Introduction

Compact groups of galaxies are small, relatively isolated systems of galaxies with the projected separations between the member galaxies comparable to the diameters of these galaxies (Rose 1977; Hickson 1982). These groups belong to the densest systems of galaxies and they may form an interesting link in the merging process which runs from galaxies in loose groups to the resulting merged systems.

The sample used in most recent studies of compact groups is that defined by Hickson (1982). This sample was selected by visually inspecting Palomar Sky plates, but using quantitative selection criteria. This sample contains about 100 groups, distributed over two-thirds of the sky. The compact groups of this sample typically consists of 4-5 galaxies, and the median separation between galaxies is 40 kpc. About 1% of the visible mass in the universe is estimated to be in such groups. The typical three-dimensional velocity dispersion of a compact group is about 300 km, implying that they contain dark matter ().

Extensive work, both theoretical and observational, has been done on compact groups in the last 15 years, but there is still controversy and debate over their true nature. The crossing time of a typical compact group is only of order 1% of a Hubble time. This means that the galaxies in these groups must merge on a short timescale. In principle, this would fit within the framework of hierarchical formation of structure in the universe: compact groups merge to form elliptical galaxies, while they are replenished by loose groups of galaxies that become more tightly bound. However, it still has to be demonstrated that the required rates follow from realistic initial conditions. The formation rates of elliptical galaxies and compact groups depend on cosmological parameters (e.g. ), and some fine-tuning may be required to reproduce the observed number of compact groups (Ramella et al. 1994; Hernquist et al. 1995).

However, the observational evidence for these high merging rates in compact groups is not very strong. Interactions between galaxies in compact groups occur, but the observed merging rates appear to be smaller than predicted (e.g. Zepf 1993). Several investigators have suggested that a high fraction of the groups in the Hickson sample are not true compact systems, but instead are chance alignments of either loose groups (e.g. Rose 1979; Mamon 1986) or as the result of a filamentary distribution of galaxies (Hernquist et al. 1995).

Considering the implications for our understanding of the formation of structure in the universe, it is important to know what fraction of the groups of galaxies that appear compact on the sky are indeed physically dense systems and how these compact groups evolve.

© Copyright Astronomical Society of Australia 1997