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

3.2. Stellar Kinematics

RADIO GALAXIES     An enormous number of papers have appeared recently on the stellar dynamics of active galaxies, and the following tentative conclusions emerge:

1. Radio galaxies associated with powerful radio sources apparently rotate more rapidly than normal elliptical galaxies. Illingworth (1977), Schechter & Gunn (1979), and others find that the rotation amplitude of a typical elliptical galaxy is 50 km s^-1. Simkin (1979) first drew attention to the differences between normal elliptical and radio galaxies in their rotation amplitudes. A survey of the recent literature (Simkin 1977, 1979, Appenzeller & Mollenhoff 1980, Efstathiou et al. 1980, Carter et al. 1981, Goss et al. 1980, Graham 1979, Heckman et al. 1981a, Jenkins 1981, Jenkins & Scheuer 1980, Schweizer 1980, and others) shows that for radio galaxies with a power 10^24.5 W Hz^-1 at 1420 MHz, the average rotation amplitude is ~ 150 km s^-1. We hasten to add that luminous radio galaxies are often associated with D and cD galaxies, and not ellipticals, and Simkin (1979) suggested that the large rotation amplitudes of strong radio galaxies are strongly related to their morphological classification as D galaxies by MMS (Section 3.1). The rotation amplitudes of these radio-loud D galaxies are apparently similar to the bulges of SO galaxies (Kormendy & Illingworth 1981).

Theoretically, however, the relatively large rotation amplitudes of radio galaxies run counter to the proposal of Sparke & Shu (1980) and Sparke (1981) in which a narrow channel in the galaxy's halo (through which the radio-emitting plasma is collimated) is produced by the very small rotation speed of the system. Rapid rotation is also in conflict with the expectation that the gas within the galaxy can freely fall into the nucleus of a galaxy as fuel.

2. Powerful radio sources (> 10²⁵ W Hz^-1 at 1420 MHz) have a projected axis of ejection aligned to within ~ 10° of the galaxy's rotation axis. Less powerful sources show little alignment (misalignments are typically 50°), and the radio, rotation, and isophotal minor axes are often all misaligned with each other (Jenkins 1981). Of course there are specific examples with a high degree of alignment in the low-luminosity radio sources.

Even though the rotation and radio axes are highly aligned in powerful radio sources, the alignment of these axes with the isophotal minor axis is variable in the few cases in which all three axes have been measured. In NGC 612 (a disk galaxy) all three axes are aligned (Goss et al. 1980); in 3C 33 the rotation and minor axes are misaligned by 20°; and in 3C 98 the misalignment is 58° (Simkin 1979, Palimaka et al. 1979).

NUCLEAR VELOCITY DISPERSION Based on observations of an anomalous increase in the stellar velocity dispersion and departures from a King model at the nucleus of M87, Sargent et al. (1978) and Young et al. (1978) proposed that a compact dark mass, probably a black hole, with a mass exceeding 10 ⁹M, lies at the nucleus of M87. Such a black hole might orchestrate the galactic activity in M87 and other active galaxies. This suggestion was supported by further observations of M87 (de Vaucouleurs & Nieto 1979, Jenkins 1980). However, this interpretation of the observations has not met universal acceptance.

Faber (1980) has questioned whether the observations in M87 are unusual and whether their interpretation in terms of a compact mass are required. Duncan & Wheeler (1980) also challenge the interpretive model; they suggest that an anisotropic velocity dispersion can explain the data equally as well. Moreover, Dressler (1980b) finds that the velocity dispersion in the inner 0".5 of M87 is lower than the compact mass model would predict. Finally Young et al. (1979) observed three other ellipticals for similar signs of a compact mass. The only other galaxy to show photometric peculiarities in its nucleus is NGC 6251, the only radio galaxy in the sample (no kinematic data were obtained). NGC 6251 is six times more distant than M87, and a detailed study of its nuclear region is difficult.

There are only two other classical radio galaxies at distances comparable to M87: M84 (Illingworth et al., in preparation) and NGC 1316 = Fornax A (Jenkins & Scheuer 1980). Neither show a similar rise in velocity dispersion near the nucleus. The kinematic anomalies in the nuclei of M87 and NGC 6251, whatever their origin, are not ubiquitous in radio galaxies.

Heckman (in preparation) has found that radio-loud elliptical galaxies have significantly larger nuclear velocity dispersions and global mass-to-light ratios than radio-quiet ellipticals of the same luminosity. It is clear from this that unless large velocity dispersions are transitory, the ability of an elliptical galaxy to produce radio emission depends more strongly on galaxy mass than luminosity.

SEYFERT DISKS     There are several suggestions that a nonaxisymmetric gravitational potential can result in gas flows that might deliver gas to the nucleus of an active galaxy, and virtually all Seyferts show signs of such a potential (Section 3.4). However, the near-nuclear dynamics of Seyfert disks have only been traced extensively in the gas (Section 4.3), which is susceptible to kinematic distortion by the active nucleus.