Annu. Rev. Astron. Astrophys. 1979. 17: 135-87 Copyright © 1979 by Annual Reviews. All rights reserved

7.4 Conclusion

For the present, we conclude that the available data continue to support the existence of high cluster mass-to-light ratios, with M / L_B 325. This value at first sight looks significantly larger than the values of 30-90 which we determined for binaries and small groups. However, part of the discrepancy is due to stellar population differences between the early-type galaxies in clusters and the spirals in small groups. The theoretical and observational data in Table 2 suggest that M / L_B for the stellar population in spirals is only half that in ellipticals and S0's. So, to properly compare small groups with great clusters, we must increase M / L_B for small groups from 30-90 to 60-180. Thus roughly 20% to 50% of the total mass can plausibly be associated with galaxies.

A further small correction must be made for ionized gas. Conventional X-ray measurements of cluster cores suggest that the mass of ionized gas is probably ~ 10% of the virial mass (Lea et al. 1973, Field 1974, Gull & Northover 1975, Malina et al. 1978), in agreement with the requirements imposed by the detection of microwave diminution by hot gas in clusters (Birkinshaw et al. 1978). Adding 10% to our previous total, we can account for roughly 30% to 60% of the virial mass, leaving a net discrepancy of approximately a factor of two. Although this difference is uncomfortably large, real difficulties still remain in the determination of cluster M / L ratios (e.g. the luminosity function). The reality of excess unseen mass in great clusters relative to small groups must therefore still be considered uncertain at the present time.

We note in passing that a controversial detection of large X-ray halos around clusters has just been reported by Forman et al. (1978). While the gas in these halos potentially could bind the cluster as a whole, the halo has little dynamical influence on the core and thus cannot ease the M / L problem in the central regions.

White (1976a, 1977) has shown through N-body models that the dark material in clusters cannot all be attached to individual galaxies, or else a marked degree of radial mass segregation should be observed owing to dynamical friction. Since little segregation is apparent in real clusters, the hidden matter must be distributed rather uniformly throughout the cluster as a whole. Hence, galaxies in clusters cannot have large massive envelopes still attached. Gallagher & Ostriker (1972) and Richstone (1975, 1976) have suggested that high-velocity encounters between galaxies might liberate these envelopes through tidal shocks, thus spreading the dark matter throughout the cluster. The exact interrelationship between the competing processes of tidal disruption and dynamical friction remains to be worked out.