THE STRUCTURE OF THE UNIVERSE 391 



When now he compared the number of stars of average 

 equal brightness with the estimates of their relative distance, 

 it was found that there was no correspondence. If we were 

 to conceive the stars distributed in a series of spheres, or spherical 

 shells, one enclosing the other after the manner of a Chinese 

 egg, the earth or the sun being supposed at the centre, it would 

 follow from the supposition of even spacing that the number 

 of stars in each shell would increase in the proportion of the 

 cube of the distance. The figures did not correspond. The 

 actual number increased much more rapidly in the lesser magni- 

 tudes than the theory would suppose. 



Towards the end of his life, therefore, Herschel seems to 

 have concluded that it was impossible to set any definite bounds 

 to the stellar system ; but he appears more or less to have 

 retained the belief that its general shape answered somewhat 

 to the conclusions of his earlier research. 



The subject appears to be one of endless fascination to the 

 astronomer, and since the time of Herschel a considerable 

 number of minds have attempted the problem. One of the 

 most notable was the elder Struve, long the director of the 

 Pulkowa Observatory. He combined the result of a count of 

 the stars of several magnitudes made by Bessel in a wide zone, 

 with the gauges of Herschel. He adopted the same theory as 

 the latter, supposing that the brightness of the stars supplies, 

 on the average, a measure of their relative distance. His con- 

 clusions were that the stellar system might be made up of a 

 series of layers of varying density, lying parallel to the plane 

 of the Milky Way. In the via lactea the stars would be densest ; 

 he would conceive that in this layer they are spread out as in 

 a wide, thin sheet, with our sun situated somewhere near the 

 middle of the layer. Were we to journey outwards in a direc- 

 tion perpendicular to this plane, we should find the stars growing 

 thinner and thinner without perhaps ever reaching a boundary. 



Against the hypothesis of Herschel, Struve, and all their 

 like, the late Richard A. Proctor, well known as a writer on 

 the more popular side of astronomy, brought forward objec- 

 tions that seem decisive, at least so far as our present know- 

 ledge extends. One was as to the assumption that the stars 

 are more or less alike in their actual light-giving power. Stars 

 that we now know to be of very nearly equal distance may vary 



