Burton — On the Supermimeraj'y Rings of the Rainhoiv. 189" 



oil. The principle might also be illustrated by melting small pieces 

 of glass into globules of various sizes, and all the phenomena referred 

 to can be seen in the dewdrops hanging from slender blades of grass, 

 &c. 



The diameter of the largest raindrop which can produce the super- 

 numerary colours (estimating it without micrometrical appliances) is 

 about aV of an inch ; but from the general appearance of the bows it 

 is probable that the drops which form them are much smaller. Dr. 

 Young has calculated the size of those necessary to cause the pheno- 

 mena described by Dr. Langwith to be about Ve" of an inch, and says it 

 would be sufficient if they were between -yV and -^V [Phil. Trans., 

 vol. xciv., p. 48). It is certain that a slight diiierence in size does 

 not interfere with the regularity of the bows ; for, if we look into 

 the web of a very minute spider, which, resembling gossamer, is to be 

 met with on the ground, we can often perceive upon it drops of dew 

 not exceeding xou of an inch in diameter, and these collectively form 

 several supernumerary rings, which appear perfectly regular and 

 concentric with the outside bow, although it can scarcely happen that 

 the drops should be exactly equal in size. 



If the colours of the iris be seen through a telescope, as in M. Babi- 

 net's experiment, in which they are observed in a descending column 

 of water let down through a small aperture, it is evident that the limit 

 at which interference ceases to take place ought to be greater than 

 when the phenomena are observed by the naked eye. Professor Potter, 

 in detailing the results of his observations, mentions^ that, when the 

 water was xo of an inch in diameter, the interference bars were plainly 

 visible, but in some other instances they did not appear at all. He 

 has not stated the cause of their non-appearance, but it was probably 

 owing to the thickness of the column of water exceeding the limit at 

 which interference occurs. 



Dr. Pemberton, in endeavouring to explain why the supernumerary 

 colours usually appear more vivid under the upper part of the bow, 

 remarks that " it is most likely they are formed in the vapour of the 

 cloud which the aii', being put in motion by the fall of the rain, may 

 carry down along with the large drops, "^ but it being certain that the 

 clouds produce no such colours, we should rather attribute the circum- 

 stance to the greater abundance of the small drops in the higher re- 

 gions, their number being diminished as they descend, on account of 

 several particles coalescing with one another, and forming drops too 

 large to produce colours by interference. It is also very probable that 

 the unusual vividness of these bows, which occasionally occurs, is 

 owing to the preponderance of small drops in the shower which pro- 

 duces them, and to their general uniformity in size. It may seem in- 

 consistent with this explanation, that the additional coIotu's often 



~ Philosophical Magazine, May, 1855, p. 321. 



^ Fhllosophical Transactions, abridged, vol. vi., p. 1-10. 



