Annu. Rev. Astron. Astrophys. 1982. 20: 431-468
Copyright © 1982 by . All rights reserved

Next Contents Previous

4. GAS AND DUST IN ACTIVE GALAXIES

The gaseous and dusty components of galaxies are generally believed to be the reservoir of fuel for nuclear activity. Clearly it is important to understand not only the sources of such material, but also how the material can find its way into the nucleus. Since gas and dust can behave as a hydrodynamic fluid, they can dissipate energy and redistribute angular momentum, and thus can fall more easily into an active nucleus than can stars. For similar reasons, a fluid can easily be disrupted by energy released from the nucleus, especially if the nucleus ejects a fluid itself. Consequently, studies of gas and dust distributions and dynamics are of great interest, but interpretation must be made cautiously.

4.1. Dust

Dust lanes and patches are common in many active and inactive ealaxies. In systems later than Sa, the normal content of dust is sufficiently large that only gross anomalies in dust content and structure are useful as potential clues to causes of galactic activity. While many of the best known elliptical radio galaxies show conspicuous dust lanes (e.g. Cen A, Fornax A, Cygnus A), there are inactive early-type galaxies having dust lanes (e.g. Hawarden et al. 1981), as well as radio galaxies without conspicuous dust lanes.

A striking correlation between dust lane orientation and the alignment of lobes in eight radio galaxies has been noted by Kotanyi & Ekers (1979). A brief literature search shows that for active ellipticals, dust lanes are strongly preferentially oriented along the minor axis (NGC 1316, 2685, 4374, 5128, 5363, Cygnus A, PKS 1934-63, and 3C 293). In one case (NGC 3665) the dust lane lies along the major axis, in another (NGC 3801) dust lanes are found along both the major and minor axis, and in 3C 305 the dust lanes do not lie along any structural or dynamical axis (Kotanyi & Ekers 1979, Sandage 1961, Goss et al. 1980, Bertola & Galleta 1978, Kormendy & Bachall 1974, van Breugel, in preparation). Hawarden et al. (1981) have studied dust lane orientations in 40 active and inactive early-type galaxies and find roughly equal numbers of dust lanes oriented along either the major or minor galaxy axis. It is clear that dust lanes in active ellipticals are not generally counterparts of the relaxed disks of spiral galaxies. As Hawarden et al. (1981), Schechter & Gunn (1978) and others have suggested, dust lanes are probably recent (108 - 109 yr), unrelaxed additions to the active galaxies, perhaps the results of a merger with a gas-rich system, and probably a by-product of the processes that activate the nucleus.

Because of its proximity (~ 5 Mpc), Cen A provides an important opportunity to study the relationships between the dust, the stellar component, the nuclear activity, and the extragalactic environment. The dust and associated gas has been studied by Appenzeller & Mollenhoff (1980), Dufour & van den Bergh (1978), Dufour et al. (1979), Graham (1979), Mollenhoff (1979, 1981), Phillips (1981), Rodgers (1978), Rodgers & Harding (1980), and Telesco (1978). There is general, but not unanimous, agreement that the thick (~ 2 kpc), patchy dust lane is the nonstellar remnant of a spiral galaxy that was tidally disrupted within the past few rotational time scales (< 109 yr) by Cen A (e.g. Dufour et al. 1979, Tubbs 1980). The "normal" disk heavy element abundances and the distribution of the dust, including the thickness and apparent warps, are taken as evidence of incomplete relaxation of the spiral's remnants as they settle into stable circular orbits. Several tests of this idea have been suggested. (a) Unless the original spiral was a very late type, its former nucleus may still be identifiable. (However, no such nucleus has yet been located.) (b) During relaxation the disk rotation axis may precess. The S-shaped morphology of the radio lobes in Cen A provides indirect corroboration that this process might occur. (c) Parts of the disrupted spiral may have been ejected from the system during relaxation, including some of the gas and dust, and may be observable in the halo of Cen A. (d) As described by Gunn (1979), some gas of the spiral can fall into the nucleus to fuel its activity; evidence of such infall may exist in the H I and molecular line spectra of Cen A (Section 4.2). The biggest problem for this merger scenario is that Cen A has no near-neighbors; thus the present chances of a spiral interacting with Cen A are negligible.

Next Contents Previous