80 THE CHEMISTRY OF THE SUN. [CHAP. 



the solar atmosphere was greatly increased by the number of 

 new coincidences observed. Cobalt remained doubtful, the solar 

 lines coincident with a considerable number of its bright lines 

 not having been observed. New coincidences in the spectra of 

 barium, copper, and zinc with dark solar lines confirmed the 

 presence of those elements in the sun's atmosphere. In the cases 

 of strontium and cadmium the number of coincidences seemed 

 ta be too small to warrant the conclusion that those metals are 

 in the sun. The other chemical elements examined, including 

 potassium, did not appear to be traceable in the solar atmosphere. 

 The case of potassium however they considered as doubtful, 

 since faint solar lines are very near the red potassium lines. 



It is not too early to note that when we pass from the 

 spectrum of the spark to the spectrum of the sun we are landed 

 in doubt in many instances. 



These then were the facts accumulated by KirchhofFs method 

 of solar observation, by which the spectra of the various solar 

 regions were not separately observed. 



Kirchhoff next discussed the bearing of this work on the 

 physical and the chemical condition of the atmosphere of the 

 sun. He says : 1 



" In order to explain the occurrence of the dark lines in the solar 

 spectrum, we must suppose that the solar atmosphere incloses a 

 luminous nucleus, producing a continuous spectrum, the brightness 

 of which exceeds a certain limit. The most probable supposition 

 which can be made respecting the sun's constitution is, that it con- 

 sists of a solid or liquid nucleus, heated to a temperature of the 

 brightest whiteness, surrounded by an atmosphere of somewhat 

 lower temperature." 



Of course this at once destroyed, at a blow, the idea of 

 Sir William Herschel that the sun was a cool habitable globe, 



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

 part i. p. 23. 



