means of conducting investigations of this kind, which his predecessors had not possessed, was offered in the method of spectrum analysis discovered by Kirchhoff; and he was first able to undertake the application of this method in the beginning of 1862. In preparation for the research he mapped the spectra of twenty-six of the chemical elements, publishing the results of his labors in the Philosophical Transactions of the Royal Society. In conjunction with Dr. William Allen Miller, he compared the spectra of some fifty stars with those of several terrestrial elements, and found that the stars are hot bodies, similarly constituted with our sun, and containing many of the substances found on the earth. In 1864 he and Prof. Miller reported to the Royal Society the results of their observations of the spectra of the planets Venus, Mars, Jupiter, and Saturn; but they had found the light from Uranus too faint to be satisfactorily examined with the spectroscope.
The study of Uranus was resumed with an improved telescope in 1871 by Mr. Huggins, and he found its spectrum to be continuous so far as the feebleness of its light permitted it to be traced, or from C to near G. A photograph of the spectrum of Sirius was obtained by Mr. Huggins and Prof. Miller in 1863, when observations were suspended. They were resumed by Mr. Huggins in 1876, with apparatus so arranged that the spectrum of the sun could be taken on the same plate, and this method was applied to other bright stars. After recording in his communication to the Royal Society his expectation, with apparatus then under construction, of obtaining finer lines which might be present in the stellar light than those that had been seen, and of extending the photographic method to stars that were less bright, Prof. Huggins referred in general terms to "the many important questions in connection with which photographic observations of stars may be of value." Another paper recording the progress of these investigations to the end of 1879 named thirteen bright stars, Venus, Mars, and Jupiter, and different small areas of the moon, to which the method had been applied. Six of the spectra belonged to stars of the white class, while Arcturus seemed to present a spectrum "on the other side of that of the sun in the order of changes from the white-star group." The photographs of the planets showed no sensible planetary modification of the violet and ultra-violet parts of the spectrum. The results of the photographs of lunar areas taken under different conditions of illumination were negative as to any absorptive action of a lunar atmosphere. The author was then preparing to attempt to obtain by photography any lines which might exist in the violet and ultra-violet spectra of the gaseous nebulae. He further pointed out "the suitability of the photographic method of stellar spec-
