FACTS AND PUZZLES CONCERNING THE STRUCTURE OF GALAXIES The views of V.A. Ambartsumyan on the origin of galaxies are often interpreted, especially in the west, as a hypothesis about the instability of multiple systems and galactic clusters. In point of fact, Ambartsumyan's ideas on this subject are much more comprehensive than any hypothesis concer ning the mechanical instability of galactic groups. The main thesis of his theory is that galactic centers are superdense formations, which are highly active. Their splitting gives rise to multiple galaxies, while the ejection of smaller masses of superdense prestellar matter produces spiral arms, globular star clusters, and diffuse matter. We would like to dwell on some consider ations related to Ambartsumyan's theory.
The interesting idea of checking statistically whether double galaxies originate simultaneously, specifically by the splitting of a single one, or whether they are produced as a result of capture was advanced by Page /1/. His preliminary findings led to some unexpected and implausible results, and it became clear that his apparently simple idea is in practice complicated by a number of factors which have to be studied.
It seems to us beyond doubt that physical double and multiple systems have a common origin. This has been well established by V.A. Ambartsu myan and others. In particular, interacting galactic pairs, which are evidently physically coupled, are so frequent that they could not possibly be produced by random encounters.
The capture theory has been thoroughly studied in connection with O.Yu. Shmidt's cosmogonic hypothesis. It was shown that for capture to take place a very unique approach of three bodies is required. The possibility of capture among the isolated galaxies of the field in which interacting galaxies are commonly encountered is therefore highly unlikely.
The velocity difference in physical pairs is also on the average lower than the peculiar velocity of the galaxies. This is an additional strong argument against the capture theory.
Less certainty prevails where the instability of clusters and groups and their positive energy are concerned. Most authors tend to draw conclusions
about the instability of the small groups brought under study on the basis of the known radial velocities. The calculations, however, involve a number of supplementary assumptions and some inaccurate data. These include the distance to the group, and the assumption that the directions linking a few galaxies in pairs are randomly distributed, with uniform probability, etc. An analysis of the possible calculation errors, carried out by Limber and Matthews /2/, shows that the results are quite unreliable. It is interesting to note that there were even cases where the determination of the radial velocities of components which seemed definitely to belong to the same group yielded values differing by hundreds and thousands of kilometers per second from the mean velocity of the group. It thus becomes doubtful that such galaxies actually belong to a group, in spite of their apparent size and position in the sky. If we want to consider them as members of a group, we must necessarily conclude that the group has been formed by an explosion. The fact that the configurations of these components do not display any visible perturbations cannot be construed as evidence that their projection in space is accidental, though; such perturbations are apparently not of gravitational origin and are sometimes absent at quite small distances, while in other cases they are observable at considerable distances between the components. Matters are further complicated when galactic clusters are considered, because it is even more difficult to single out the member of a cluster. The size of any cluster is always much less than the distance to it. Thus the members of a cluster cannot be differentiated from a background galaxy situated, say, five cluster radii away from it along the line of sight. This is borne out by the fact that, for instance, there are many investigators who no longer consider the cluster in Virgo as a single cluster.