Herschel, as has been already remarked, had sought a cosmology and he had found a cosmogony. Nevertheless, he did not abandon his attempt to discover the structure of the sidereal system. In 1817 and 1818, when nearly eighty years of age, he communicated two remarkable papers to the Royal Society on the extent and condition of the Milky Way and on the relative distances of clusters of stars. In these papers Herschel explained his new method of star-gauging, which some writers have confused with his first. The two methods, however, were quite distinct. In the first, one telescope was used in different regions of the heavens; whereas in the second, various telescopes were turned on the same region. The new method assumed the distribution of the stars to approximate to a certain properly-modified equality of scattering, and also a certain equality of real brightness. In the paper of 1817 he applied this new principle to the Milky Way, and in the paper of 1818 to star-clusters, assuming that the relative distances of "globular and other clusters" can be determined by the telescopic powers necessary to reveal and resolve them, provided that the component stars are, generally speaking, comparable to Sirius in size. Proctor, writing in 1872, contended that "the principle is unsound and that Herschel himself would have abandoned it had he tested it earlier in his observing career". Most writers have agreed with his estimate: yet recent work on star-clusters[1] would seem to indicate that Herschel's second method was not so unsound as has been generally believed. Herschel propounded no hypothesis to take the place of the disc-theory. Indeed, his later view was that the sidereal system was much more extended in the plane of the Galaxy than he had previously believed. "The utmost stretch of the space-penetrating power of the 20-foot telescope could not fathom the profundity of the Milky Way." In 1817, he gave expression to the
- ↑ By Dr. Harlow Shapley, at Mount Wilson Observatory, California.
