vi.] THE SOLAR ATMOSPHERE. 81 



with trees, and flowers, and everything such as we know of 

 here. If the atmosphere were in a state of sufficient incan- 

 descence to give these phenomena it was absolutely impos- 

 sible that anything below that atmosphere should not be at 

 the same time at a higher temperature. He says, 1 "The 

 height of the solar atmosphere, judging from the phenomena 

 observed in a total eclipse of the sun, is not small in com- 

 parison with the radius of that body ; and hence the distances 

 which two rays have to pass, one of which proceeds from 

 the centre, and the other from the edge of the sun's disk, do 

 not greatly differ." That was a reply to an objection which had 

 been urged to the effect that if a dark line had been produced 

 by anything absorbing in the atmosphere of the sun, there 

 would be a very considerable difference between the spectrum 

 of the sun's limb and the spectrum of the sun's centre, for 

 the same reason, mutatis mutandis, that the sun is white at noon- 

 day and reddish at sunset ; for since our atmosphere is thin, 

 the light passes through a greater stratum in the latter case 

 than in the former. At the sun the light would have to do the 

 same thing, and we should get, therefore, a greater darkening of 

 the limb than is actually observed. His words are : " We 

 must remember that the lowest layers of our terrestrial atmo- 

 sphere, or those in which the distance traversed by the light 

 increases most rapidly when it approaches most nearly the 

 horizon, are those which by virtue of their density must exert 

 the most powerful absorptive action. In the solar atmosphere, 

 on the contrary, it is those layers which are elevated to a 

 certain position above the solid crust of the sun which act 

 most energetically in producing the dark lines ; for the lower 

 layers which possess a temperature but slightly different from 

 that of the mass, effect but little alteration on the light." 



His theory therefore places the region where this absorption 



1 Researches on the Solar Spectrum and the Spectra of the Chemical Elements, 

 part i. p. 22. 



G 



