of Energy in the Spectra of Solids. 431 



depends upon temperature, and formula (11) gives the law of 

 this dependence : — Whatever the law of emission may be, the 

 wave-length corresponding to the maximum energy is inversely 

 proportional to the square root of the absolute temperature of the 

 source. 



The only experimental measurements, so far as I know, 

 which have been published hitherto upon this subject are con- 

 tained in a recent diagram by Prof. Langley*. In order to 

 compare these data with the relationship #A.f U ax= const, equi- 

 valent to (11), I have deduced from this diagram, by a gra- 

 phical process, the following values of X max : — 



Temperatures. 







i \ 

 c°. 6°. 



Amai' 



<^max- 



178 451 



4-90 



10828 



330 603 



4-05 



9891 



525 798 



3-63 



10515 



815 1088 



3-27 



11634 



Although the constancy of #/V4a K is far from being perfect, 

 yet in view of the manner in which these four values have 

 been obtained, and the uncertainty attaching to the determi- 

 nation of elevated temperatures, we may say that these obser- 

 vations appear to confirm our theory. 



4. Measurement of Temperatures. — If we suppose that this 

 law were confirmed by more numerous observations, it would 

 then furnish us with a very simple method of measuring the 

 high temperatures of radiating bodiesf . This method would 

 be superior to the spectro-photometric method elaborated by 

 M. Crova, inasmuch as it is not dependent upon the choice of 

 the tints compared, and would give, not " optical degrees," 

 but absolute temperatures. Knowing the constant of equa- 

 tion (11), it would suffice, in order to measure the temperature 

 of a radiating body, to determine the wave-length corre- 

 sponding to the maximum intensity of its normal spectrum. 



If we take account, as much as possible, of the absorptions 

 suffered by the solar rays, and if we admit that the great den- 

 sity and enormous mass of the sun ought to give to its primi- 

 tive spectrum all the qualities of continuous spectra, we may 

 apply this method to calculate the temperature of the sun. 

 According to the observations of Prof. Langleyl, the maximum 



* Loc. cit. p. 462. 



t The idea of a similar method, but without any definite law, has been 

 propounded by M. Crova, loc. cit. p. 479. 



X S. P. Langley, ' Researches on Solar Heat,' a report of the Mount 

 "Whitney Expedition, p. 144, pi. xv. curve iii. (Washington, 1884). 



