use the statistical samples de?ned in Fig. 2 to put limits on the total number of compact new cluster members. In the upper region in the Figure (above the dashed line) we found 7 new members having observed 327 galaxies not listed as members in the FCC; the probability of getting a new member is therefore p ≈ (7/327) = 0.02. Using this as a binomial probability the expected number of new members among the remaining 141 galaxies we didn’t observe in this region is therefore equal to 3 and the 98 per cent upper limit on the additional new members is equal to 7. There is
c 0000 RAS, MNRAS 000, 000–000

Our spectroscopic data not only allow us to con?rm the cluster membership of the Fornax galaxies, but they also give us the opportunity to revise the classi?cations of individual cluster members on the basis of measured spectral indicators of star formation. The FCC galaxies were given morphological classi?cations in the extended Hubble system following the work by Sandage (see FCC). For the purpose of our current analysis we have converted these classi?cations to a one-dimensional sequence of “revised morphological types” or “t-types” following de Vaucouleurs et al. (1991, RC3 hereafter). The correspondences between the FCC types and the t-types we have used are given in Table 8 which is based on Table 2 of the RC3 galaxy catalogue except for the early type dwarfs which are not de?ned in that table. We assigned dE and dS0 galaxies a t-type of ?5 based on the classi?cation of some dwarf galaxies in the Local Group. The one cluster member classi?ed “?” we assigned a value of 10 (Im). The one line detectable in all our spectra is the Hα emission line. We measured Hα equivalent widths interactively using the IRAF task splot, de?ning the continuum level from regions both sides of the Hα line, avoiding the [N II] lines. In Fig. 5 we plot the equivalent widths of Hα as a function of the newly assigned galaxy t-types; the Hα equivalent widths are listed in Table 4. There is a general correlation: most of the late-classi?ed galaxies show emission lines and the dE/dS0 galaxies mostly have low equivalent widths of Hα. However several of the early type galaxies do have signi?cant emission in Hα, notably the new cluster member FCCB 2144 with an equivalent width of 65? that was preA viously classi?ed as “E or dE(M 32)”. Our 3-sigma detection for Hα emission corresponds to an equivalent width of 1.0?. A We classify the 42 galaxies with Hα emission above this limit as late types and the remainder (66) as early types. The new spectroscopic classi?cations reveal a much higher incidence of star-forming dwarf galaxies in the cluster than implied by the original morphological classi?cations, as summarised in Table 9. This trend is discussed further in Section 5.1 where we also consider the e?ect of any bias on the new classi?cations caused by the limited aperture size of our observations. It is clear that reliable morphological classi?cations would need higher resolution and/or dynamic range material than a wide ?eld photographic plate, i.e. the

8

Drinkwater et al.
in Paper III) so as to give a larger sample (62 dwarfs) for our analysis, but there is a natural break in the luminosity distribution at this limit and it matches the luminosity of the brightest local group dwarfs (Mateo, 1998). Our analysis of the galaxy velocities has revealed the ?rst signi?cant evidence for substructure in the Fornax Cluster. This is presented fully in Paper III, so we only summarise it in this section.

4.1

Cluster Membership

Figure 5. Equivalent widths of Hα emission as a function of morphological t-types of 108 Fornax cluster galaxies. The equivalent widths are measured from our FLAIR-II spectra with positive values indicating emission. The t-types are derived from the FCC morphological classi?cations as discussed in the text. Galaxies with widths greater than our 3σ detection limit of 1? (indicated A by the dashed line) are plotted as triangles and the rest as circles with the larger symbols in each case for the giant galaxies. The three galaxies with possible BCD classi?cations are indicated by ?lled triangles. The scale is logarithmic for widths greater than our 1σ limit of 0.3? (indicated by the horizontal axis) and linear A for smaller values.

The overall distribution of the cluster members is shown in Fig. 6 as cone diagrams in Right Ascension and Declination. The Figure shows all objects measured between velocities of zero and 7 000 km s?1 , to emphasise the clear demarcation between the cluster and foreground stars on the one hand and background galaxies on the other. The Figure also differentiates between early- and late-type cluster members as discussed in the next section. The locations of the nine new cluster members are indicated by circles; their positions are not signi?cantly di?erent to those of the previously classi?ed cluster members. Our new measurements represent a large increase in the number of con?rmed dwarf galaxies in the Fornax Cluster, from 26 with redshifts in the FCC to 62 in our sample. However as only 9 of our sample were not already classi?ed as cluster members in the FCC, this does not represent a signi?cant change in the cluster luminosity function. At fainter levels (bJ > 18, MB > ?13.5) the galaxy counts in the FCC tail o? rapidly so it is presumably incomplete.

4.2

Substructure and Dynamics

Table 9. Comparison of new spectroscopic classi?cations with FCC morphological types. The dwarf sample is de?ned as galaxies with bJ ≥ 15.0. Sample dwarf dwarf giant giant total FCC type early late early late – N 50 12 23 23 108 Spectroscopic type absorption emission 39 1 21 5 66 11 11 2 18 42

FCC classi?cations were in?uenced by the data characteristics as noted by Ferguson at the time.

4

CLUSTER STRUCTURE

The 108 con?rmed cluster galaxies we have measured constitute a very useful sample to examine the structure of the cluster as they were uniformly selected over a very large ?eld. This is also the ?rst spectroscopic sample of cluster galaxies to contain a signi?cant fraction of dwarf galaxies. For the purposes of this paper we de?ne dwarf galaxies to be those with bJ ≥ 15 (MB ≥ ?16.5) in our sample. This is slightly brighter than we have used elsewhere (bJ ≥ 15.5

We present histograms of the velocity distributions of cluster galaxies and various subsamples in Fig. 7. The total cluster sample has a marginally non-Gaussian velocity distribution at the 91 per cent con?dence level using the W-test (Royston, 1982). The mean velocity is 1493 ± 36 km s?1 and the velocity dispersion is 374 ± 26 km s?1 . The other subsamples (late- and early-types; dwarfs and giants as de?ned in Section 3) are all consistent with Gaussian distributions. There is however an indication in Fig. 7 of di?erences in the velocity dispersions of the subsamples. The velocity dispersion of the dwarfs (409 ± 37 km s?1 ) is larger than that of the giants (324±34 km s?1 ) at the 90 per cent con?dence level as measured by the F-test (Press et al., 1992). The larger velocity dispersion of the dwarfs, combined with their more extended spatial distribution (Section 4.3) suggests that they are infalling whereas the giants form a virialised population (see Paper III). The dispersions of the late-type (405 ± 45 km s?1 ) and early-type samples (356 ± 31 km s?1 ) are not signi?cantly di?erent. In Paper III we describe in detail the evidence for substructure in this galaxy distribution. We used the KMM mixture modelling algorithm as described by Colless & Dunn (1996) to identify a robust partition of the cluster into a 92-member main cluster centred close to NGC 1399 with cz = 1478 km s?1 and σcz = 370 km s?1 and a 16-member subcluster centred about 3 degrees to the South West with cz = 1583 km s?1 and σcz = 377 km s?1 . The partition is indicated in Table 4 in the form of a probability for each cluster
c 0000 RAS, MNRAS 000, 000–000

Dwarf galaxies in Fornax

9

Figure 6. Distribution of cluster galaxies. Cluster members are indicated as early types (crosses), late-types (triangles) and new members (additional circles). Solid circles indicate foreground stars and background galaxies.

Fig. 8. A simple two-body dynamical model allows for solutions with the subcluster either in front of the main cluster (infalling) and behind the main cluster (moving away) but the infalling solution is slightly more probable. We use these dynamical data in Paper III to make new estimates of the cluster mass using both virial mass estimators and the velocity amplitude method of Diaferio (1999). The cluster mass within a projected radius of 1.4 Mpc is (7 ± 2) × 1013 M⊙ corresponding to a mass-to-light ratio of 300 ± 100 M⊙ / L⊙ .

4.3

Spatial Distribution

Figure 7. Histograms of Fornax Cluster galaxy velocities. In each panel the unshaded histogram shows the full sample of 108 galaxies. The shaded histograms show various subsets: the nine new cluster members (top panel), the 42 late-type galaxies (middle) and the 56 dwarf galaxies (bottom panel).

member that it is a member of the main cluster. Note that the subcluster does not contain the giant barred spiral NGC 1365 which is closer to the centre of the ?eld, but it does contain the active radio galaxy Fornax A (NGC 1316) as well as a high concentration of other late-type galaxies. The two-sigma limits of the cluster and subcluster are shown in
c 0000 RAS, MNRAS 000, 000–000

The projected galaxy distribution in Fig. 8 shows evidence that the late-type galaxies are more widely distributed than the early types. This is con?rmed in Fig. 9, a plot of the normalised cumulative radial distributions for di?erent galaxy subsamples. Note that for this radial analysis we restrict our analysis to the 92 members of the main cluster. Even after the removal of the SW subcluster which is dominated by late-type galaxies, the late-type galaxy subsample is significantly more extended than the rest of the cluster galaxies. The two distributions di?er at the 99 per cent con?dence level measured by the KS test. The dwarf galaxies are also more extended than the giant galaxies (middle panel of the ?gure) at the 99 per cent con?dence level. Restricting the analysis to the dwarf galaxies alone (lower panel of the ?gure) the fraction of late-type dwarfs also decreases towards the cluster centre, but given the smaller sample (56 dwarfs) the di?erence is only at the 85 per cent con?dence level.

10

Drinkwater et al.
5 5.1 STAR FORMATION IN DWARF GALAXIES New star forming galaxies

Figure 8. Projected spatial distribution of Fornax cluster galaxies. Galaxies spectroscopically classi?ed as early type are plotted as circles and late types as triangles. The symbol sizes of the latetype galaxies are scaled by the logarithm of their star formation rates (ranging from 10?3 to 6 M⊙ yr?1 ) derived from their magnitudes and Hα equivalent widths as described in Section 5.3. The dashed ellipses show the 2-sigma limits of the two subclusters.

Figure 9. Cumulative radial distributions of various galaxy samples in the 92-member main part of the Fornax Cluster. In each plot the total distribution is shown as a solid line. These are compared to the early (dotted) and late-type (dashed) subsamples (upper panel), the dwarf (dotted) and giant (dashed) subsamples (middle panel) and the dwarf early (dotted) and dwarf late-type (dashed) subsamples (lower panel).

As described above, we reclassi?ed the measured Fornax Cluster galaxies as late or early types from their spectroscopic properties, independent of their morphologies. This spectral reclassi?cation is based on our FLAIR-II data, i.e. spectra taken through 6.7 arc second ?bres centred on the individual galaxies. We therefore need to consider how representative the spectra are of the whole galaxies, especially as star formation is often centrally-concentrated in dwarf galaxies. Gallagher & Hunter (1989) obtained Hα imaging and spectroscopy for a sample of about 30 Virgo Cluster dwarf irregulars. They found that the Hα emission was signi?cantly centrally-concentrated in 30% of their sample. In the remaining galaxies the distribution of Hα emission followed that of the red continuum light, in which case central measurements of Hα equivalent width would be representative of the whole galaxy. A spectroscopic analysis of the nuclei of Fornax dwarf galaxies (Held & Mould, 1994) found evidence for young stellar population in some of them. Our sample includes two of the same galaxies (FCC 207 and 261) in which we detected weak Hα emission. Held & Mould (1994) remarked that FCC 207 is blue in U-B colour indicating a young population although their spectra did not extend far enough into the red to allow them to detect the Hα emission. Ideally we would avoid any bias by classifying galaxies as a function of their total Hα emission, normalised in some way by the galaxy size or mass. The measurement of Hα equivalent width (EW) is one example of this normalisation. The only reliable way to measure the total Hα emission is with large-aperture spectroscopy or narrow-band Hα imaging. This is not available for our sample, although we plan to obtain Hα images of these galaxies in the future. If the Hα emission were centrally concentrated then our EW measurements would over-estimate the average values for each galaxy. We can estimate a lower limit for the total EW by considering an extreme case where all the emission is in the central region. The total EW is then just our measured value multiplied by the fraction of the total galaxy (continuum) light sampled by the aperture of the spectrograph. The e?ective aperture of the the FLAIR-II system is a convolution of the 6.7 arc second ?bre diameter with image movement during the observation (due to tracking errors and di?erential atmospheric refraction). These errors are di?cult to quantify (Parker & Watson, 1995a), but the image translation during long exposures is at least as large as the ?bre apertures because we have observed signi?cant ?ux variations between di?erent targets between exposures. The e?ective aperture is therefore at least 15 arc seconds in diameter. To calculate the fraction of light observed for each galaxy, we use the exponential scale lengths already measured from the APM data (shown in Fig. 11) and an aperture radius of 8 arc seconds. For a dwarf with scale length of 4 arc seconds this implies we have observed 60% of the light and that the lower limit of the the total EW is 60% of that measured in our spectra. We estimated lower limits of total EW for all 11 dwarfs which we had reclassi?ed as late-type: only one of them resulted in a value lower than our 1? cuto?: FCC 36 origiA nally classi?ed as “dE4 pec,N” in the FCC. So, although our
c 0000 RAS, MNRAS 000, 000–000

Dwarf galaxies in Fornax
FLAIR-II spectra are biased towards light from the galaxy cores (around 8 arc seconds or 800 pc in radius), the detections are strong enough that they would have a signi?cant detection of Hα emission when averaged over the whole galaxy. We are therefore con?dent in our use of spectral classi?cations, although Hα imaging will give us more accurate numbers in the future. Our spectral reclassi?cation results in a higher fraction of late-type galaxies than implied by morphological classi?cation, especially for the dwarfs (see Table 9). Of the 62 dwarfs measured, 12 were morphologically classi?ed as latetype in the FCC (t ≥ 0), but we ?nd that twice this number (22) have detected Hα emission. The fraction of late-type dwarfs in our sample is therefore 35 per cent compared to 19 per cent using the FCC classi?cations. For the total cluster sample in the FCC, the fraction (Im types compared to dE+dS0+Im) is 13 per cent. We compared the properties of the 11 newly identi?ed star-forming dwarfs with the other cluster galaxies. Their radial and velocity distributions are not signi?cantly di?erent to the other galaxies, but we note that their Hα equivalent widths are signi?cantly lower at a KS signi?cance of 94 per cent. As might be expected, it is the dwarfs with lower star formation rates that were not identi?ed by morphology in the FCC. The exception to this is the high star formation galaxies which were previously misclassi?ed as background objects (see Paper I). We show bJ images of all 22 star forming dwarf galaxies in Fig. 10 in order of increasing Hα equivalent width. The low equivalent width dwarfs do resemble dE types, but we do not identify any signi?cant di?erence in the measured morphological parameters scale length and magnitude between the old and new star-forming dwarfs. The new star-forming dwarfs have a slightly brighter mean surface brightness, but the di?erence is only signi?cant at the 84 per cent level. 5.2 Blue compact dwarf galaxies

11

Figure 11. Diagram of surface brightness against scale length for dwarf cluster members (bJ > 15.0). The classi?cations are shown as early types (circles) and late-types (triangles). (a) All probable dwarf cluster members from the FCC using morphological classi?cation; the three BCDs are shown as ?lled triangles. (b) All dwarfs spectroscopically classi?ed from our FLAIR-II observations. The size of the triangles is scaled to the logarithm of the Hα equivalent width.

Blue compact dwarfs (BCDs) are de?ned as high surface brightness, compact, star-forming dwarf galaxies Thuan & Martin (1981). Ferguson (1989b) identi?es a total of 35 candidate BCDs in the FCC, but remarks that there are very few BCDs in the Fornax Cluster (only ?ve of the candidates are classi?ed as probable cluster members). Our spectroscopy of 19 of the candidate BCDs has found that nearly all of these are background galaxies, with only three actually in the cluster–FCC 32, FCC 33, and FCC 35. All three have detected Hα emission so their BCD classi?cation is con?rmed. Note that FCC 33 is actually too bright to fall in our dwarf sample. Interestingly all three of these are in fact members of the spiral-rich “Fornax SW” subcluster identi?ed in Paper III and not the main cluster. The morphological properties of the FCC-classi?ed BCDs are compared to the other Fornax dwarf galaxies in Fig. 11(a) which plots central surface brightness against scale length. All three BCDs have relatively high surface brightness (compared to the other FCC dwarfs) and high Hα equivalent widths, consistent with a high rate of star formation. However when we include the new cluster members and spectral classi?cations in our comparison (Fig. 11(b)) the FCC-classi?ed BCDs are not so distinctive as a class; there are several other star-forming dwarfs with higher surface brightness, although FCC 35 still stands out due to its
c 0000 RAS, MNRAS 000, 000–000

exceptionally high Hα equivalent width. There is a significant variation of Hα widths with scale lengths: the mean equivalent width is 37? for dwarfs with scale lengths less A than 4.5′′ compared to 10? for larger dwarfs; the distribuA tions di?er at a KS signi?cance of 92 per cent. This suggests that the star forming region is very compact. Apart from FCC 35 we cannot identify the FCCclassi?ed BCDs as a separate class from our measured parameters (scale length, magnitude, surface brightness or Hα equivalent width) in our FLAIR-II sample. There are other new cluster members that could also be classi?ed as BCDs, especially those in the low scale length, high surface brightness region of Fig. 11(b). This agrees with the conclusions of Marlowe et al. (1999) from a study of nearby blue “amorphous” star-forming dwarf galaxies. They found that blue amorphous galaxies were indistinguishable from blue starburst dwarfs selected by other means such as BCDs and H II galaxies. On the basis of our measurements we can make a stronger statement for the star-forming dwarfs in the Fornax Cluster: there is no evidence for two distinct populations of “compact” and “irregular” star-forming dwarfs. Instead we ?nd that their distribution of scale sizes is continuous and is not signi?cantly di?erent to that of the dwarf ellipticals. The use of “BCD” as a separate class is confusing here, as it implies a dichotomy not seen in our data. We will therefore avoid any morphological sub-classi?cations of the late-type dwarfs in our discussion of their evolution below; we limit ourselves to a simple classi?cation of the dwarfs into just two classes: early and late-types based on quantitative Hα equivalent width measurements.

12

Drinkwater et al.

Figure 10. Fornax cluster dwarf galaxies showing signi?cant star formation plotted in order of increasing Hα equivalent widths from left-to-right and top-to-bottom. The two galaxies previously classi?ed as BCDs are FCC 32 and FCC 35. These Digitized Sky Survey (DSS) blue optical images are each about 3 arc minutes across.

5.3

Evolution of dwarf galaxies

In previous studies of dwarf galaxies in the Virgo cluster (Drinkwater & Hardy, 1991; Drinkwater et al., 1996) possible evolutionary relationships between BCDs, dwarf elliptical and dwarf irregular galaxies were examined. It was concluded that there were very few if any star-forming dwarfs in Virgo of intermediate size between the BCDs and dwarf irregulars and in consequence that there was no existing population of star-forming galaxies that could be progenitors of the dE cluster population (see also Bothun et al., 1986). However in the Virgo analysis (Drinkwater et al., 1996), the sample was only complete for very compact galaxies and relied on morphological classi?cations for the intermediate sized dwarfs. Our current sample of Fornax Cluster dwarfs includes the full range of scale sizes and, in consequence, requires us to revise the conclusions of Drinkwater et al. (1996). To see the e?ects of our improved sample we compare Fig. 11(b) with the earlier plot of Drinkwater & Hardy (1991, Fig. 4) based on CCD measurements of individual Virgo dwarfs. From the Virgo sample it was concluded that the star forming dwarfs–BCDs and irregulars, as de?ned by morphological classi?cations–occupied two extreme regions of the diagram at small and large scales respectively with no late-type dwarfs at intermediate scales. In our new data the region of the diagram previously only occupied by dEs of intermediate scale sizes (5-10 arc seconds) is also populated with star forming dwarfs: galaxies with the whole range of scale lengths measured have detected emission lines. On the basis of this diagram alone we can now revise the conclusion of Drinkwater et al. (1996) that there is no current population of star-forming progenitors of cluster dwarf ellipticals. As we now ?nd star-forming dwarfs at all scale lengths, it is much more plausible to model dwarf galaxy evolution by transformations from late- to early-type dwarfs. To determine the factors that drive the evolution we must however examine the distribution of galaxies within the cluster. The projected distribution of all Fornax galaxies in Fig. 8 suggests immediately that the star-formation activity in the cluster is not concentrated in the same regions as the early-type galaxies. The late-type galaxies are plotted as symbols with sizes related to their absolute star formation rates. These were calculated using a relation derived from Kennicutt (1992): 2.7 × 10?12 (LB /LB⊙ )EW (Hα) M⊙ yr?1 . The Figure shows the large amount of star formation tak-

ing place in the SW subcluster, 3 M⊙ yr?1 out of a total of 11 M⊙ yr?1 for the whole cluster. The giant galaxies are included in this total star formation rate, so it is entirely dominated by NGC 1365 (nearer the cluster centre) producing 6 M⊙ yr?1 . If we concentrate instead on the dwarf galaxies alone and ignore the spiral-rich SW subcluster, the radial distributions in Fig. 9(c) con?rm the de?cit of star-forming dwarfs at small projected radii from the centre of the cluster. The fraction of late-type dwarfs falls from 44 per cent at radii greater than 105 arc minutes to 25 per cent at smaller radii (the radial distributions di?er at a Kolmogorov-Smirnov con?dence level of 85 per cent). This is consistent with the normal density-morphology relation seen for more luminous cluster galaxies. The 5 star-forming dwarfs found at small projected radius from the centre have lower surface brightness and star formation rates than the other star-forming dwarfs, but the di?erence is not statistically signi?cant because of the small sample. Our data are therefore consistent with the cluster core being entirely devoid of star-forming dwarfs, the 5 “central” dwarfs being physically located in front of or behind the core. The lack of any star-forming dwarfs in the cluster centre provides strong evidence against a very simple evolutionary model in which all dwarfs undergo short, sporadic bursts of star formation (e.g. Davies & Phillipps, 1988). A number of processes have been proposed to explain the density-morphology relation, mainly in the context of the giant galaxy population of clusters. There is strong evidence from radio observations that ram pressure stripping can remove the gas from spirals that pass near the cluster cores (van Gorkom, 1996), although it has been argued that this is not such an important process for the dwarf galaxy population of poor clusters like Virgo and Fornax (Ferguson & Binggeli, 1994) because of the lower central gas density. More recently there has been considerable discussion of dynamical evolution whereby tidal forces from either the cluster or the larger member galaxies can result in signi?cant morphological evolution, enough to convert dwarf irregulars to dwarf ellipticals (Moore et al., 1998; Mayer et al., 2001). The second process (“tidal stirring”) was proposed in the context of galaxies in the Local Group rather than a rich cluster, but it may also apply in a poor cluster such as Fornax whose central mass is not much greater than that of the local group. The cumulative radial distribution of late-type dwarfs in Fig. 9(c) exhibits a sharp jump between projected radii of
c 0000 RAS, MNRAS 000, 000–000

Dwarf galaxies in Fornax

13

Figure 12. Radial distribution of dwarf galaxies in the Fornax Cluster. The plots give the number of galaxies in annuli of equal area, so the numbers in each bin are directly comparable. In (a) the dwarfs are separated by their FCC morphological classi?cations (248 early-type and 31 late-type dwarfs). In (b) the dwarfs are separated by their spectral classi?cations from our FLAIR data (40 early-type and 22 late-type dwarfs). The excess of latetype dwarfs at radii between 100 and 130 arc minutes is evident in both plots.

ter core once had similar properties then some active process must have removed the gas as they fell into the cluster; these star formation rates are not high enough to exhaust the gas in a Hubble time. The dwarf galaxies in Fornax appear to trace a complex evolutionary path. In the cluster core there is virtually no current star formation, although Held & Mould (1994) showed that several of the nucleated dwarf ellipticals have experienced recent star formation activity. This is the classical density-morphology relation seen here for dwarf galaxies, driven by gas removal and morphological transformation. In the outer regions of the cluster however, at a radius of about 600 kpc, there is a concentration of star-forming dwarfs. This resembles a similar concentration of post starburst galaxies at similar radii around rich clusters (Dressler et al., 1999). We plan to obtain blue spectra of the Fornax dwarfs in order to look for post starburst galaxies. This combination of enhanced star formation in the cluster fringes and complete suppression in the cluster core is very consistent with the results of Hashimoto et al. (1998) who measured the spectral properties of a very large sample of galaxies from the Las Campanas Redshift Survey to determine how star formation rate depends on the local galaxy density. They found that starburst activity was enhanced in regions of intermediate density, but that star formation was suppressed in regions of high density like the centres of galaxy clusters.

6

COMPACT DWARF ELLIPTICAL (M 32-LIKE) CANDIDATES

100 and 120 arc minutes. This corresponds to an increased density at that particular radius as is shown in Fig. 12 which plots the surface density of the dwarfs as a function of projected radius. There is a large increase in the number of late-type dwarfs at projected physical radii around 600 kpc which is evident in the FCC data as well as our smaller FLAIR sample. This concentration resembles that of post starburst galaxies at similar radii around rich clusters reported by Dressler et al. (1999). Some process such as interaction (e.g. Moss & Whittle, 1993) is triggering star formation as the dwarfs reach this particular radius. The Fornax dwarfs show signi?cant evidence of current infall compared to the virialised giant cluster galaxies (see Paper III). Assuming there is no signi?cant gas stripping at this large distance from the cluster centre, we can estimate the timescale for the current star formation in some of these dwarfs for which HI measurements are available (Schr¨der o et al., 2001). Using their HI masses, we can estimate the depletion time scales assuming the current rates of star formation estimated from the Hα data and an extra factor of order 2 to allow for recycling of the gas (Kennicutt et al., 1994). The depletion time scales are 5 × 109 yr (FCC 35), 3 × 1010 yr (FCC 113), 1 × 1010 yr (FCC 139), 5 × 1011 yr (FCC 302) and 2 × 1010 yr (FCC 306). (The time scales for the giant star-forming galaxies span a similar range.) Apart from FCC 35 (a special case as it is in the active subcluster) these timescales are all long, of order both the Hubble time and the orbital period of the cluster at this radius (3 × 1010 yr). If the early-type galaxies already in the clusc 0000 RAS, MNRAS 000, 000–000

We observed 97 of the M 32-like cdE candidates listed in the FCC (Tables 13 and 3) and successfully determined redshifts for 76 of these. Only one of these was a cluster member (FCC B2144), but it had a blue spectrum with emission lines, and therefore is a BCD-like galaxy, not a normal red dwarf elliptical (details in Paper I). The remaining 75 cdE candidates found to be background objects are indicated in Table A1. The FCC criteria for cdE candidates included possible background galaxies, although it was noted that most of these were probably isolated background ellipticals. We calculated absolute magnitudes and estimated scale lengths and surface brightnesses for all 75 cdE candidates in the background sample using a Hubble constant of 75 km s?1 Mpc?1 . These are shown in Fig. 13 along with all the other galaxies we measured for comparison. Most of the cdE candidates have redshifts of 10 000 km s?1 or more and absolute magnitudes brighter than MB =?18 and are giant ellipticals as suggested. Two of the candidate cdEs were somewhat closer: FCCB 278 at v=7960 km s?1 , MB =?17.5 and FCCB 712 at v=6670 km s?1 , MB =?17.2 with scale lengths of 1000 and 780 pc respectively. These are both much more luminous and larger than M 32 which has an absolute magnitude MB =?15.8 (Mateo, 1998) and scale length of 80 pc (see below). We therefore conclude that none of the candidate cdEs listed in the FCC resemble M 32 (see locations of ?lled symbols in Fig. 13). Following the philosophy of this paper of replacing morphological classi?cations by quantitative measures, have included in Fig. 13 the parameters of all the compactappearing galaxies we measured in addition to the cdE

14

Drinkwater et al.
Fornax cluster. A similar search for M 32-like galaxies in the Leo group was unsuccessful (Ziegler & Bender, 1998), so this may be a general result. These observations give some support to the conclusions of Nieto & Prugniel (1987) that tidal stripping is not a realistic formation scenario for such objects. However we do note that tidal stripping does appear to be a realistic explanation (Bekki et al., 2001) for the formation of a less luminous class of compact galaxy recently found in the Fornax Cluster (Drinkwater et al., 2000a)

7

CONCLUSIONS

Figure 13. Structural parameters of candidate cdE (M 32-like) galaxies with measured redshifts as a function of absolute magnitude. In each plot background galaxies are plotted as circles and cluster members as squares; the candidate cdE galaxies (as listed in the FCC) are plotted as ?lled symbols. Upper panel: the distribution of central surface brightness with the position of M 32 given by a mean surface brightness which is a lower limit to the central surface brightness (see text). Lower panel: the distribution of physical scale lengths. The position M 32 would occupy is indicated by the large asterisk and the region of M 32-like objects discussed in the text is enclosed by dashed lines. Note that these structural parameters were calculated from exponential ?ts to the photographic images and have not been corrected for seeing or cosmological fading.

candidates. We also indicate the position that M 32 would occupy in the diagram, although the values of its surface brightness and scale length parameters are not directly comparable as they are based on images of much higher physical resolution than our data. We use the data of Faber et al. (1989) who measure an e?ective radius of 140 pc (corresponding to an exponential scale length of 80 pc) and a mean surface brightness within this radius of ?B = 18.7 mag arcsec?2 which gives a lower limit for the central surface brightness. As discussed above, none of the cdE candidates have sizes and luminosities comparable to M 32 but there are a number of compact cluster galaxies in our sample with similar luminosities as indicated by the dashed line in the Figure. There are 20 galaxies in this region; nine (including the only background galaxy FCCB 138) have late-type spectroscopic classi?cations, but the rest are early-type dwarf galaxies (most with FCC classi?cations of dE,N). These are potentially similar to M 32, but their much larger scale lengths according to Fig. 13 show that they are not like M 32, as we might expect since they were not classi?ed as cdE candidates in the FCC. As a result of our investigation of 75 cdE candidates from the FCC and our own, larger, sample of compact galaxies, we conclude that there are no galaxies like M 32 in the

In this paper we have made the ?rst detailed study of the Fornax Cluster in which morphological and membership classi?cations have been replaced by quantitative image parameters and spectroscopic memberships. The image parameters were obtained from complete catalogues of digitised sky survey plates, allowing for a statistical approach to our analysis. The spectroscopic data have allowed us to con?rm the membership of many cluster galaxies originally estimated from visual morphological classi?cation (Ferguson, 1989b). In our search for compact cluster galaxies we have found 9 new compact cluster dwarf galaxies previously classi?ed as background galaxies. We used our spectroscopic data to estimate star formation rates from Hα emission equivalent widths. The total star formation rate for the cluster is dominated by the giant spiral NGC 1365, but the remaining star formation is concentrated in a separate subcluster we have identi?ed 3 degrees South West of the main cluster, centred on NGC 1316 (Fornax A) as described in Paper III. Among the dwarf galaxies we ?nd a much higher incidence of detected star formation than was suggested by the original morphological classi?cations. In the 62 dwarf galaxies we measured 35 per cent are star-forming but only 19 per cent were classi?ed as late types. Our spectral classi?cation is based on emission from the cores of each galaxy, but we estimate that a correction to include the whole galaxy would change at most one of our classi?cations. Only three of the BCD candidates were found to be cluster members. Our data do not support the existence of a well-de?ned separate class of BCD-like objects, but rather we detect actively star forming dwarf galaxies over the whole range of measured sizes. We have also shown that there is no group of very compact BCDs without corresponding earlytype dwarfs at the same scale sizes. The distribution of scale sizes is consistent with model of dwarf galaxy evolution involving morphological transformation from late- to earlytype galaxies. This is in contrast to earlier work on Virgo Cluster dwarfs (Drinkwater et al., 1996) which was not based on spectroscopic classi?cations. The fraction of star-forming dwarf galaxies in Fornax drops signi?cantly towards the cluster centre. This observation alone is su?cient to rule out simple evolutionary models where the dwarfs form a single population with all experiencing regular short bursts of star formation. We estimate long gas depletion timescales for the few dwarfs with detected H I emission. This implies that an active process of gas depletion is needed to transform them to quiescent galaxies like the central dwarfs. We conclude that star formation in dwarf cluster galaxies is a complex process in?uenced by their enc 0000 RAS, MNRAS 000, 000–000

Dwarf galaxies in Fornax
vironment: there are less star forming dwarfs at the cluster centre presumably due to some form of gas stripping. High rates of star formation are found at the edges of the cluster, perhaps stimulated by tidal interactions. Our data also allowed us to look at the candidate compact elliptical galaxies listed by Ferguson (1989b). We ?nd that none of these have properties like M 32: they are all background giant ellipticals except for one cluster member which is less luminous than M 32 and is also a spectroscopic late type galaxy. We broadened our search to the larger sample of compact galaxies we observed and found that, although some of the brighter cluster dwarfs had similar luminosities to M 32, none of them had similarly small scale lengths. We therefore con?rm our earlier conclusions Drinkwater & Gregg (1998) that there are no galaxies like M 32 in the Fornax Cluster. We do note however that our 2dF observations have revealed a population of very compact objects at the centre of the cluster which resemble lowluminosity (?13 < MB < ?11) versions of M 32 (Drinkwater et al., 2000a).

15

ACKNOWLEDGEMENTS We are very grateful to all the sta? of the UK Schmidt Telescope for the assistance with our many observing runs. The referee of this paper made a number of suggestions which have greatly improved the presentation of this work. We also wish to thank Jon Davies for a copy of his galaxy catalogue with the APM photometry, Bryn Jones for discussions of morphological t-types, and Marion Schmitz (of NED) for checking our galaxy lists prior to publication. Part of this work was done at the Institute of Geophysics and Planetary Physics, under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48. This material is based upon work supported by the National Science Foundation under Grant No. 9970884. The Digitized Sky Surveys (DSS) were produced at the Space Telescope Science Institute under U.S. Government grant NAG W-2166. The images of these surveys are based on photographic data obtained using the Oschin Schmidt Telescope on Palomar Mountain and the UK Schmidt Telescope. The plates were processed into the present compressed digital form with the permission of these institutions. The UK Schmidt Telescope was operated by the Royal Observatory Edinburgh, with funding from the UK Science and Engineering Research Council (later the UK Particle Physics and Astronomy Research Council), until 1988 June, and thereafter by the Anglo-Australian Observatory. The blue plates of the southern Sky Atlas and its Equatorial Extension (together known as the SERC-J), as well as the Equatorial Red (ER), and the Second Epoch [red] Survey (SES) were all taken with the UK Schmidt.

References Bekki K., Couch W. J., Drinkwater M. J., 2001, ApJ, 89, in press Binggeli B., Popescu C. C., Tammann G. A., 1993, A&AS, 98, 275
c 0000 RAS, MNRAS 000, 000–000

Binggeli B., Sandage A., Tammann G. A., 1985, AJ, 90, 1681 Binggeli B., Sandage A., Tarenghi M., 1984, AJ, 89, 64 Bothun G. D., Mould J. R., Caldwell N., MacGillivray H. T., 1986, AJ, 92, 1007 Bureau M., Mould J. R., Staveley-Smith L., 1996, ApJ, 463, 60 Colless M., Dunn A. M., 1996, ApJ, 458, 435 Couch W. J., Ellis R. S., MacLaren I., Malin D. F., 1991, MNRAS, 249, 606 Davies J. I., Phillipps S., 1988, MNRAS, 233, 553 Davies J. I., Phillipps S., Cawson M. G. M., Disney M. J., Kibblewhite E. J., 1988, MNRAS, 232, 239 de Vaucouleurs G., de Vaucouleurs A., Corwin J. R., Buta R. J., Paturel G., Fouque P., 1991, Third reference catalogue of Bright galaxies. New York: Springer-Verlag Diaferio A., 1999, MNRAS, 309, 610 Dressler A., Smail I., Poggianti B. M., Butcher H., Couch W. J., Ellis R. S., Oemler A. J., 1999, ApJS, 122, 51 Drinkwater M., Hardy E., 1991, AJ, 101, 94 Drinkwater M. J., Currie M. J., Young C. K., Hardy E., Yearsley J. M., 1996, MNRAS, 279, 595 Drinkwater M. J., Gregg M. D., 1998, MNRAS, 296, L15 Drinkwater M. J., Gregg M. D., Colless M., 2001, ApJ, 548, L139 Drinkwater M. J., Gregg M. D., Holman B. A., 1997, in ASP Conf. Ser. 116: The Nature of Elliptical Galaxies; 2nd Stromlo Symposium, p. 287 Drinkwater M. J., Jones J. B., Gregg M. D., Phillipps S., 2000a, Publications of the Astronomical Society of Australia, 17, 227 Drinkwater M. J., Phillipps S., Jones J. B., Gregg M. D., Deady J. H., Davies J. I., Parker Q. A., Sadler E. M., Smith R. M., 2000b, A&A, 355, 900 Faber S. M., 1973, ApJ, 179, 423 Faber S. M., Wegner G., Burstein D., Davies R. L., Dressler A., Lynden-Bell D., Terlevich R. J., 1989, ApJS, 69, 763 Ferguson H. C., 1989a, Ap&SS, 157, 227 —, 1989b, AJ, 98, 367 Ferguson H. C., Binggeli B., 1994, A&A Rev., 6, 67 Ferguson H. C., Sandage A., 1988, AJ, 96, 1520 Gallagher J. S., Hunter D. A., 1989, AJ, 98, 806 Hashimoto Y., Oemler A. J., Lin H., Tucker D. L., 1998, ApJ, 499, 589 Held E. V., Mould J. R., 1994, AJ, 107, 1307 Hilker M., Infante L., Vieira G., Kissler-Patig M., Richtler T., 1999, A&AS, 134, 75 Holman B., Drinkwater M., Gregg M., 1995, in ASP Conf. Ser. 84: IAU Colloq. 148: The Future Utilisation of Schmidt Telescopes, p. 107 Irwin M., Maddox S., McMahon R., 1994, Spectrum, 2, 14 Irwin M. J., Davies J. I., Disney M. J., Phillipps S., 1990, MNRAS, 245, 289 Kennicutt R. C., 1992, ApJ, 388, 310 Kennicutt R. C., Tamblyn P., Congdon C. E., 1994, ApJ, 435, 22 Kurtz M. J., Mink D. J., 1998, PASP, 110, 934 Loveday J., 1996, MNRAS, 278, 1025 Loveday J., Peterson B. A., Maddox S. J., Efstathiou G., 1996, ApJS, 107, 201 Maddox S. J., Efstathiou G., Sutherland W. J., Loveday J., 1990, MNRAS, 243, 692

16

Drinkwater et al.

Marlowe A. T., Meurer G. R., Heckman T. M., 1999, ApJ, 522, 183 Mateo M. L., 1998, ARA&A, 36, 435 Mayer L., Governato F., Colpi M., Moore B., Quinn T., Wadsley J., Stadel J., Lake G., 2001, ApJ, 547, L123 Michard R., Nieto J. ., 1991, A&A, 243, L17 Moore B., Lake G., Katz N., 1998, ApJ, 495, 139 Morshidi-Esslinger Z., Davies J. I., Smith R. M., 1999, MNRAS, 304, 297 Moss C., Whittle M., 1993, ApJ, 407, L17 Mould J. R., Hughes S. M. G., Stetson P. B., Gibson B. K., Huchra J. P., Freedman W. L., Kennicutt R. C., Bresolin F., Ferrarese L., Ford H. C., Graham J. A., Han M., Hoessel J. G., Illingworth G. D., Kelson D. D., Macri L. M., Madore B. F., Phelps R. L., Prosser C. F., Rawson D., Saha A., Sakai S., Sebo K. M., Silbermann N. A., Turner A. M., 2000, ApJ, 528, 655 Nieto J. ., Prugniel P., 1987, A&A, 186, 30 Parker Q. A., Watson F. G., 1995a, FLAIR System Users Guide. AAO User Manual 31.1, Anglo-Australian Observatory, 1995 —, 1995b, Proc. SPIE, 2476, 34 Press W. H., Teukolsky S. A., Vetterling W. T., Flannery B. P., 1992, Numerical recipes in C. The art of scienti?c computing. Cambridge: University Press, —c1992, 2nd ed. Royston J. P., 1982, ppl. Stat., 31, 176 Schr¨der A., Drinkwater M. J., Richter O.-G., 2001, A&A, o submitted, in press Thuan T. X., Martin G. E., 1981, ApJ, 247, 823 Tody D., 1993, in ASP Conf. Ser. 52: Astronomical Data Analysis Software and Systems II, Vol. 2, p. 173 van Gorkom J., 1996, in ASP Conf. Ser. 106: The Minnesota Lectures on Extragalactic Neutral Hydrogen, p. 293 Ziegler B. L., Bender R., 1998, A&A, 330, 819

APPENDIX A: BACKGROUND GALAXIES The majority of the galaxies we measured background objects behind the Fornax Cluster as expected from our compact selection criteria. We list these 408 background galaxies in Table A1 below. We note that there are two background clusters identi?ed behind the central region of the Fornax Cluster: a poor cluster at z = 0.11 (Hilker et al., 1999) and a more distant cluster (“J1556.15BL”) at unknown redshift (Couch et al., 1991). Neither of these are evident in our data, the ?rst because the members are fainter than our magnitude limit and the second, presumably, because it is at too high a redshift.

c 0000 RAS, MNRAS 000, 000–000

Dwarf galaxies in Fornax
Table A1. Catalogue of all background galaxies with measured redshifts
FCC FCCB 0102 FCCB 0116 FCCB 0119 FCCB 0126 FCCB 0127 FCCB 0138 FCCB 0153 FCCB 0162 FCC 24 FCCB 0185 FCCB 0183 FCCB 0178 FCCB 0187 FCCB 0192 FCCB 0208 FCCB 0206 FCCB 0213 FCCB 0216 FCCB 0219 FCCB 0248 FCCB 0245 FCCB 0249 FCCB 0255 FCCB 0252 FCCB 0258 FCCB 0265 FCCB 0266 FCCB 0272 FCCB 0274 FCCB 0276 FCCB 0278 FCCB 0288 FCCB 0290 FCCB 0305 FCCB 0298 FCCB 0309 FCCB 0311 FCCB 0315 FCCB 0331 FCCB 0332 FCCB 0328 FCCB 0339 FCCB 0336 FCCB 0337 FCCB 0346 FCCB 0347 FCCB 0358 FCCB 0359 FCCB 0362 FCCB 0373 FCCB 0379 FCCB 0378 FCCB 0376 FCCB 0389 FCCB 0388 FCCB 0391 FCCB 0403 FCCB 0413 FCCB 0420 FCCB 0422 FCCB 0424 Name 1 FCCB 0438 FCCB 0446 FCCB 0443 FCCB 0447 FCCB 0448 FCCB 0441 FCCB 0445 FCCB 0453 FCCB 0450 FCCB 0454 FCCB 0459 FCCB 0458 FCCB 0466 FCCB 0469 FCCB 0465 FCCB 0477 FCCB 0475 FCCB 0471 FCCB 0482 FCCB 0490 FCCB 0501 FCCB 0502 FCCB 0513 FCCB 0516 FCCB 0517 FCCB 0522 FCCB 0532 FCCB 0535 FCCB 0542 FCCB 0548 FCCB 0547 FCCB 0556 FCCB 0561 FCCB 0567 FCCB 0566 FCCB 0570 FCCB 0576 FCCB 0582 FCCB 0580 FCCB 0594 RA (J2000) Dec 03:22:26.6 03:22:42.7 03:22:46.6 03:22:52.9 03:22:57.8 03:23:02.4 03:23:19.0 03:23:23.7 03:23:31.7 03:23:37.6 03:23:38.2 03:23:38.2 03:23:38.2 03:23:39.6 03:23:51.3 03:23:52.1 03:23:56.4 03:23:58.7 03:23:59.5 03:24:24.0 03:24:25.6 03:24:26.5 03:24:36.6 03:24:36.8 03:24:39.3 03:24:44.8 03:24:45.5 03:24:48.0 03:24:52.0 03:24:56.4 03:24:58.0 03:25:12.1 03:25:16.4 03:25:23.5 03:25:23.5 03:25:30.5 03:25:34.3 03:25:38.9 03:25:43.3 03:25:45.0 03:25:46.8 03:25:53.4 03:25:53.6 03:25:53.6 03:25:59.0 03:26:02.2 03:26:08.2 03:26:13.0 03:26:16.8 03:26:17.1 03:26:20.3 03:26:22.7 03:26:23.5 03:26:29.5 03:26:33.0 03:26:35.1 03:26:41.7 03:26:48.7 03:26:51.5 03:26:51.5 03:26:53.2 03:27:00.2 03:27:00.3 03:27:09.2 03:27:09.3 03:27:09.4 03:27:11.5 03:27:11.5 03:27:12.9 03:27:16.7 03:27:18.0 03:27:22.6 03:27:25.2 03:27:26.8 03:27:30.8 03:27:30.9 03:27:35.0 03:27:36.5 03:27:37.7 03:27:38.0 03:27:43.7 03:27:45.5 03:27:57.9 03:27:58.3 03:28:08.0 03:28:13.4 03:28:14.4 03:28:15.6 03:28:24.3 03:28:27.6 03:28:38.4 03:28:43.4 03:28:44.3 03:28:46.5 03:28:54.0 03:28:56.5 03:28:56.7 03:28:57.5 03:29:05.0 03:29:10.0 03:29:10.2 03:29:22.3 ?36:15:30 ?36:26:26 ?36:21:12 ?36:46:03 ?34:17:16 ?36:54:09 ?36:00:22 ?37:32:10 ?34:36:34 ?36:05:50 ?35:06:53 ?33:27:42 ?36:53:14 ?36:54:31 ?35:05:34 ?34:05:14 ?34:54:29 ?34:19:40 ?36:03:57 ?37:20:32 ?34:22:26 ?36:35:27 ?34:06:14 ?32:34:13 ?35:33:53 ?36:42:48 ?36:19:42 ?36:18:14 ?37:04:32 ?34:39:17 ?34:47:17 ?33:05:02 ?32:28:14 ?37:15:55 ?33:03:25 ?34:52:08 ?33:52:21 ?32:01:07 ?37:31:26 ?37:35:60 ?33:58:56 ?35:40:14 ?33:43:37 ?34:19:52 ?36:01:43 ?34:06:35 ?37:43:10 ?34:48:40 ?33:33:02 ?37:54:14 ?36:48:05 ?35:02:28 ?34:05:03 ?36:10:19 ?32:40:05 ?33:30:21 ?34:59:35 ?34:15:12 ?34:22:24 ?35:14:37 ?34:20:04 ?35:48:55 ?36:35:42 ?37:09:11 ?34:47:41 ?38:19:05 ?37:27:52 ?32:42:16 ?33:24:10 ?36:51:57 ?34:41:47 ?33:56:40 ?35:59:49 ?33:40:32 ?36:35:02 ?37:36:28 ?33:03:58 ?37:27:23 ?33:48:24 ?32:46:02 ?35:01:49 ?37:36:15 ?36:03:53 ?35:56:34 ?35:02:44 ?34:12:35 ?34:38:37 ?36:03:46 ?34:52:25 ?34:55:45 ?32:49:37 ?33:59:52 ?32:44:54 ?37:13:34 ?36:34:47 ?38:04:18 ?36:05:05 ?37:42:46 ?35:19:34 ?35:52:35 ?34:57:41 ?37:06:60 cz 25572 19364 19231 12131 15727 4184 19729 12113 13222 19296 26186 20402 12315 19608 26163 19553 19408 19229 19350 24905 19371 17043 15265 19156 29831 28839 19634 19421 22077 32670 7957 19875 19095 16334 19675 15359 14306 13763 16664 24721 34526 15902 19490 15251 19421 19506 24981 15655 20554 19297 19222 15650 19686 31840 22219 14783 27896 26202 9351 19731 14742 25548 13660 19202 13538 9287 17520 16671 19001 19167 32180 9392 6640 39044 33769 9253 34898 18887 14121 20428 15508 9331 24336 24294 31301 19550 14735 30321 15107 15401 19235 20549 19238 19310 19276 19855 19156 17456 15943 16759 15360 30911 σcz 99 59 43 45 21 34 85 46 14 55 44 36 51 41 65 86 61 88 33 99 42 37 101 58 39 78 62 42 42 99 38 66 74 37 62 54 41 30 39 64 63 50 73 44 58 29 69 78 74 80 37 83 102 44 66 42 51 105 60 29 33 44 34 64 51 65 45 34 71 60 65 45 32 102 106 39 96 24 190 62 86 32 88 90 51 39 68 52 28 79 62 40 34 42 65 52 39 73 76 40 46 107 cdE FCC FCCB 0599 FCCB 0598 FCCB 0600 FCCB 0605 FCCB 0616 FCCB 0624 FCCB 0631 FCCB 0638 FCCB 0634 FCCB 0653 FCCB 0650 FCCB 0651 FCCB 0654 FCCB 0657 FCCB 0674 FCCB 0672 FCCB 0679 FCCB 0676 FCCB 0693 FCCB 0692 FCCB 0691 FCCB 0702 FCCB 0703 FCCB 0707 FCCB 0712 FCCB 0718 FCCB 0727 FCCB 0740 FCCB 0734 FCCB 0731 FCCB 0743 FCCB 0747 FCCB 0753 Name 2 FCCB 0754 FCCB 0757 FCCB 0761 FCCB 0774 FCCB 0777 FCCB 0786 FCCB 0792 FCCB 0793 FCCB 0797 FCCB 0801 FCCB 0805 FCCB 0817 FCCB 0816 FCCB 0815 FCCB 0829 FCCB 0838 FCCB 0847 FCCB 0854 FCCB 0858 FCCB 0859 FCCB 0860 FCCB 0871 FCCB 0874 FCCB 0883 FCC 123 FCCB 0882 FCCB 0902 FCCB 0897 FCCB 0909 FCCB 0920 FCCB 0916 FCCB 0921 FCCB 0926 FCCB 0925 FCCB 0931 FCCB 0939 FCCB 0938 FCCB 0943 FCCB 0946 FCCB 0950 FCCB 0948 FCCB 0951 FCCB 0962 FCCB 0963 FCCB 0960 FCCB 0968 FCCB 0986 FCCB 0994 FCCB 0981 FCCB 0984 FCCB 0993 FCCB 1009 FCCB 1004 FCCB 1014 FCCB 1013 FCCB 1020 FCCB 1027 FCCB 1029 FCCB 1034 FCCB 1056 FCCB 1081 FCCB 1083 FCCB 1087 FCCB 1090 FCCB 1079 FCCB 1084 FCCB 1085 FCCB 1094 RA (J2000) Dec 03:29:23.6 03:29:25.9 03:29:28.1 03:29:31.0 03:29:47.9 03:29:55.4 03:30:01.4 03:30:08.2 03:30:10.7 03:30:24.5 03:30:25.3 03:30:25.7 03:30:27.3 03:30:30.5 03:30:41.9 03:30:44.4 03:30:45.0 03:30:49.7 03:30:57.7 03:31:00.1 03:31:01.2 03:31:04.6 03:31:08.6 03:31:12.6 03:31:19.6 03:31:26.3 03:31:30.3 03:31:40.6 03:31:41.5 03:31:41.8 03:31:48.2 03:31:49.5 03:31:51.8 03:31:54.9 03:31:55.6 03:31:59.5 03:31:59.5 03:32:04.1 03:32:08.2 03:32:17.0 03:32:24.1 03:32:24.9 03:32:34.1 03:32:35.9 03:32:49.4 03:32:51.1 03:32:51.5 03:32:53.3 03:33:00.6 03:33:04.4 03:33:13.4 03:33:14.1 03:33:19.3 03:33:19.6 03:33:24.7 03:33:36.2 03:33:40.6 03:33:41.4 03:33:43.4 03:33:46.9 03:33:53.9 03:33:56.4 03:34:01.4 03:34:03.3 03:34:06.5 03:34:07.5 03:34:09.3 03:34:10.0 03:34:11.5 03:34:21.7 03:34:24.8 03:34:26.0 03:34:28.3 03:34:31.6 03:34:32.5 03:34:37.7 03:34:46.0 03:34:48.0 03:34:49.2 03:34:50.8 03:35:05.9 03:35:06.0 03:35:06.0 03:35:08.6 03:35:13.3 03:35:17.6 03:35:19.4 03:35:19.5 03:35:20.0 03:35:29.6 03:35:34.4 03:35:39.4 03:35:47.5 03:36:07.1 03:36:22.3 03:36:25.1 03:36:27.5 03:36:27.7 03:36:27.9 03:36:30.2 03:36:30.9 03:36:38.3 ?37:21:28 ?35:08:29 ?34:29:12 ?37:28:31 ?36:46:57 ?34:39:33 ?35:36:26 ?38:08:58 ?34:14:59 ?36:22:35 ?33:38:53 ?33:58:58 ?34:40:15 ?35:22:08 ?36:55:16 ?34:38:45 ?38:11:24 ?34:38:44 ?37:19:01 ?35:28:09 ?34:23:28 ?37:55:29 ?37:07:09 ?36:14:26 ?35:35:01 ?35:05:07 ?37:42:04 ?37:21:57 ?34:17:03 ?32:34:03 ?34:23:45 ?37:04:10 ?38:14:52 ?34:30:45 ?36:21:07 ?35:24:34 ?36:25:17 ?37:59:58 ?37:47:21 ?37:22:42 ?37:48:14 ?38:05:47 ?33:18:19 ?36:33:10 ?32:40:07 ?34:51:07 ?34:29:30 ?33:08:22 ?35:27:07 ?37:07:13 ?35:04:29 ?36:34:02 ?35:20:43 ?35:40:32 ?32:38:24 ?35:10:03 ?33:00:54 ?38:21:52 ?35:51:33 ?34:19:44 ?35:51:43 ?32:39:37 ?33:55:24 ?37:19:43 ?33:54:32 ?34:32:23 ?34:05:54 ?33:39:14 ?37:48:50 ?35:45:09 ?33:04:30 ?36:49:09 ?36:53:26 ?36:52:20 ?35:03:55 ?35:33:30 ?36:21:51 ?35:33:31 ?33:13:22 ?36:02:03 ?35:51:31 ?37:54:51 ?33:38:12 ?32:16:16 ?32:25:26 ?36:15:07 ?33:06:42 ?36:22:38 ?35:59:30 ?33:13:55 ?36:18:54 ?36:17:48 ?32:58:42 ?33:24:02 ?38:10:19 ?36:53:30 ?37:15:43 ?37:57:25 ?33:58:28 ?33:44:01 ?33:27:26 ?34:22:56 cz 12069 13926 31614 19126 29276 28591 16719 29280 16333 19206 36348 36699 19017 13335 14002 28736 12321 28444 18018 18974 21139 17661 34247 6543 6666 21130 23734 50131 24942 22336 15838 18290 12129 21972 29967 20979 30043 20140 24327 19194 24476 17712 27130 11986 19756 21467 14374 38232 14369 23859 21510 30148 21157 19122 22145 21396 26930 20210 15483 21982 15463 22471 24673 23435 27963 21186 19833 28322 19136 21452 13972 18958 23740 18946 19618 28887 20894 19118 30514 19213 39410 19073 15970 13908 13726 20881 30604 18888 15564 29463 12111 14341 38808 22243 19068 18824 21315 19234 21953 36261 28100 17777 σcz 27 38 73 43 77 97 20 93 67 41 102 52 32 22 21 99 43 35 56 76 61 24 45 53 47 33 45 91 56 80 39 27 64 33 35 23 71 59 43 46 51 67 78 22 111 60 42 37 47 22 35 37 35 58 58 93 105 55 60 57 42 68 42 64 81 61 54 53 39 35 81 50 123 79 21 63 58 68 89 40 95 64 77 40 25 41 99 52 61 139 104 48 60 30 61 45 106 44 96 105 87 48 cdE

17

*

*

*

*

*

* *

*

* * * *

*

*

*

*

* *

*

* *

*

* * *

*

*

* *

*

*

* * *

* * *

*

*

*

*

*

*

*

Notes: the names of the galaxies not listed in the FCC are given in Table 7. The asterisks indicate cdE candidates from the FCC.

c 0000 RAS, MNRAS 000, 000–000

18

Drinkwater et al.

Table A2. (cont’d) Catalogue of all background galaxies with measured redshifts
FCC FCCB 1100 FCCB 1103 FCCB 1106 FCCB 1107 FCCB 1109 FCCB 1110 FCCB 1116 FCCB 1134 FCCB 1143 FCCB 1144 FCCB 1147 FCCB 1156 FCCB 1155 FCCB 1158 FCCB 1173 FCCB 1162 FCCB 1171 FCCB 1169 FCCB 1178 Name 3 FCCB 1186 FCCB 1193 FCCB 1195 FCCB 1192 FCCB 1200 Name 4 FCCB 1202 FCCB 1212 FCCB 1209 FCCB 1225 FCCB 1221 FCCB 1230 FCCB 1234 FCCB 1238 FCCB 1244 FCCB 1240 FCCB 1257 FCCB 1263 FCCB 1273 FCCB 1267 FCCB 1271 FCCB 1278 FCCB 1276 FCCB 1279 FCCB 1284 FCCB 1299 FCCB 1303 FCCB 1306 FCCB 1316 FCCB 1308 FCCB 1315 FCCB 1319 FCCB 1321 FCCB 1320 FCCB 1323 FCCB 1335 FCCB 1330 FCCB 1332 FCCB 1339 FCCB 1348 FCCB 1349 FCCB 1353 FCCB 1357 FCCB 1355 FCCB 1378 FCCB 1373 FCCB 1383 FCCB 1377 FCCB 1380 FCCB 1384 FCCB 1396 FCCB 1389 FCCB 1404 FCCB 1399 FCCB 1406 FCCB 1407 FCCB 1411 FCCB 1408 FCCB 1421 FCCB 1419 FCCB 1430 FCCB 1436 FCCB 1426 FCCB 1446 FCCB 1452 FCCB 1454 FCCB 1456 FCCB 1467 Name 5 FCCB 1487 FCCB 1498 FCCB 1523 FCCB 1520 FCCB 1528 FCCB 1538 FCCB 1532 FCCB 1533 FCCB 1541 FCCB 1548 FCCB 1562 FCCB 1577 FCCB 1594 RA (J2000) Dec 03:36:41.6 03:36:43.0 03:36:47.6 03:36:49.7 03:36:53.2 03:36:53.2 03:36:54.0 03:37:08.6 03:37:17.3 03:37:19.9 03:37:22.5 03:37:23.2 03:37:28.6 03:37:32.1 03:37:32.8 03:37:33.4 03:37:33.5 03:37:35.3 03:37:42.7 03:37:45.0 03:37:50.1 03:37:50.9 03:37:51.4 03:37:52.0 03:37:54.9 03:37:57.1 03:38:00.6 03:38:02.1 03:38:05.9 03:38:08.1 03:38:10.3 03:38:14.0 03:38:14.3 03:38:15.9 03:38:16.6 03:38:18.1 03:38:28.3 03:38:31.3 03:38:36.4 03:38:38.4 03:38:39.2 03:38:42.8 03:38:44.3 03:38:45.1 03:38:48.2 03:38:52.9 03:38:59.1 03:39:03.7 03:39:08.2 03:39:08.4 03:39:11.6 03:39:12.8 03:39:14.8 03:39:17.7 03:39:20.0 03:39:22.8 03:39:26.0 03:39:26.4 03:39:31.6 03:39:31.6 03:39:38.2 03:39:39.0 03:39:41.6 03:39:43.4 03:39:45.9 03:39:46.0 03:39:52.3 03:39:52.4 03:39:55.9 03:39:56.5 03:40:02.1 03:40:03.8 03:40:05.2 03:40:05.6 03:40:07.1 03:40:09.1 03:40:10.4 03:40:13.6 03:40:19.5 03:40:24.7 03:40:27.4 03:40:30.4 03:40:32.7 03:40:42.8 03:40:49.7 03:40:53.1 03:40:55.8 03:41:03.8 03:41:11.5 03:41:13.9 03:41:21.1 03:41:35.0 03:41:36.5 03:41:36.5 03:41:42.2 03:41:43.0 03:41:43.3 03:41:48.2 03:41:57.2 03:41:59.9 03:42:06.6 03:42:32.6 ?34:59:27 ?34:45:34 ?34:32:29 ?33:22:46 ?32:59:20 ?33:33:16 ?35:35:52 ?36:50:10 ?35:09:26 ?33:15:20 ?33:02:29 ?37:35:08 ?33:02:46 ?32:26:02 ?37:53:28 ?32:21:18 ?36:18:26 ?34:39:51 ?35:19:11 ?34:48:34 ?33:11:57 ?34:29:17 ?35:47:35 ?33:36:45 ?36:43:52 ?37:25:43 ?33:34:17 ?35:59:59 ?32:25:32 ?36:19:26 ?33:17:19 ?32:51:40 ?33:38:43 ?34:06:55 ?37:33:33 ?33:07:15 ?34:37:49 ?35:14:18 ?36:04:14 ?31:53:48 ?32:54:37 ?34:20:37 ?32:39:18 ?32:47:54 ?32:42:09 ?37:30:09 ?36:15:03 ?35:44:20 ?35:52:17 ?33:30:16 ?33:03:15 ?34:07:09 ?35:15:48 ?32:51:27 ?34:27:38 ?38:12:12 ?34:19:18 ?34:51:52 ?33:06:43 ?36:53:18 ?32:37:08 ?35:25:14 ?33:52:37 ?32:25:50 ?37:50:41 ?36:10:47 ?36:59:43 ?33:13:14 ?32:24:49 ?35:53:16 ?37:47:31 ?32:56:53 ?37:47:32 ?36:17:12 ?37:08:59 ?36:05:58 ?37:49:40 ?33:43:42 ?37:14:17 ?32:34:49 ?37:49:56 ?37:49:40 ?33:05:13 ?33:55:47 ?35:33:24 ?34:25:28 ?33:10:01 ?34:15:15 ?36:45:50 ?37:45:05 ?36:41:01 ?36:16:50 ?33:40:58 ?38:00:53 ?37:55:29 ?34:07:26 ?34:29:41 ?36:48:51 ?32:45:08 ?38:01:57 ?36:38:20 ?33:12:17 cz 19240 38389 19244 28146 35175 22083 20941 18833 18909 15765 11702 24491 11884 31599 33243 11813 25261 37055 36840 26232 11755 36748 33550 15747 34831 16923 15879 36380 35737 15477 32624 32302 15929 14480 13743 32595 16343 40600 18655 13263 32491 16016 39020 32170 39398 18980 34702 14838 25865 15698 32347 16002 18459 28276 21624 30910 37032 26789 32728 14546 38362 13850 16356 29330 25015 14693 42520 27581 63367 35941 13809 17794 13800 40193 24391 25652 14099 22104 14241 28579 12983 12855 21844 15986 14669 52838 42490 36937 31302 34196 15170 6794 21744 13664 13681 12161 37316 8377 11534 13540 15426 19474 σcz 36 65 37 51 114 25 76 28 39 43 28 34 36 74 59 50 69 56 42 133 68 115 63 72 157 31 51 42 123 89 58 64 19 40 38 64 53 52 48 80 60 33 57 77 63 23 79 47 44 58 97 50 64 57 111 53 64 71 83 106 47 37 34 58 66 64 69 33 142 42 55 55 49 45 45 70 47 48 82 47 44 39 59 39 87 41 97 74 36 59 28 102 50 45 38 62 43 69 64 38 57 57 cdE FCC FCCB 1611 FCCB 1618 FCCB 1619 FCCB 1621 FCCB 1632 FCCB 1626 FCCB 1631 FCCB 1635 FCCB 1643 FCCB 1648 FCCB 1654 FCCB 1660 FCCB 1663 FCCB 1668 FCCB 1666 FCCB 1676 FCCB 1680 FCCB 1684 FCCB 1681 FCCB 1686 FCCB 1682 FCCB 1700 FCCB 1695 FCCB 1707 FCCB 1714 FCCB 1711 FCCB 1727 FCCB 1739 FCCB 1749 FCCB 1754 FCCB 1762 FCCB 1769 FCCB 1767 FCCB 1784 FCCB 1797 FCCB 1799 FCCB 1798 FCCB 1810 FCCB 1814 FCCB 1816 FCCB 1848 FCCB 1854 FCCB 1857 FCC 311 FCCB 1892 FCCB 1898 FCCB 1905 FCCB 1903 FCCB 1921 FCCB 1919 FCCB 1911 FCCB 1932 FCCB 1939 FCCB 1941 FCCB 1940 FCCB 1942 FCCB 1946 FCCB 1951 FCCB 1974 FCCB 1986 FCCB 1983 FCCB 1984 FCCB 1987 FCCB 1988 FCCB 1993 FCCB 2007 FCCB 2005 FCCB 2020 FCCB 2032 FCCB 2037 FCCB 2050 FCCB 2047 FCCB 2051 FCCB 2058 FCCB 2064 FCCB 2063 FCCB 2067 FCCB 2087 FCCB 2080 FCCB 2092 FCCB 2098 FCCB 2096 FCCB 2111 FCCB 2138 FCCB 2174 FCCB 2180 FCCB 2203 FCCB 2197 FCCB 2199 FCCB 2225 FCCB 2230 FCCB 2234 FCCB 2235 Name 6 FCCB 2239 FCCB 2248 FCCB 2251 FCCB 2263 FCCB 2276 Name 7 FCCB 2289 FCCB 2290 RA (J2000) Dec 03:42:43.4 03:42:50.0 03:42:53.6 03:42:54.1 03:42:54.5 03:42:54.6 03:42:54.6 03:43:07.1 03:43:07.9 03:43:08.3 03:43:16.3 03:43:21.0 03:43:23.1 03:43:29.0 03:43:29.6 03:43:31.6 03:43:33.5 03:43:34.1 03:43:35.0 03:43:39.8 03:43:40.2 03:43:46.1 03:43:47.4 03:43:57.9 03:44:00.6 03:44:04.5 03:44:12.0 03:44:21.0 03:44:29.0 03:44:30.1 03:44:41.6 03:44:44.9 03:44:45.7 03:44:51.5 03:45:00.6 03:45:01.8 03:45:04.9 03:45:17.4 03:45:23.2 03:45:26.6 03:45:50.1 03:45:59.5 03:45:59.9 03:46:18.2 03:46:29.1 03:46:30.6 03:46:32.1 03:46:34.6 03:46:41.0 03:46:41.5 03:46:43.8 03:46:49.3 03:46:56.0 03:46:56.5 03:46:56.8 03:46:58.2 03:47:01.4 03:47:09.2 03:47:18.5 03:47:35.8 03:47:37.6 03:47:38.0 03:47:40.0 03:47:42.6 03:47:44.2 03:47:50.8 03:47:52.4 03:48:06.1 03:48:15.5 03:48:16.4 03:48:28.9 03:48:29.2 03:48:30.8 03:48:34.9 03:48:39.3 03:48:39.5 03:48:40.9 03:48:56.5 03:48:57.2 03:49:02.2 03:49:02.6 03:49:03.5 03:49:11.9 03:49:43.7 03:50:22.2 03:50:29.4 03:50:44.4 03:50:45.1 03:50:49.2 03:51:02.0 03:51:11.9 03:51:14.1 03:51:15.3 03:51:17.1 03:51:21.0 03:51:22.3 03:51:24.9 03:51:37.4 03:51:50.1 03:51:59.5 03:52:13.2 03:52:16.4 ?36:16:08 ?35:07:57 ?34:11:60 ?34:47:51 ?37:28:30 ?36:17:56 ?37:10:57 ?32:41:35 ?34:23:52 ?37:07:36 ?33:37:54 ?36:18:18 ?35:12:24 ?33:41:04 ?33:27:01 ?36:39:06 ?36:18:28 ?37:17:03 ?35:48:02 ?34:40:06 ?32:37:25 ?37:37:43 ?33:54:49 ?36:12:20 ?35:35:28 ?32:41:14 ?37:07:34 ?35:31:56 ?36:56:09 ?38:16:49 ?35:41:31 ?36:38:41 ?34:55:26 ?38:34:32 ?37:36:25 ?38:22:57 ?35:30:18 ?32:12:48 ?35:45:57 ?34:15:45 ?38:08:35 ?34:34:06 ?35:32:14 ?33:45:48 ?34:24:21 ?34:22:46 ?36:53:41 ?34:23:01 ?36:56:43 ?35:51:35 ?32:48:44 ?36:26:45 ?35:14:15 ?35:46:17 ?35:15:03 ?34:47:59 ?34:26:27 ?32:49:22 ?38:22:47 ?37:29:31 ?34:16:55 ?34:21:26 ?35:25:03 ?35:06:24 ?36:08:47 ?37:38:31 ?36:06:60 ?33:50:02 ?34:30:14 ?36:05:25 ?33:19:44 ?32:05:20 ?32:14:21 ?33:43:55 ?34:02:48 ?33:44:13 ?35:54:45 ?35:50:59 ?32:37:17 ?35:20:57 ?36:26:08 ?35:36:21 ?36:33:40 ?36:22:20 ?33:56:53 ?34:15:13 ?37:22:32 ?35:56:27 ?33:33:59 ?36:01:39 ?33:50:10 ?34:25:27 ?33:49:02 ?35:45:49 ?32:46:00 ?37:07:40 ?37:20:22 ?34:54:11 ?37:36:25 ?36:14:20 ?33:25:09 ?32:07:22 cz 23108 18469 16161 19245 6411 23392 15099 38715 38289 26513 18228 12150 23378 27643 22718 14054 25428 34346 30490 12759 19578 13214 52379 25083 18250 49995 18379 34769 13145 25287 33718 22825 35093 6578 15278 14684 35294 12111 30418 37560 13437 18268 33119 47658 16285 16130 18101 15664 18258 23558 35132 21247 37814 25791 37596 27823 15993 27361 16325 23404 19188 16381 35385 15255 23677 14153 23180 33124 15806 30270 35242 21665 21731 16005 49439 16193 18764 23125 22075 12181 34292 18103 5679 34835 7096 38068 23150 23107 27597 18100 19380 16313 19285 18102 22340 22473 16669 18466 19162 14505 22390 10083 σcz 99 74 38 71 45 98 23 77 48 57 70 57 102 56 112 61 78 68 97 28 18 45 95 59 69 75 87 69 38 55 48 37 88 38 37 38 81 73 63 38 51 75 55 60 31 73 105 61 47 71 68 116 153 88 74 88 112 65 134 82 38 134 55 30 64 69 36 77 48 118 57 62 62 12 47 27 33 104 53 31 152 46 35 63 47 135 86 32 127 22 77 43 43 51 47 42 32 35 39 34 181 62 cdE

* * *

*

*

*

*

*

*

*

*

* *

* *

*

*

*

*

*

*

*

* *

*

*

* *

*

* *

Notes: the names of the galaxies not listed in the FCC are given in Table 7. The asterisks indicate cdE candidates from the FCC.

c 0000 RAS, MNRAS 000, 000–000

Dwarf galaxies in Fornax

19

Figure A1. Distribution of the observed galaxies.

c 0000 RAS, MNRAS 000, 000–000


赞助商链接

更多相关文章:
更多相关标签:

All rights reserved Powered by 甜梦文库 9512.net

copyright ©right 2010-2021。
甜梦文库内容来自网络,如有侵犯请联系客服。zhit325@126.com|网站地图