The plates accompanying this article show two spiral nebulae, the great nebula in Andromeda, with its plane considerably tilted to us, coming comparatively early in the evolutionary sequence. Here the nucleus is large and the centre of the nebula structure less, and only in the outer parts are condensations beginning to form. The other plate shows an open view of a more developed spiral in Ursa Major. It has a much smaller nucleus and the spiral arms are full of condensations throughout their length, the outermost of them becoming stellar in character. Dark mark ings will be seen in the Andromeda nebula, and these are found in many of the spirals. They suggest irresistibly the presence of obscuring clouds similar to those seen in the Milky Way. Many of the extra-galactic nebulae which are seen edge-on are found to be crossed by dark bands, clearly caused by surrounding rings of dark matter hiding some of the luminous nebulosity nearer their nuclei. A typical example is N.C.G. 4594 for an illustration of which, and of several other spiral nebulae, see COSMOGONY.
Spectroscopic observation of the extra-galactic nebulae is dif ficult on account of their faintness, but the central parts of those that are sufficiently bright have all been found to show continu ous spectra, crossed by absorption lines, similar in nature to those of stars like our sun. A few also show bright lines in their spectra; but these come from individual condensations, and not from the nebula as a whole. Thus these nebulae must be formed either of bodies similar to the stars or of matter that derives its light from stars involved in it, in a way similar to that postulated for those of the diffuse nebulae that show similar spectra.
The velocities of these nebulae in the line of sight, revealed by the spectroscope, come out very large, the average being over 500 idiom. a second, many times that of any of the bodies within our galactic system. They also appear almost without exception to be receding from us. No entirely satisfactory explanation of this is yet forthcoming, but it seems possible that these observa tions do not represent real velocities of the nebulae. There are reasons, derived from the general theory of relativity, for suppos ing that extremely distant bodies might appear to be receding rapidly from us. Whatever is the final interpretation of these ob servations, they differentiate the extra-galactic nebulae from any bodies in our stellar system.
During recent years several examples of two types of variable star, which otherwise are only found closely confined to the plane of the Milky Way, have been discovered in some of the larger spirals. As there are none of these stars in the regions around, it seems certain that they lie actually in the nebulae and are not merely seen projected on them. These variables are of the kind known as Cepheids, or are novae (new stars). The Cepheid vari
ables in the Galaxy have the peculiarity that their periods of variation are closely correlated with their actual luminosities, whence from a knowledge of the period and the apparent bright ness of such a star, both easily observable quantities, its distance can be inferred. Assuming that the Cepheids found in the spirals are of the same nature as those in the Galaxy, and that conse quently this law holds for them also, Hubble has determined the distances of the great spiral in Andromeda and of that in Tri angulum to be each of the order of 900,000 light-years. These distances are confirmed by the brightness, when at their maximum luminosity, of the novae in them, if we may assume that this will be on the average the same as it is with those in the Galaxy. And these are the two largest and most conspicuous of the spiral nebulae, and therefore may be among the nearest to us.
The evidence then points unmistakably to the extra-galactic nebulae being bodies at truly immense . distances, reckoned in millions of light-years, and in some cases perhaps in tens of mil-. lions; for, though on this point we have as yet no certain knowl edge, it is possible that all these bodies are of much the same size, the minute apparent size of so many of them being only the effect of immense distance. They must then be of vast actual dimensions, comparable to, though apparently not quite so large as, our own stellar systeml, which they resemble in so many ways; in their lenticular shape; in their colour, their spectra being sim ilar to the integrated spectrum of the Milky Way; in the presence in them of dark clouds of obscuring matter; and in the resolva bility of the more advanced of them into separate stars, which seem similar in character to those in our stellar system. The ab sence of a nucleus in our own system shows, however, that it is not a typical spiral nebula, as does the much greater brightness of the central parts of even the advanced type of spirals.
Thus the problem of the nature of the extra-galactic nebulae seems definitely settled in favour of the "island universe" theory, as the supposition that they are bodies external to our own stellar universe has been called. But as H. Shapley, who has made a special study of the structure of the universe, has said "If we call them islands, the galaxy is a continent." There is no sign of the nebulae thinning out as far as we can penetrate space, and we must picture the whole of known space as an ocean filled with these island universes, and our own stellar system among them but of sufficiently greater size to deserve the more dignified ap pellation of continent. It is a stupendous thought that the light which now reaches us from the • outermost parts of this cosmos started on its journey millions of years ago, and that our photo graphs of these distant universes give us pictures of them, not as they are now, but as they were before the dawn of the human race on this small earth of ours. But then in the ages occupied by stellar evolution a million years is but a day.
(H. K.-S.) NEBULAR THEORY: see COSMOGONY.