But the rings of Saturn are easily the most fascinating features of the system. They can never be seen as if we were directly above or beneath the planet so they never appear circular, as they really are in space, but always oval or elliptical in shape. The minor axis or greatest breadth is about one-half the major axis or length. The latter is the outer ring's actual diameter, and it amounts to 170,000 miles, or two and one-half times the diameter of Saturn's globe. There are in fact no less than four rings; an outer ring, sometimes seen to be divided near its middle; an inner, broader and brighter ring; and an innermost dusky, or crape ring, as it is often called. This comes within about 10,000 miles of the planet itself. After the form and size of the rings were well made out, their thickness, or rather lack of thickness, was a great puzzle. If a model about a foot in diameter were cut out of tissue paper, the relative proportion of size and thickness would be about right. In space the thickness is very nearly 100 miles, so that, when we look at the ring system edge-on, it becomes all but invisible except in very large telescopes. Clearly a ring so thin cannot be a continuous solid object and recent observations have proved beyond a doubt that Saturn's rings are made up of millions of separate particles moving round the planet, each as if it were an individual satellite. Ever since 1857 the true theory of the constitution of the Saturnian ring has been recognized on theoretic grounds, because Clerke-Maxwell founded the dynamical demonstration that the rings could be neither fluid nor solid, so that they must be made up of a vast multitude of particles traveling round the planet independently. But the physical demonstration that absolutely verified this conclusion did not come until 1895, when, as we have said in a preceding chapter, Keeler, by radial velocity measures on different regions of the ring by means of the spectroscope, proved that the inner parts of the ring travel more swiftly round the planet than the outer regions do. And he further showed that the rates of revolution in different parts of the ring exactly correspond to the periods of revolution which satellites of Saturn would have, if at the same distance from the center of the planet. The innermost particles of the dusky ring, for example, travel round Saturn in about five hours, while the outermost particles of the outer bright ring take 137 hours to make their revolution. For many years it was thought that the Saturnian ring system was a new satellite in process of formation, but this view is no longer entertained; and the system is regarded as a permanent feature of the planet, although astronomers are not in entire agreement as to the evolutionary process by which it came into existence--whether by some cosmic cataclysm, or by gradual development throughout indefinite aeons, as the rest of the solar system is thought to have come to its present state of existence.