It seemed, however, possible to M. Janssen that these bands owned a joint solar and terrestrial origin. Oxygen in a fit condition to produce them might, he considered, exist in the outer atmosphere of the sun; and he resolved to decide the point. No one could bring more skill and experience to bear upon it than he.[692] By observations on the summit of the Faulhorn, as well as by direct experiment, he demonstrated, nearly thirty years ago, the leading part played by water-vapour in generating the atmospheric spectrum; and he had recourse to similar means for appraising the share in it assignable to oxygen. An electric beam, transmitted from the Eiffel Tower to Meudon in the summer of 1888, having passed through a weight of oxygen about equal to that piled above the surface of the earth, showed the groups A and B just as they appear in the high-sun spectrum.[693] Atmospheric action is then adequate to produce them. But M. Janssen desired to prove, in addition, that they diminish proportionately to its amount. His ascent of Mont Blanc[694] in 1890 was undertaken with this object. It was perfectly successful. In the solar spectrum, examined from that eminence, oxygen-absorption was so much enfeebled as to leave no possible doubt of its purely telluric origin. Under another form, nevertheless, it has been detected as indubitably solar. A triplet of dark lines low down in the red, photographed from the sun by Higgs and McClean, was clearly identified by Runge and Paschen in 1896[695] with the fundamental group of an oxygen series, first seen by Piazzi Smyth in the spectrum of a vacuum-tube in 1883.[696] The _pabulum vitæ_ of our earth is then to some slight extent effective in arresting transmitted sunlight, and oxygen must be classed as a solar element. The rays of the sun, besides being stopped selectively in our atmosphere, suffer also a marked general absorption. This tells chiefly upon the shortest wave-lengths; the ultra-violet spectrum is in fact closed, as if by the interposition of an opaque screen. Nor does the screen appear very sensibly less opaque from an elevation of 10,000 feet. Dr. Simony's spectral photographs, taken on the Peak of Teneriffe,[697] extended but slightly further up than M. Cornu's, taken in the valley of the Loire. Could the veil be withdrawn, some indications as to the originating temperature of the solar spectrum might be gathered from its range, since the proportion of quick vibrations given out by a glowing body grows with the intensity of its incandescence. And this brings us to the subject of our next Chapter. FOOTNOTES: [Footnote 596: _Phil. Mag._, vol. xlii., p. 380, 1871.] [Footnote 597: _Astr. Nach._, No. 3,053, _Amer. Jour._, vol. xlii., p. 162; Deslandres, _Comptes Rendus_, t. cxiii., p. 307.] [Footnote 598: _Proc. Roy. Society_, vol. lxi., p. 433.] [Footnote 599: _Phil. Mag._, vol. xlii., p. 377.] [Footnote 600: Frost-Scheiner, _Astr. Spectroscopy_, pp. 184, 423.]