There is another circumstance connected with the movement of the sun which should here be mentioned. It is well known that the sun has a velocity which carries it on through space at the rate of half a million miles a day. In this movement the whole solar system, of course, participates. This movement of translation of our system must also be a result of the movements of the two original colliding masses. These two masses imparted to the system, which resulted from their union, both the lineal velocity with which it advances through space, and also that moment of momentum which is of such vast importance in the theory. A consideration of the probabilities of the case make it quite certain that the celestial bodies we see are as nothing compared with the dark bodies we do not see. The stars we see are moving, and the natural assumption is that the dark objects with which the heavens teem are also in motion. We shall, under these conditions, not feel any insuperable difficulty in the supposition that collisions between different bodies in the heavens may have taken place from time to time. We remember that these bodies are moving in all directions, and at extremely high velocities. We are quite willing to grant the excessive improbability that any two bodies particularly specified should come into collision. Within view of our telescopes we have, however, a hundred millions of stars, and if we multiply that figure even by millions, it will still, we may well suppose, not be too large to express the number of bodies which, though contained within the region of space ranged over by our telescopes, are still totally invisible. In these circumstances, we may admit that occasional collisions are not impossible. Please note the strength which the argument derives from the enormous increase in our estimate of the number of bodies, when we include the dark objects as well as the stars. If we were asked whether it would ever be possible for two bright stars to come into collision, we might well hesitate about the answer. We know, of course, that the stars have proper motions; we know, too, that the stars, in this respect unlike the planets, have no definite directions of movement under the control of a supreme co-ordinating attraction. Some stars move to the right, and some to the left, some one way and some another; but even still, notwithstanding their great number, the extent of space is such that the stars keep widely apart, and thus collisions can hardly be expected to take place, unless perhaps in a cluster such as that shown in Fig. 53. We have no reason to think that a collision between two actual bright stars was the origin of the primæval nebula of our system. But when we reflect that the stars, properly so called, are but the visible members of an enormously greater host of objects, then the possibilities of occasional collision between a pair of these incomparably more abundant dark bodies seems to merit our close attention.