424 Mr. J. C. McConnel on the 



sun as 23°. For this purpose I have found the naked eye 

 most efficient in winter, though not in summer. 



It is thus obvious that the distribution of colour depends 

 mainly on the distance from the sun when that distance is 

 small, the colours being arranged in circles. But further out 

 it depends more on the variation of some property of the cloud, 

 since the colours follow its edge. In both theories this pro- 

 perty is the average size of the particles, which is no doubt 

 generally greater in the interior of a cloud. So far we have but 

 little evidence as to the proper order of the colours. But along 

 the edge of a large cloud we may expect the particles to be 

 of tolerably uniform size, and the disturbing element to be 

 thereby removed. On one occasion, when such an edge 

 passed almost through the sun, I noted down the colours in 

 order : white, yellow, red ; blue, green, yellow, pink ; green, 

 pink. This list consists evidently of three successive spectra 

 similar to those seen in Newton's rings or in various diffrac- 

 tion experiments. Its accuracy was fully borne out by a 

 number of similar observations. The blue, however, is often 

 replaced by a brilliant purple, due to the first and second 

 spectra overlapping. 



We may now, I think, fairly conclude that, if the sky were 

 overspread by a thin cloud of particles of uniform size, and 

 the eye were not troubled by the glare, we should see a series 

 of coloured rings — blue, white, yellow, red ; blue, green, 

 yellow, pink ; green, pink ; green, pink ; &c. The nearest 

 approach to this in my experience was in April at Bern. 

 With the aid of dark spectacles, I saw nearly complete circles 

 of yellow, orange, red, purple, and green. 



In the summer I have rarely seen colour with the unaided 

 eye, and even with spectacles the iridescence, though seldom 

 absent, is generally insignificant. The most vivid effects are 

 given by unmistakable ice-clouds, while unmistakable water- 

 clouds often show no effects at all. In the high altitudes, 

 in which I have been living since I began to pay special 

 attention to these appearances, the majority of the clouds are 

 no doubt ice ; and for some time I was uncertain whether the 

 colours were ever shown by clouds composed entirely of 

 water. However, one warm day in June, these doubts were 

 set at rest. I was standing on a high ridge, the temperature 

 perhaps being 60° Fahr., when some thin clouds came drift- 

 ing over, perhaps a hundred feet above my head ; I put on 

 dark spectacles, and immediately circles of colour appeared — 

 yellow, orange, red, faint green, in order outwards. Un- 

 luckily I took no measures, but the colours were about the 

 usual distance from the sun. On another occasion, with similar 





