As distinguished from the crape ring, the bright rings must have a considerable closeness of texture; for the shadow of the planet may be seen projected upon them, and their shadows in turn projected upon the surface of the planet (see Plate XV., p. 236). According to Professor Barnard, the entire breadth of the ring system, that is to say, from one side to the other of the outer ring, is 172,310 miles, or somewhat more than double the planet's diameter. In the varying views which we get of Saturn, the system of the rings is presented to us at very different angles. Sometimes we are enabled to gaze upon its broad expanse; at other times, however, its thin edge is turned exactly towards us, an occurrence which takes place after intervals of about fifteen years. When this happened in 1892 the rings are said to have disappeared entirely from view in the great Lick telescope. We thus get an idea of their small degree of thickness, which would appear to be only about 50 miles. The last time the system of rings was exactly edgewise to the earth was on the 3rd of October 1907. The question of the composition of these rings has given rise to a good deal of speculation. It was formerly supposed that they were either solid or liquid, but in 1857 it was proved by Clerk Maxwell that a structure of this kind would not be able to stand. He showed, however, that they could be fully explained by supposing them to consist of an immense number of separate solid particles, or, as one might otherwise put it, extremely small satellites, circling in dense swarms around the middle portions of the planet. It is therefore believed that we have here the materials ready for the formation of a satellite or satellites; but that the powerful gravitative action, arising through the planet's being so near at hand, is too great ever to allow these materials to aggregate themselves into a solid mass. There is, as a matter of fact, a minimum distance from the body of any planet within which it can be shown that a satellite will be unable to form on account of gravitational stress. This is known as "Roche's limit," from the name of a French astronomer who specially investigated the question. There thus appears to be a certain degree of analogy between Saturn's rings and the asteroids. Empty spaces, too, exist in the asteroidal zone, the relative position of one of which bears a striking resemblance to that of "Cassini's division." It is suggested, indeed, that this division had its origin in gravitational disturbances produced by the attraction of the larger satellites, just as the empty spaces in the asteroidal zone are supposed to be the result of perturbations caused by the Giant Planet hard by.