190 Proceedings of the Royal Irish Academy. 



appear very vivid in heavy rains consisting of large drops, but this 

 circumstance is easily accounted for. In some of the heaviest showers 

 which I have seen, when the rain was carefully observed falling 

 between the eye and an aperture in a wall, &c., behind which a dark 

 screen was placed, there could be plainly noticed, amongst the large 

 drops, a great number of small ones, which, being impeded by the air, 

 did not descend with the velocity of the others. These small drops 

 may also be occasionally observed in the open air, when the heavy 

 drops, descending in oblique lines, are crossed nearly horizontally by a 

 multitude of spherules wafted by the wind in various directions. It 

 is then these minute particles, which, being present in every shower, 

 produce, when more or less uniform in dimensions, the colours of the 

 supernumerary bows. 



It would appear, from Dr. Young's theory, that in the iris formed 

 by very small drops there ought to be a dark space between the 

 primitive and supernumerary rings ; but however small may be the 

 particles which we examine, these colours lie close to one another, 

 unless, indeed, the primitive bows be supposed to have vanished, and 

 that all the rings are " spurious," being produced by rays of unequal 

 lengths. This theory, also, does not explain the fact which appears to 

 be well established, that the ordinary rainbow actually occurs within, 

 the calculated place of the caustic ; neither does it explain the mixing 

 of the colours at the outer edge of the bow produced by the small 

 drops, which circumstance, taken in connexion with the diminished 

 intensity of the light, would seem to indicate that, in this case, the 

 rays of equal lengths have, at least, partially disappeared, although 

 manifestly this does not take place in the large drops. 



IS'OTE ADDED 11^ THE PEESS. 



The phenomena here described are those which are seen in 

 globular drops ; but as small particles of water adhering to a hair are 

 generally distorted into a spheroidal form, the bows produced by them 

 are somewhat different in figure. Also, if the hair does not pass 

 through the centre of the drop, one of the bows is sometimes absent, 

 or is divided into branches. The large drops are not sensibly dis- 

 torted ; and on this account, when I first perceived that they did not 

 form additional rings, I considered that the production of the latter 

 might be a consequence of the spheroidal figure of the small drops : 

 repeated experiments, however, convinced me that such could not be 

 the case, and that the different appearances observed in both cases 

 were entirely dependent on the size of the particles. 



