Next Contents Previous

5. THE DISTRIBUTION OF MASS FOR GALAXIES IN CLUSTERS

5.1 Velocity Dispersions for Elliptical Galaxies in Clusters

The velocity dispersion profiles for most elliptical galaxies are flat or falling. However, Dressler (1979) and Carter et al. (1985) have observed rising rotation curves in the outer regions of several cD galaxies. This provides evidence that the cluster environment can affect the distribution of mass for galaxies in the centers of clusters, and more specifically, provides support for the cannibalism or tidal accretion theory for cD formation. Dressler (1979) argues that three components are required to fit his data; an inner region similar to normal elliptical galaxies (M/L approx 10), an intermediate region that might represent accreted cluster members (M/L approx 35), and an extended cluster-filling halo (M/L approx 500). Although this would agree with several of the arguments presented in Section 4.1, more extensive observational data is clearly needed to make the case based on velocity dispersion gradients alone. Observations of the globular cluster system (Mould, Oke, and Nemec 1987) and the X-ray halo around M87 (Fabricant and Gorenstein 1983) also provide evidence that the mass around central cluster galaxies can grow to prodigious amounts.

On a smaller scale, Dressler (1984b, 1987) and Dressler et al. (1987) found that the Faber-Jackson and the Dn-sigma relations for both ellipticals and spiral bulges in clusters are similar to their counterpart relations in the field. There is no apparent difference in the Faber-Jackson relation for cD galaxies once light from the outer envelope is removed from the comparison (Malumuth and Kirshner 1981). Even severely interacting galaxies have a similar Faber-Jackson relation (Lake and Dressler 1986). It appears that the strongly bound stars near the centers of galaxies are not unduly affected by the environment. However, Djorgovski, de Carvalho, and Han (1989) find tentative evidence that both the Dn-sigma and Tully-Fisher relations may be affected by the cluster environment at the few percent level, and caution that the assumption that these relations are universal may lead to systematic errors in recent observations of large-scale motions.

These studies all use measurements of the central velocity dispersion where environmental effects may be minimized. The question of what is happening to the velocity dispersion gradients in the outer regions of normal cluster ellipticals has not yet been addressed.

Next Contents Previous