Annu. Rev. Astron. Astrophys. 1991. 29: 499-541 Copyright © 1991 by Annual Reviews. All rights reserved

3.6 Observations of Clusters

Clusters of galaxies are the largest structures in the universe that unambiguously appear dynamically bound. They can thus provide estimates of the mass distribution over scales on the order of several Mpc. The density perturbations that give rise to the formation of clusters and allow them to decouple from the Hubble expansion have very long growth times, and most clusters have not yet attained the degree of dynamical equilibrium that is accompanied by a smooth, spheroidal appearance. Thus, the understanding of their dynamical state necessitates having rich kinematical information in order to determine cluster morphology and membership, and eventually to disentangle substructures from one another and yet obtain statistically sound estimates of the characteristics of each. Although important contributions were made through the years, especially in the study of rich clusters like Coma, until the relatively recent work of Kent & Gunn (1982) for the Coma cluster and Kent & Sargent (1983) for the Perseus cluster, no truly detailed study was available. Several other clusters have since been studied in comparable detail, i.e. with samples that include a few hundred redshifts per cluster, among them the nearby systems Virgo, Centaurus, and Hydra I = A1060 and (thanks to multiobject spectrographs) several more distant clusters.

COMA, PERSEUS, AND VIRGO Kent & Gunn (1982) assembled a data base of about 300 radial velocities in Coma, complete to m_cgcg = 15.7 within 3° of the cluster center, and to m_cgcg = 15.0 within 6°. Their discussion of distribution models for the galaxies and for the mass indicated a core radius of 170-200 h^-1 kpc (8.5'-10'), a total mass within the inner 3.5 h^-1 Mpc (~ 3°) of 1.5 x 10¹⁵ h^-1 M, a mass to light ratio M / L_B ~ 360 h, and that the matter distribution was not significantly different from that of the galaxies. Hughes (1989) critically analyzed their models and those presented in more recent work. This analysis confirmed the results of Kent & Gunn (1982).

In spite of the low galactic latitude and high extinction of the Perseus cluster - which exceeds half a magnitude at the cluster center - 187 redshifts of cluster members were available to Kent & Sargent (1983). At 5470 km s^-1, Perseus has the highest velocity dispersion among nearby clusters (1300 km s^-1). Its global parameters are comparable to those of Coma: a core radius of 170 h^-1 kpc (11'), a virial mass of 1.7 x 10¹⁵ h^-1 M, and a mass to light ratio M / L_V ~ 600 h.

In the case of Virgo, a large redshift data base has currently been accumulated. The major recent contributions have been made by Karachentsev & Karachentseva (1982), Huchra (1985), and Hoffman et al (1987, 1989). Huchra compiled existing data and completed a sample of 471 galaxies within a 6° radius from M87 to a limiting magnitude of m_cgcg = 15.5. Binggeli et al (1985) produced a catalog of 2096 galaxies in an area centered on the Virgo cluster, based on 2.5-m duPont telescope plates. Using this catalog, Hoffman and coworkers (1989 and refs. therein) observed (at 21-cm at Arecibo) all dwarf irregular galaxies brighter than b_T = 17.0 - a sample of nearly 300 objects. H I observations are also available of all 100 or so bright spirals deemed to be cluster members. Overall, redshifts are known for over 400 Virgo cluster members, which constitute the highest level of detail and completeness recorded for any cluster. Binggeli et al (1987) have written a comprehensive morphological and kinematical study of this cluster.

OTHER CLUSTERS A detailed study of the Centaurus cluster (based on about 180 cluster members) was presented in a series of three important papers by Lucey et al (1986 and refs. therein). A2670 has been thoroughly studied by Sharples et al (1988), on the basis of a sample of 220 cluster members. Many other clusters have been studied in some detail data sets typically include a few dozen members per cluster. Notable among them are: the results of Fitchett & Merritt (1988) for Hydra I (A1060, based on redshift data of 95 cluster members as given by Richter 1987), the work by Fabricant et al (1989) on A2256 (86 cluster redshifts), those by Chapman et al (1988) on A194 (74 cluster members), Ostriker et al (1988) on A539 (86 members), Cristiani et al (1987) on Klemola 22, Soucail et al (1988) on A370, Metcalfe et al (1987) on Shapley 8, Dixon et al (1989) on A2197 and A2199 (based on redshift data of Gregory & Thompson 1984, 1986), and (Bothun et al 1983) on Cancer. Other contributions of considerable interest, albeit of somewhat less wealth in the data base, are available for the clusters A1146 (Melnick & Quintana 1985), A1142 (Geller et al 1984), the Eridanus cluster (Willmer et al 1989), A744 (Kurtz et al 1985), A1099 and A1016 (Chapman et al 1987), AC103 (Sharples et al 1985), A262 (Giovanelli et al 1982), A347 and A1367 (Moss et al 1988), NGC 5846 (Haynes & Giovanelli 1990b), and Klemola 27 (Richter 1984). Significant samples including many clusters have been collected by Vettolani et al (1990), Proust et al (1987, 1988), Rhee & Katgert (1988), Green et al (1988), and Owen et al (1988). Struble & Rood (1987) and Andernach (1990) have prepared recent compilations of redshifts and velocity dispersions for Abell clusters.

Two recent contributions stand out for their statistical wealth. To complement Dressler's (1980) morphological study, Dressler & Schectman (1988a, b) have measured 1268 redshifts in 15 clusters. Most recently, Zabludoff et al (1990) presented an analysis of a compilation of 3250 redshifts for galaxies in 69 nearby Abell clusters (including 359 new redshifts).

MULTIPLEXING The high concentration of bright galaxies found in cluster cores is ideally suited for the opportunities offered by multislit and multifiber devices. Thus, Colless & Hewett (1987) measured 604 radial velocities in 14 rich southern clusters - ranging in redshift from 14,000 to 44,000 km s^-1. This approach is representative of modern cluster surveys, such as those of Teague et al (1990), Mazure et al (1989), Batuski and coworkers (personal communication), and Guzzo et al (1990). Teague et al (1990) obtained 1034 redshifts (805 cluster members) in a sample of 10 southern clusters, with samples of more than 100 in 3 of them. Guzzo et al (1990) are completing a redshift survey of 150 clusters extracted from the Edinburgh/Durham Southern Galaxy Catalogue (Heydon-Dumbleton et al 1989), carried out with the EFOSC spectrograph in multislit mode at ESO and the Autofib multifiber system at the AAT. This effort should result in a sample of over 2000 redshifts. Such a rapidly growing cluster data base will not only lead to an improved understanding of cluster structure, dynamics, and evolution, but also to a much sounder determination of the clustering properties of the universe at the largest scalelengths.