ena of dissociation are seen in a wonderful degree in the sun, the fixed stars, and the nebulæ. It is not necessary to recall to you the marvelous field of celestial chemistry which the spectroscope, in the hands of Kirchhoff and his followers, has made known to us, nor the proofs that the solar atmosphere contains in a dissociated state very many of the elements which in our own planet are met with in a free state only in the laboratory of the chemist. It is instructive to compare the spectra of the various fixed stars with each other, from white stars like Sirius, to yellow stars like Aldebaran and our own sun, and red stars like Alpha Orionis and Antares, and to note in these three classes an increasing complexity of chemical composition. In the first, with a predominance of hydrogen, we see only faint lines of magnesium, sodium, calcium, iron, and a few other metals, while in the second, though free hydrogen still abounds, the number of metallic elements is greatly augmented, and finally in the red stars hydrogen is seen only in combination, as aqueous vapor, the metals are wanting, and the metalloids and their compounds appear. If, in accordance with the nebular hypothesis, we look upon these different types of stars as representing successive stages in the process of condensation from nebula to planet, we may also see in them a gradual evolution of the more complex from the simple forms of matter by a process of celestial chemistry. Such was the view put forward by F. W. Clarke in January, 1873, and some months later by Lockyer, who has reiterated and enforced these suggestions, and, moreover, connected them with the speculations of Dumas on the composite nature of the elements. The white stars are the hottest, and in the atmosphere of these bodies the various metals, according to Lockyer, make their appearance in the order of their vapor-densities.
I ventured, in 1867, while speculating on the phenomena of dissociation, to remark that, although from the experiments of the laboratory we can only conjecture the complex nature of the so-called elementary substances, we may expect that their "further dissociation in stellar or nebulous masses may give us evidence of matter still more elemental." Now, while the nebulæ, when scanned by the spectroscope, show us only the lines of hydrogen and nitrogen, the two lightest forms of gaseous matter known to chemistry, it is remarkable that the recent studies of the solar chromosphere reveal to us the existence of an unknown gaseous element which, from its extension beyond even the layer of partially cooled hydrogen, must, according to the deductions of Mr. Johnson Stoney, be still lighter than this gas. The green line by which this substance is distinguished is not as yet identified with that of any terrestrial element. Is it not possible that we have here that more elemental form of matter which, though not seen in the nebulæ, is liberated by the intense heat of the solar sphere, and may possibly correspond to the primary matter conjectured by Dumas, having an equivalent weight one-fourth that of hydrogen? Mention
