On the Colour of the Atiiiospherc. 197 



colour of the sky results from a mixture of light and shade ; and 

 that of Muncke, that that colour is merely subjective, or arises from 

 an ocular deception. Tlie remaining theories are : 



(1.) That the colour of the sky is that transmitted by pure air, 

 and that all the tints it displays are modifications of the reflected 

 and transmitted colours. This is more or less completely the opi- 

 nion of Mariotto, Bouguer, Euler, Leslie, and Brandes. 



(2.) That the colours of the sky are explicable by floating va- 

 pours acting as thin jdates do, in reflecting and transmitting com- 

 plementary colours. This is the theory of Newton and most of 

 his immediate followers, and more lately of Nobili. 



(3.) On the principle of opalescence and of specific absorption, 

 depending on the nature and unknown constitution of floating par- 

 ticles. This head is intended to embrace the various opinions of 

 .Melvill, Delaval, Count Maistre, and Sir D. Brewster. 



To the last named philosopher, huwever, the merit is due of 

 iiaving conspicuously turned attention to the important, com- 

 plex, and hitherto unexplained phenomena of absorption, which 

 lie has proved to be totally inconsistent with Newton's theory of 

 the colours of Nature, (considered as those of thin plates) ; and he 

 lias fiirther demonstrated the inapplicability of it in the case of the 

 colours of the atmosphere, by shewing that their constitution is 

 wholly distinct from that which any modification of Newton's theory 

 \vould assign, by a series of experiments of which as yet the re- 

 sults only are announced. 



Since, then, the constitution of the atmospheric colours analyzed 

 by the prism resembles that produced by absorption, the question 

 i-, To what medium are we to refer that absorptive action? Evi- 

 (lontly not to pure air, since a distant light is red in a fog, and in 

 dear weather white, or nearly so. The author is disposed to at- 

 tribute the eff'ect to the presence of vapour in the very act of con- 

 ilensation. This intermediate or colorific stage occurs between 

 i!ie colourless and transparent form of steam wholly uncondensed, 

 .iiid that which may be termed the state oi proximate condensation 

 in which it is seen to issue from the spout of a tea-kettle, when it 

 i.; likewise colourless, but semiopake. During the transition, it 

 ^vas shewn in the former paper that steam becomes intensely red, 

 ji:id remains transparent. The absorjitive action resembles then, 

 ^•| far, that of the atmosphere observed under certain meteorologi- 

 <• d conditions ; the dark lines and bands noticed by Sir David 

 f'rewster in the atmospheric spectra have not been discovered, and 

 .•<'i far the analogy is as yet imperfect.* 



In applying this theory to the colours of sunset in particular, the 

 author quotes many acknowledged facts to prove that the redness 



' Borne plausible reasons are assigned why these bands should not liave 

 ..(>i>uared in the experiment as it was made, when steam in every stage of 

 f"iidensation must iioccssarily have been present; nor does it seem easy to 

 di'vise a form of experiment free from this objection. A very important 

 observation would be to examine the spectrum produced by a distant arti- 

 licial light seen through a red fog. 



