303 



Article XL 



On the Nature of those Constituents of the Atmosphere by 

 which the Reflexion of the Light within it is effected. By 

 R. Clausius. 



[From Poggendorff 's^^wwa^ew, vol. Ixxvi. p. 161.] 



IT is a known property of our atmosphere that it is not abso- 

 lutely transparent, but that it enfeebles the light considerably, 

 and that hence the sun must appear less bright at the earth's 

 surface than to an observer placed beyond the limits of the 

 atmosphere. This diminution is due for the most part to 

 a reflexion which takes place everywhere throughout the 

 atmosphere, which may be inferred from the quantity of re- 

 flected light sent from it to the surface of the earth. That the 

 light of the firmament is actually reflected light, is pretty mani- 

 fest of itself, and is rendered certain by the phaenomena of 

 polarization which it has lately been discovered to exhibit. 



The question as to the real origin of this reflexion within the 

 atmosphere now presents itself to us, and its solution would be 

 of two-fold interest. In the first place, the theoretical investi- 

 gation regarding the strength and nature of the reflexion must 

 be based upon the knowledge of its cause; secondly, this 

 knowledge w^ould be of service in many other considerations 

 regarding the atmosphere, and serve more particularly to 

 establish a connexion between its optical and meteorological 

 phenomena. 



With regard to this subject, w^hich is intimately connected 

 with the blue colour of the firmament and the morning and 

 evening red, various conjectures have been expressed by many 

 investigators, but none of these have been sufficiently established 

 to win general recognition, and the question is still the subject 

 of very divergent and indefinite notions. Probable, however, as 

 it may appear from this that a decision of the question in the 

 present state of science is impossible, and that we must await 

 the result of new observations, I still believe that conclusions 

 may be drawn, even now, from known premises which lead to 



