384 PROFESSOR FORBES ON THE COLOURS OF THE ATMOSPHERE. 



the character of the second order ; whilst he attributes the tints of flocculent 

 clouds, partially illuminated by the sun or moon, to the first order ; in other words, 

 he supposes the vesicular vapour of which he speaks, to have double the thick- 

 ness in an azure sky, than in the midst of a fog, whilst Newton expressly assigns 

 the blue of the first order to the air, because " it ought to be the colour of the 

 finest and most transparent skies in which vapours are not anlved at that gross- 

 ness requisite to reflect other colom-s, as we find it is by experience." This is only 

 one of the various contradictions into which the artist-like view of matching 

 colours by external resemblances, and assuming a common origin, has led the in- 

 genious author. The application of the colours reflected from vapours to measure 

 the thickness of the vesicles* was, we have seen, completely anticipated by 

 Keatzenstein, and the generality of the application disproved by Melvill half 

 a century ago, when he speaks of the theory of the " gaudy colours" of the clouds 

 arising, " like those of the soap bubble, from the particular size of their parts." 



I have perused Nocili's Memoir with a most anxious wish to arrive at his true 

 meaning, disembarrassed of the somewhat poetical vagueness of his own expres- 

 sions, and the serious mistakes of his translator ; and I believe his view to be this : 

 — There are both transmitted and reflected tints in the sky. The transmitted ones 

 are complementary to the blue of the sky, and therefore, acccording to Nobili, of 

 the second order, whilst all the fiery tints which particularly characterize sunset 

 as contrasted with the dawn, are colours of the Jirst order reflected from the vesi- 

 cular vapours of clouds. 



An ingenious paper by Count Xavieb de Maistre on the colour of air and 

 water, appeared in the Bibliotheque Uiiiverselle for November 1832.f With regard 

 to the atmosphere, the author's theory is so far similar to that of Delaval, that 

 its colom- is to be ascribed to the peculiar state of the particles of water contained 

 in it acting on the principle of opalescence, the reflected light being blue and the 

 transmitted orange. He thence refers to the colours of sunset, and adds, — " But 

 it often happens that the colours are not obsei-ved, and the sun sets without pro- 

 ducing them. It is not, therefore, to the pm-e air alone that we must attribute 

 the opaline property of the atmosphere, but to the mixture of air and vapour in 

 a particular state, which produces an effect analogous to that of the powder of 

 calcined bones in opaline glass. Neither is it the quantity of water which the air 

 contains that occasions these colom's, for when it is very humid, it is more trans- 



lu tlie translation of the paper in Taylor's Scientific Memoirs, i. 99, by an oversight, the maxi- 

 mum thickness of the cloudy vesicles is stated at the ten-millionth of an inch, instead of ten millionths 

 of an inch, or a hundred times greater, as in the original. There is even a slight mistake in the latter ; 

 the tint he describes corresponding to plates of water, not of air, would require a thickness of seven 

 millionths. 



+ Translated in the Edin. New Phil. Journal, vol. xv. 



1 



