7. SUMMARY AND CONCLUSIONS

The basic results of the recent efforts concerning observations of gas deficiency in cluster spirals can be summarized as follows:

1. Spiral galaxies that pass within the central regions of rich clusters with moderate X-ray luminosities may lose as much as ninety percent of their interstellar HI gas.

2. The HI is preferentially removed from the outer portions of the disk so that the deficient galaxies have shrunken HI disks. The inner disks are only moderately HI deficient. The stripping mechanism displaces the affected galaxies along tracks of nearly constant M_HI / D²_HI.

3. The molecular component of the ISM is relatively unaffected both in terms of its content and its distribution. Galaxies that are HI-poor by a factor of ten may be gas poor by only a factor of two or three.

4. The presence of spiral structure and normal CO content argue for a recent timescale, but most highly HI deficient galaxies show reddened colors indicating that disk fading has already set it. The shift in colors can be explained by a reduction in the star formation rate by a factor of two about 10⁹ years ago.

5. The star formation rate can also be enhanced by the stripping process. Individual objects may show signs of starburst or Seyfert activity over short time scales. Star formation without replenishment is a further cause of eventual gas depletion.

6. Galaxy-ICM interactions are the most likely cause of gas removal although the efficiency of competing mechanisms - ram pressure sweeping, evaporation or turbulent viscosity - is not yet well evaluated. Among clusters, the fraction of galaxies observed to be deficient is greatest where the X-ray luminosity is highest, but there is no evident dependence on , V² or galaxy mass.

7. Early-type spirals are more HI poor relative to their counterparts in lower density regions than are late-type spirals. There are suggestions that this dichotomy may result from fundamental differences in the orbital characteristics of galaxies along the spiral sequence.

8. The stripping mechanism is effective only within the small volume occupied by the cluster core and is unlikely to be responsible for the morphological segregation seen on supercluster scales.

After 15 years of effort, the cluster samples are significant but not terribly large, and statistical studies are still plagued by small numbers, especially when one is trying to investigate individual variables for which subsamples must be compared. It is also important to keep in mind that specific objects - whether single galaxies within a cluster or individual clusters - may not be representative of the universe at large. There are several additional points that best illustrate our uncertainties:

9. No single physical gas sweeping process can easily reproduce all of the observations.

10. In fact, all responsible processes (e.g., those under the broad category of galaxy-ICM interactions) might show the observed dependence of the degree of HI deficiency on distance from the cluster center.

11. Many galaxies remain unaltered over their lifetime, although morphological segregation is observed over supercluster scales.

12. The direct relationship between atomic hydrogen, molecular hydrogen and star formation is unclear. It is not clear how the removal of over ninety percent of a galaxy's atomic gas, the long term potential for future star formation, may effect the inner molecular component, and whether the general heating of the interstellar medium during the sweeping process may alter the mass spectrum of molecular clouds. In interpreting the CO data for Virgo objects, one must recognize the uncertainty in applying a universal conversion from CO to H₂.

13. It appears that early-type galaxies are even more depleted that later-type spirals so that the evolution even within a class is dependent to some extent on the local environment.

The future offers us a number of opportunities to follow the study of gas deficiency. The study of the HI content of cluster spirals and especially of early type objects will be greatly enhanced by the increases in sensitivity gained by the Arecibo gregorian feed upgrade and by the construction of the new Green Bank telescope. Aperture synthesis observations of both HI and CO in additional galaxies in Virgo, in Hydra and in other nearby clusters are vital to our understanding both of the sweeping mechanism and its effect on the conversion of gas to stars in galaxies. Further clues to the star formation process will be sought through careful studies of indicators such as H emission with the spatial dimension included. Possible environmental influences on the dark matter distribution are critical both to our understanding of galaxy formation and to application of the Tully-Fisher relation to obtain the distance scale.

The Hubble Space Telescope will contribute enormously to our ability to study galaxies in clusters at higher redshifts when we know galaxies were not all like the objects we see in clusters today. Of particular relevance will be the continued study of the blue cluster objects seen at z > 0.3. Are they spirals falling into the core and suffering HI depletion both by induced star formation and sweeping?

Based on the indirect evidence of the observed HI deficiency in cluster spirals, it now seems well established that selected spirals that pass through the cores of rich clusters lose significant portions of their cool interstellar gas. We are led to return to the long-standing debate over whether stripped spirals are responsible for the S0 class. Based on the usual arguments about the occurrence of S0's in the field and the fundamental differences in the dominance of disk and bulge components (Dressler 1980), it does not seem likely that all S0's are stripped spirals. The loss of nearly all of the HI gas, despite the retention of the molecular component, must affect the galaxy's future evolutionary path. It still remains difficult to see how one could turn an Sc into an S0, but the possibility that the early type spirals may preferentially evolve towards the S0 class because they follow radial orbits (Dressler 1986) is intriguing. de Freitas et al. (1985) have noted already the tendency for cluster S0's to have flatter axial ratios than field ones, implying a contribution to the S0 population of stripped spirals. In comparing the morphology-density relation in clusters with high and low X-ray luminosity separately, GH85 have noted a decrease in the population of spirals and a corresponding increase in the population of S0's, for the same galaxy density, in the clusters with high X-ray luminosity. While we can recognize candidates for stripping and plausible galaxy-ICM interactions that could result in adequate gas removal, the same stripping mechanism(s) cannot be responsible for the S0's seen in less dense regions. Hence, we conclude that there are effective mechanisms for environmentally-driven galaxy evolution in operation in cluster cores containing a hot, healthy ICM, but the regimes of density within which such mechanisms could be of significance in enhancing the morphological segregation represent only a small fraction of the volume of the universe.

I thank R. Giovanelli, T. L. Herter and M. S. Roberts for many discussions on the continuing questions about stripping. My talk was greatly aided by the contribution of unpublished data by L. Cayatte, C. Balkowski and J. van Gorkom and V. Rubin. This work has been supported in part by NASA-JPL contract no. 957289. The study of Sa and Sc galaxies has been conducted in collaboration with C. Magri as part of his dissertation research at Cornell University.