THE PLANET SATURN. 361 



distance just under a diameter and a quarter from the planet's 

 center. Within this distance, then, no satellite of any consider- 

 able size can circulate for the reasons above stated. 



Now, the most remarkable fact remaining is that the outer 

 edge of Saturn's ring system lies just within this limit, so that 

 the conclusion as to its nature seems to point to the " meteoric 

 theory/' as it is called, as the only possible one. Either a satellite 

 has been drawn within the fatal circle and disrupted, or the mate- 

 rials now present as a ring have been prevented from uniting to 

 form a single satellite, as they might otherwise have done. 



So much, then, for theory. The next point is, What proof can 

 we get to substantiate it ? This might seem at first a hopeless 

 task, but that wonderful instrument, the spectroscope, has recently 

 given us direct testimony on the subject. 



One of the peculiarities of the spectroscope is its ability to 

 detect the motion of a luminous body in the line of sight, by the 

 shifting of the dark (Fraunhofer) lines of its spectrum from their 

 normal position as seen in the spectrum of direct sunlight. Ad- 

 vantage was taken of this fact by Mr. J. E. Keeler, who obtained 

 photographs of the spectrum of Saturn and its rings which plain- 

 ly showed that the shifting of the lines due to the motion of the 

 rings was greater in each case for the inner edge than for the 

 outer, proving conclusively that the portions of the ring nearer 

 the planet move faster than those farther away. 



Let us see what this means. In the first place, if we suppose 

 the rings to be solid, it is evident that they must rotate as a whole, 

 the angular velocity of all parts being the same, but the linear or 

 actual velocity being much greater at the outer edge of the ring 

 than the inner, because of the greater circumference of the circle 

 traveled over in rotation. 



If, on the other hand, the ring is composed of separate par- 

 ticles, each in effect a little moon, it is apparent that the nearer 

 these tiny satellites are to the planet the faster they must revolve 

 to overcome the increasing pull of the planet and save themselves 

 from being drawn to destruction upon its surface. In this case, 

 therefore, the inner edge of the ring will have a much greater 

 velocity than the outer. 



Thus we see that the two theories require opposite condi- 

 tions to obtain, and that the proof given by the spectroscope 

 confirms directly the approximate correctness of the "meteoric 

 theory." 



This latter theory offers a ready explanation for the curious 

 " crepe " ring. Shading off gradually as this ring does from the 

 inner edge of the bright one, it is natural to suppose that it is a 

 portion of the former ring in which the fragments or " meteorites " 

 are more sparsely distributed, their numbers growing gradually 



VOL. LI. 28 



