Cause of Iridescence in Clouds. 425 



clouds, I found the boundary between red and green to be 

 about 6° from the sun. 



The peculiar brilliance of the effects during the Engadine 

 winter is, no doubt, partially due to a subjective cause. The 

 eye, continually exposed to the reflexion from the snow, 

 accomodates itself to the intense light, and is better fitted to 

 deal with the bright glare near the sun. Some eyes, perhaps, 

 have not this power of adaptation, for, even at St. Moritz in 

 the winter, some persons are quite unable, without glasses, to 

 see the colours. The near proximity also of the ice -clouds 

 adds greatly to the display, for, though the brightness is not 

 increased, each colour is spread over a larger portion of the 

 sky, and thereby rendered much more effective. 



Having now put the reader in possession of the facts to be 

 explained, I will give my reasons for thinking Dr. Stoney's 

 explanation untenable in the face of these facts. In the first 

 place, it does not apply to the very similar phenomena seen in 

 water-clouds, so that in this case, at any rate, he would have 

 to resort to diffraction. Secondly, though it gives the colours 

 in their proper order, it leads to a hopelessly exaggerated 

 idea of scale. 



Let cf> be the angle of incidence on the plate of ice, 

 <p / „ ,, refraction, 



yfr the angular distance of the cloud from the sun. 



Then we have 



ifr=180°-2<£ 

 sin <p=fi sin <j>' 



where n is the order of the spectrum and t the thickness of 

 the plate. Let us suppose the sky to be covered with a thin 

 cloud, composed of plates of such thickness as to give the 

 blue of any order at ty = 5° ; then taking fi=l'Sj and the 

 wave-length of red light half as great again as that of blue, 

 we find for the red of the same order ^ = 136°. And this is 

 neglecting the variation of /u, with \, so that really ty would 

 be greater still. Thus a single spectrum would be spread 

 over almost the whole expanse of the sky. It is needless to 

 say that this is utterly at variance with the facts I have 

 described. 



Thirdly, the brightness of the colours does not fall off in 

 the theory of thin plates nearly as rapidly as we find in 

 observation. 



As the plate is inclined at different angles to the sun, the 

 quantity of light reflected passes through a series of maxima 



Phil. Mag. S. 5. Yol. 24. No. 150. Nov. 1887. 2 F 



