3.3. Observations of Ring Kinematics
Some of the earliest studies of ring galaxies showed that the rings both expanded and rotated (e.g., Theys and Spiegel 1976; Fosbury and Hawarden 1977). Most optical spectroscopic studies of ring galaxies have involved obtaining long slit spectra at various position angles around the rings and assumed that the rings are true circles seen at some arbitrary angle (Jeske 1986; Few, Madore and Arp 1982; Charmandaris and Appleton 1994). If this simple assumption is made, the kinematics of the radial velocity field of ring galaxies can then be interpreted as both rotation and expansion of the ring. Typical ratios of radial expansion to rotation velocity in rings range from 5, in the case of NGC 985 ring to 0.1 in VIIZw466 (from Jeske 1986). In Figure 11 we show this ratio plotted against the linear diameter of the ring. On naive grounds one might expect that the smaller rings may show larger radial expansion velocities than the larger rings, since the smaller rings presumably have responded to the perturbation most recently (1)
Figure 11. The ratio of V(expansion) to V(rotation) versus linear diameter from Appleton and Marston (1995) for a small northern sample of ring galaxies. On naive grounds it might be expected that the expansion velocity might dominate over rotation for small (young) rings as compared with larger more evolved systems. However no clear trend emerges (see text).
However, it seems that no clear trends emerge from this plot. NGC 985 and Vela (Taylor and Atherton 1984) show the largest expansion velocities relative to their small rotational velocities. However, it must be remarked that both these galaxies show marked deviations from simple ellipses and the assumption that their geometry is that of a deprojected circle may be incorrect. In Vela, the velocity field may be further complicated by the existence of a weak bar. Certainly for NGC 985 there is evidence that the ring is a tightly wrapped spiral probably resulting from a highly off-center collision (Appleton and Marcum 1993). In general, radial velocities tend to be comparable to or somewhat less than the local ring rotational velocity. Such values suggest that the perturbation is in most cases relatively strong.
Very few kinematic studies of ring galaxies have been published which provide full 2-dimensional information about the dynamics of ring galaxies. In cases where detailed H velocities are available around the entire ring (e.g., the TAURUS imaging of the Vela ring by Taylor and Atherton 1984) there are strong indications that a simple expanding and rotating circular ring is only a first-order description of these data. Deviations are not altogether surprising. Firstly, if the companion makes a non-central collision, the ring will not be symmetrical, either in appearance or in the subsequent expansion of the ring into the disk. Secondly, if the companion is massive enough, it will induce large warping and other non-planar effects into the disk. The combination of these two effects is to lead to azimuthally dependent structure in the radial velocity field of the expanding ring.
In order to fully compare models to observation, it is necessary to detect the motion of the target disk both inside and outside the ring. Since the H emission is exclusively found in the ring component, observations have to rely on either detection of faint stellar absorption lines inside the rings (e.g., as in IIZw28, Sargent 1977; Appleton and Charmandaris, in preparation) or from interstellar gas. As discussed above, very few galaxies have yet been mapped in the 21 cm HI line or in the CO millimeter line.
Ghigo et al. (1994) have performed VLA HI mapping of the ring galaxy NGC 2793 with low spatial resolution. This galaxy has a companion which lies a few ring diameters away on the minor axis and appears to be a classical ring galaxy. The rotation curve is found to be rising from the inner regions to the outer edges of the ring, with evidence for significant non-planar motions in the gas. The best fit to a rotating and expanding ring was found in the regions of the optical ring, where the expansion velocity was found to be about one-half the rotation velocity at that radius. Asymmetries in the velocity field were observed and are to be expected in a collision in which the companion drives a `banana' type wave of the kind discussed by Appleton and Struck-Marcell (1987b).
1 The theory predicts that this depends sensitively on the structure of the gravitational potential of the ring galaxy (see Section 4). Back.