214 PROGRESS OF SCIENCE IN THE CENTURY. 



tested by Professor Miller; and Sir Gabriel Stokes 

 suggested in 1850, as Angstrom did in 1853, that 

 the double D line must be due to the absorptive 

 action of sodium vapour in the sun^s atmosphere. 

 Interesting also in this connection was Swan's ex- 

 planation that the appearance of the two yellow 

 sodium lines in all sorts of flames was due to the 

 almost universal distribution of common salt (sodium 

 chloride) in the earth's atmosphere. 



In 1849 Foucault had shown, without seeing the 

 importance of the fact, that the D lines were dark- 

 ened when the sunlight was passed through an elec- 

 tric arc which gave bright sodium lines in its spec- 

 trum. It was reserved for Kirchhoff ten years later 

 to show clearly what this meant. 



Thus spectrum analysis " has grown out of some 

 apparently insignificant and disconnected observa- 

 tions made by Marcgraf, Ilerschel, and others upon 

 the light emitted by flames coloured by certain salts. 

 The spectra of such flames were investigated by 

 various physicists, among whom Talbot, Miller, and 

 Swan deserve first mention ; but it was only after 

 Ivirchhoff (in 1860) had made and proved the def- 

 inite statement that every glowing vapour emits rays 

 of the same degree of refrangibility that it absorbs, 

 • — that spectrum analysis became developed by Bun- 

 sen and himself into one of the great branches of 

 science." * Again we find an illustration of the 

 historical fact that apparently trivial beginnings 

 often lead to great issues, and should never be judged 

 hastily. 



Bunsen and KircliJioff. — These two investigators 

 were the first to show conclusively that definite 



* E. von Meyer. History of Chemistry. Trans. 1891, 

 p. 445. 



