of the Atmospheric Absorption. 293 



almost to the actual molecule, whose vibration is felt in the 

 purely selective absorption of some single ray. The effect of 

 the action of the grosser dust-particles, then, is to produce 

 a general and comparatively indifferent " absorption " of all 

 rays, after which the spectrum would everywhere seem equally 

 less bright and less hot. The effect of the molecular absorp- 

 tion is to fill the spectrum with evidences of the selective 

 action in the form of dark telluric lines, taking out some 

 kinds of light and heat, and not others, so that after absorp- 

 tion what remains is not only less in amount, but altered in 

 kind. Between these two extreme examples of absorption, we 

 repeat, an unlimited number of others must exist ; but we 

 shall need here for simplicity to first treat the whole as coming 

 under one or the other of these two types, a procedure already 

 more accurate than the primitive one followed by Bouguer, 

 Herschel, and Pouillet, but which we recognize as still but a 

 convention, which is imposed on us provisionally by the 

 actual complexity of nature. 



It will be seen now more clearly that the whole primitive 

 process followed by Pouillet is an assumption, for it is simply 

 taken for granted that the same proportion will be- absorbed 

 by one like stratum as by another. On actually trying the 

 experiment, however, with media in the laboratory, Melloni 

 long since observed that like proportions were not absorbed 

 by like strata ; and the reason w r as found in the fact just 

 noted, that radiant energy is not a single emanation, but the 

 sum of an infinity of diverse ones, each with its own separate 

 rate of absorption. It follows that the coefficient of trans- 

 mission is truly constant only in the case of the absolutely 

 homogeneous ray, which the ordinary photometer or thermo- 

 meter cannot in the least discriminate, and which the finest 

 linear thermopile or bolometer can but approximately discern ; 

 and hence that the original light of the star or heat of the 

 sun, and the amount absorbed, can at best only be found 

 approximately. However familiar this principle may be in 

 some departments of physics, astronomers and meteorologists 

 have been slow in making this application to the present case, 

 but have continued to deduce the brightness of the stars, or 

 the Solar Constant, from observations in which the radiation 

 is either treated as absolutely homogeneous, or in which its 

 non-homogeneity is scarcely recognized as a factor of im- 

 portance *. 



* Exceptions to this remark are, however, to be made in favour of the 

 very early work of Principal Forbes (Royal Society's 'Philosophical 

 Transactions,' May 1842), and of the more recent labour of M. Crova 

 (Academie des Sciences de Montpellier, 1876). See also the excellent 



