Annu. Rev. Astron. Astrophys. 1984. 22: 445-70 Copyright © 1984 by Annual Reviews. All rights reserved

3.2 Tidal Streams in Nearby Groups

Within the Local Group, several examples testify to the disruption of the outer H I disks of galaxies. The bridge between the Magellanic Clouds (65) and the Magellanic Stream (78) is likely the remnant of a recent close passage of the Clouds past the Milky Way; the high-velocity H I clouds seen on the galactic periphery may also be debris from that encounter (42). Moreover, the outer warps of the Milky Way (64), M31 (81), and M33 (90) may be induced rather than generic.

Figure 1. Contours of H I column density (in units of TA dV) in the Leo triplet of galaxies (clockwise from top: NGC 3628, NGC 3623, NGC 3627), superimposed on an enlargement of the Palomar Sky Survey. A faint optical plume coincides with the H I appendage to the northeast of NGC 3628 (60).

Perhaps the best cases of morphological disruption are the H I appendages, which have been mapped in detail because of their large angular size and favorable viewing geometry. The presently identified systems with intergalactic H I are listed in Table 1. Among the best known of the two-dozen streams are the M81 / M82 / NGC 3077 system (3, 111), the Leo triplet NGC 3623 / 3627 / 3628 (shown in Figure 1; 60, 92), the ``Antennae'' NGC 4038 / 4039 (110), and the NGC 4631 / 4656 system (115). The detailed mapping of the H I distribution permits not only the numerical simulation of the tidal disruption of the sky distribution, but also the tracing of the velocity field. Computer simulations of hypothetical encounters have been developed by numerous authors, especially Toomre & Toomre (108), whose model nicely reproduced the details of the optical images of the ``Mice'' NGC 4676 and the ``Antennae.'' The observed velocity fields merge with the parent galaxies in numerical models of several encounters. Even with relatively simplistic simulations that ignore the subtler effects of magnetic fields, induced shocks, and self-gravitation within arms, the gross characteristics can be reproduced, although the finer details as seen in such streams as the tail of NGC 3628 (60) elude such models. It thus seems clear that tidal encounters can remove a substantial fraction even as much as 50%, of the interstellar H I from a perturbed galaxy.

Most of the H I streams that have been well mapped emanate from galaxies showing optical peculiarities that first aroused the interest of observers. In many cases, the neutral hydrogen appendages coincide with optical plumes. Recent high-resolution observations show that the optical and H I tails of NGC 4747 are offset by 11° in position angle (116); the relationship of the stars and gas seen in the plumes is obviously complex; The detailed mapping of a significant number of interacting systems that do not necessarily show optical plumes or wisps has been restricted by the difficulty in achieving the required sensitivity and angular resolution. Several surveys of binary galaxies (62) and loose groups (54) made with the Arecibo telescope have shown that the phenomenon is common, although by no means ubiquitous. Undoubtedly, some of our understanding of the frequency of interactions is encumbered by the severe selection restrictions that pervade recognition of tidal activity, including favorable viewing geometry, the presence of substantial interstellar gas in at least one of the affected systems, and the relatively short time frame over which the tidal remnant is likely to be prominently separate from the galactic disk but yet not too dispersed. Prediction of the occurrence of tidal activity in any one system is hampered by the absence of three-dimensional perspective in the true distance between galaxies, in their orbital parameters and directions of motion, and in the relative sense of rotation and orbital motion. A vivid example occurs in the Leo triplet, shown in Figure 1: the third member of the triplet, NGC 3623, appears unperturbed, while the disruption of NGC 3627 and NGC 3628 is in marked contrast because of the different orientations of spin and orbit.

While the evidence in support of a tidal encounter origin is compelling in many appendages, the chaotic and distinct nature of some extended H I distributions makes their interpretation still uncertain. The case for discrete, massive ( 10⁸ M) H I clouds (79) is unconvincing (35, 61). At the same time, however, the high-velocity gas near M51 (59, P. Appleton and R. Davies, private communication), the complex H I distribution in Stephan's quintet (1), and the recently discovered intergalactic ``cloud'' in the Leo group (95) do not fit straightforward tidal models involving recent encounters between two neighboring galaxies. It is clear that a single two-dimensional snapshot and a few radial velocities are not enough to convict. In these cases, the puzzle and the frustration of our limited view remain.