Annu. Rev. Astron. Astrophys. 1988. 26: 509-560 Copyright © 1988 by . All rights reserved

6.1. The Morphology-Density Relation

Following earlier work discussed in Section 1 (cf. also Hubble & Humason 1931, Morgan & Lesh 1965, Abell 1965, Oemler 1974, Melnick & Sargent 1977) Dressler (1980) discovered that the (projected) local density of galaxies governs the mixture of Hubble types in any local environment of a cluster, independent of the cluster's global environment. Subsequent work by Bhavsar (1981), de Souza et al. (1982), and Postman & Geller (1984) has extended Dressler's relation to groups and the general field.

But the Dressler effect holds not only for the main Hubble types but also for spiral subclasses as well, as convincingly shown by Giovanelli et al. (1986) in their study of the Pisces-Perseus supercluster and anticipated by Gisler (1980). [Barred spirals, strangely, may be an exception (Kumai et al. 1986).] A strong morphology-density relation has also been found for the dwarf galaxies (dE vs. Im) in the Virgo cluster (Binggeli et al. 1988). The whole Hubble sequence seems affected, down to very fine morphological detail.

As the shapes of the LFs for the different types vary little (or, by our working hypothesis, not at all) for the two different environments, which now represent merely two different mean densities (low and high) in the continuous range of D, one can estimate the relative frequency of a given type by reading the amplitude at the relevant magnitude M_BT. It is clearly visible from Figure 1 not only how strongly the early types (E, S0, dE) are suppressed in the field, but also how much the late types (Sa + Sb, Sc, Sd + Sm, Irr) are boosted - the later the type, the larger the boost. (Note the logarithmic scale of Figure 1, which conceals the importance of the effect.)