390 



SCIENCE. 



[N. S. Vol. VIII. No. 195. 



The beautiful pbenomenon of the dis- 

 placement of the spectral lines through 

 motion in the line of sight has given rise to 

 many important and Interesting results, 

 but certainly none more striking than that 

 offered by Professor Keeler's spectroscopic 

 proof of the meteoric constitution of the 

 rings of Saturn. It was suggested soon 

 after the discovery of the rings that they 

 must be made up of discrete particles re- 

 volving in zones about the planet, which, 

 from their smallness and great distance 

 from us, gave the appearance of a system 

 of solid rings encircling Saturn. This 

 had been shown by Clerk Maxwell to be a 

 mathematical necessity, and as the rings 

 lay within Eoche's limit, within which a 

 large solid body would be broken up in re- 

 volving about a planet by the unequal at- 

 traction of the planet itself, it was certain 

 that the rings must consist of small indi- 

 vidual bodies. It remained for the spec- 

 troscope, through the aid of photography, 

 to add its testimony to that of mathematical 

 analysis. The problem offered to the spec- 

 ^troscope was simply to show whether the 

 inner or the outer portion of the rings 

 moved the faster. Should they revolve as 

 a solid body the outer edge must necessarily 

 have the greater velocity. But if they are 

 made up of individual particles, then the 

 attraction of the planet would cause those 

 inearest to it to move the fastest, or, in other 

 words, the inner part of the rings must 

 have the greater velocity. This beautiful 

 problem was successfully solved by the 

 photographs of the spectrum of the rings 

 obtained by Professor Keeler, where the 

 displacement of the spectral lines by mo- 

 tion in the line of sight showed that the 

 inner portion of the rings moved faster 

 than the outer, and hence that the rings 

 must consist of small bodies responding in- 

 dividually to the attraction of the planet. 



The discovery of variable stars by photog- 

 raphy can be compared with the wholesale 



business in commercial circles, because of 

 the great number that are found on the 

 various plates. These stars are not only 

 found by the actual variation of their light, 

 as shown by the size of their images on 

 different plates, but many of them also 

 show peculiarities in their spectra which at 

 once stamp them as being members of a 

 certain class of variable stars. So expert 

 has Mrs. Fleming, of the Harvard College 

 Observatory, become in detecting these 

 bodies by their spectra that she instantly 

 recognizes them at a glance among hun- 

 dreds of other spectra on the same plate. 



The most interesting and important of 

 these Harvard College variable-star discov- 

 eries are found in the photographs of the 

 globular clusters taken by Professor Bailey 

 with the 13-inch telescope at Arequipa, 

 Peru. It was found that a great many of 

 the small stars that make up these clusters 

 varied regularly and rapidly in their light, 

 and in some cases a large percentage of 

 the entire mass of stars was variable. So 

 abundant are these variables, indeed, that 

 as many as a hundred of them have been 

 found in a space in the sky that would be 

 covered by a pin's head held at the distance 

 of distinct vision. 



The clusters most prolific in variables 

 are M 3, Omega Centauri, M 5, and a few 

 others of this class. Perhaps the most 

 remarkable circumstance, outside of the 

 actual grouping of variables in such great 

 numbers, is the fact that not a single varia- 

 ble star has been found in the great cluster 

 of Hercules, the best known of these ob- 

 jects, and apparently like them in all other 

 respects. Professor Pickering finds every 

 star in this cluster constant in its light 

 from the photographic evidence extending 

 through ten years. * This would seem to 

 mark this great cluster as being physically 

 very different from the others referred to. 



* Professor Bailey has since found that two of the 

 stars of this cluster are slightly variable. 



