![]() | Annu. Rev. Astron. Astrophys. 1982. 20:
431-468 Copyright © 1982 by Annual Reviews. All rights reserved |
The structure and mass of an active galaxy are often correlated with the degree of its activity. Since galactic properties such as these are generally believed to be nearly permanent features of a galaxy, we now investigate what role they may play in causing, or at least facilitating, activity. We must note, however, the comments recently made by Simkin et al. (1980): Active galaxies may be undergoing major and perhaps repeated morphological changes, some of which are transitory activity. The discussion here (for obvious reasons) emphasizes active galaxies that are not interacting strongly with other galaxies.
RADIO SOURCES Conventional astronomical folklore is replete with stories about the relationship between morphology and activity in galaxies. For example, it is widely accepted that powerful extragalactic radio sources are associated exclusively with elliptical galaxies. However, the earliest comprehensive optical morphological survey of extragalactic radio sources (Mathews et al. 1964 - thereafter MMS) reached different conclusions. They found that there was a good correlation between the radio luminosity of a source and its optical identification; in order of increasing radio luminosity, the typical identification was spiral, E, D (or related cD or db classes), N galaxies, and QSOs. Quantitative studies, e.g. surface photometry, strongly confirm that global differences between ellipticals and cDs exist (Hoessel 1981, Hoessel et al. 1981). However, it has never been confirmed that the radio galaxies classified as D by MMS are significantly different photometrically from normal E galaxies, although Simkin (1979) has presented kinematic evidence, as we discuss in Section 3.2.
In any event there are several examples of extended, luminous radio sources associated with galaxies that are clearly not simple ellipticals. Some of these galaxies are classified peculiar ellipticals (e.g. Cen A, Cyg A, and Fornax A; see MMS). and others have been classified as disk systems, e.g. NGC 612 (Ekers et al. 1978), 3C 293 (Sandage 1966), NGC 6240 (Fosbury & Wall 1979), 3C 305 (Sandage 1966, Heckman et al. 1981a). To these could perhaps be added 3C 120 (Arp 1975, 1981, Heckman & Balick 1979) and PKS 0400-181 (Shaver 1981).
The radio properties of Seyfert galaxies (typically early-type spirals; see the next section) are germane. Seyferts have radio luminosities intermediate between radio sources in similar non-Seyfert spirals and classical radio galaxies (e.g. de Bruyn & Wilson 1976, 1978). Many of them exhibit a strong morphological resemblance to radio galaxies but in miniature (size scales are smaller by ~ 102; Wilson & Willis 1980, Wilson et al. 1980, Wilson & Ulvestad 1981, Ulvestad et al. 1981). The frequent appearance of colinear double- or triple-radio structures suggests that the nuclei of disk galaxies sometimes have collimated ejecta just like the brighter radio galaxies (although perhaps at a lower power level) and/or that the relatively dense extranuclear gas in the disk inhibits the full development of the radio source. The radio axis is uncorrelated with any isophotal axis of the stars (see summary by Ulvestad et al. 1981). On the other hand, many other radio Seyferts bear less resemblance to radio galaxies than to "normal" spirals (e.g. Condon et al. 1981) in terms of radio properties, and the radio emission in these objects could be due to "starbursts" rather than nuclear activity. The morphology and power of the radio sources in Seyferts do not appear to be correlated.
The radio properties of nearby "normal" galaxies clearly depend on Hubble type. Proceeding from late-type to early-type disk galaxies, the nuclear radio source becomes stronger, morphologically simpler, and more compact (e.g. Hummel 1980, van der Hulst et al. 1981, Condon et al. 1982). Ultracompact flat-radio-spectrum sources are seen almost exclusively in early-type galaxies (Ekers 1978a, b, c, Condon & Dressel 1978, Heckman 1980b). Hummel (1980) and Kotanyi (1981) find elliptical galaxies (especially round ones) to be more radio-loud than SOs. However, Dressel (1981) could not confirm these last results.
In summary, radio power appears to increase with the prominence of the bulge component in galaxies, and radio sources of greater-than-galaxy size seldom appear in galaxies with prominent stellar disks. Apparently some features of systems with disks thwart the development of a large, powerful radio source, independent of the power of the nuclear machine.
SEYFERT GALAXIES The propensity of Seyfert galaxies for early-type spirals has long been known (e.g. Adams 1977), a result which was quantified by Heckman (1978) and from a homogeneous data sample, by Simkin et al. (1980). Here, of course, we ignore the Seyferts with highly disturbed or peculiar morphology.
The absence of Seyfert activity in Sc or later-type galaxies is in accord with conclusions regarding the incidence of radio emission discussed in the previous section. The apparent rarity of Seyfert activity in ellipticals is perhaps more surprising. However, emission-line nuclei (ELNs), as a rule of thumb, are known to be deficient in rich clusters where elliptical galaxies are the most abundant (Section 2.1). Thus the absence of Seyferts in ellipticals may be in part an indirect effect of the cluster environment. Also, there are many classical radio galaxies (ellipticals or related N and D galaxies) whose emission-line spectra are essentially the same as Seyfert nuclei (Grandi & Osterbrock 1978). Finally, 10% of the known Seyfert nuclei lie in galaxies that could perhaps be classed as ellipticals - Mrk 6, Mrk 50, Mrk 78, Mrk 290, Mrk 298, and Mrk 509 (Adams 1977), NGC 2110 (Bradt et al. 1978), and IC 5063 (Caldwell & Phillips 1981, but see also Danziger et al. 1981). Nonetheless it is clear that the typical Seyfert galaxy has a more developed disk than a typical radio galaxy.
Simkin et al. (1980) and Su & Simkin (198O) have suggested that Seyfert galaxies, when examined carefully, possess disks that differ morphologically from non-Seyferts, and hence lie off the usual Hubble sequence (see Section 3.5).
NUCLEAR EMISSION-LINE GALAXIES We now consider "normal" galaxies with weak ELNs. As discussed by Heckman (1980b) and Stauffer (1981), ELNs in late-type galaxies are generally explainable as nebulae photoionized by hot stars and thus not truly active by our criteria. However, the ELNs in early-type galaxies are characterized by low-level, low-ionization, and probably shock-related activity. (Heckman designated such nuclei as LINERs, for Low-Ionization Nuclear Emission-Line Regions.) That LINERs occur preferentially in the same types of galaxies as Seyferts was cited by Heckman to argue that LINERs are true activity at a low level. A close study of LINERs may reveal whether they represent pre-eruptive active galaxies.
BL LAC OBJECTS The fuzz around a small number of BL Lac objects has been studied in order to determine the nature of the underlying galaxy. In all such cases the galaxy strongly resembles an E galaxy (Miller et al. 1978, Weistrop 1981, Ulrich 1978, Miller 1981), both in terms of its morphology and its stellar content.