|Annu. Rev. Astron. Astrophys. 1988. 26:
Copyright © 1988 by . All rights reserved
7.2. Changes in Morphological Mix
7.2.1. THE BUTCHER-OEMLER EFFECT A new direction for tests of evolution in the look-back time was set out by Butcher & Oemler (1978a, b, 1984a, b) in their studies of the fraction of blue galaxies to the total population of clusters as a function of redshift. They maintain that there is evidence for secular morphological evolution in the cores of clusters because the blue to red ratio is a progressive function of redshift, starting at the very low redshift of z 0.15! A large literature has grown covering the effect, coming down on both sides of the issue. The criticisms that seemed most serious to Butcher & Oemler were made by Mathieu & Spinrad (1981) and Dressler & Gunn (1982, 1983) concerning cluster membership, and by Wirth & Gallagher (1980) concerning the proper normalization of the blue-red ratio in nearby clusters. On the other side, confirmation support was given by Couch & Newell (1984), who expanded the sample. These authors also answered a criticism concerning a color bias effect with redshift caused by different K corrections for different morphological types (Section 4), a problem raised by DeGioia-Eastwood & Grasdalen (1980). Furthermore, the effect appears not to be present in every cluster; a notable example is the very compact cluster 0016+16 found by Kron (z = 0.54), as discussed by Koo (1981). A comprehensive discussion of various selection effects that cloud the issue of the reality of the effect is given by Koo (1988a).
7.2.2. THE DRESSLER-GUNN EFFECT In a spectroscopic study of the Butcher-Oemler effect, Dressler & Gunn (1982, 1983) discovered that the percentage of emission-line galaxies in distant clusters was higher than in control fields at low redshift. A literature has also grown on this subject both for cluster galaxies and now, with the great recent emphasis, for the field [see Ellis (1988) for a review]. Dressler & Gunn's discovery has been interpreted (Dressler & Gunn 1982, 1983, Dressler 1984) as evidence for recent (last ~ 2 × 109 yr) star formation in otherwise "dead" E galaxies. This active evolution is different from the slow passive evolution of old stellar systems mentioned before. If the abnormal activity occurs in otherwise dead E galaxies, bursts of star formation must be postulated at the relevant look-back time.
But a word of caution is in order. No morphological information on galaxy types is yet available for the galaxies with active spectra found by Dressler & Gunn. As pointed out by Koo (1988a), Osterbrock's (1984) bias caution may hold the key to explaining the Dressler-Gunn statistics as a selection effect giving a false impression of an emission-line excess in the field. Seyfert galaxies are among the brightest and bluest galaxies in any volume of space. Hence, in any surveys that are flux limited, they will dominate the statistics in numbers appearing to increase in percentage over intrinsically fainter galaxies at faint apparent magnitudes. It remains a problem to determine if this selection bias can affect Dressler and Gunn's conclusion concerning the unexpectedly strong spectral activity as a function of redshift for field galaxies.
However, much additional work supports the reality of the Dressler-Gunn interpretation of recent starburst activity in some, but not all, otherwise dead ellipticals in clusters with z > 0.3. Ellis (1988) reviews the situation as it appeared in mid-1987. Observations of ~ 100 galaxies in Abell cluster 370 (z = 0.37) by McLaren et al. (1988) suggest that UV light from young blue stars has been added recently to only 15% of the E galaxies [the remaining 85% remaining normal, reminiscent of Hamilton's (1985) result]; but in this 15% this is a phenomenon of starbursting that is unknown in local E galaxies observed so far with the IUE satellite, which again suggests evolution in the look-back time.