TRANSACTIONS OF SECTION A. 657 



the primary and separated from it by a space of about three degrees. It consists of 

 a ring of green succeeded on the inside by a ring of red light, each being about half 

 a degree in breadth and concentric with the principal bow. Phenomena of this 

 kind came under my notice on the mornings of the 22nd and 23rd September, 1862, 

 and 14th July, 1876, in the neighbourhood of Kanturk, in the county of Cork. On 

 some of these occasions I remarked that the outer verge of the green ring was 

 better defined than the inner verge of the red one. The latter was sometimes 

 irregular, owing apparently to the presence of small needles or lines of grey or 

 whitish light crossing the arc. The appearances in each instance were visible for 

 nearly an hour. 



Having some time ago searched through various scientific periodicals up to the 

 year 1880, I found therein only two instances in which corresponding phenomena 

 had been recorded, one of which is in the description of a lunar fogbow seen off 

 the coast of California, in 1860, by Lieut. J. M. Gillis, of the American Navy : in 

 this no colours were observed ('American Journal of Science,' vol. xxx. p. 305). 

 The other description referred to is contained in Col. Sykes' paper ' On the 

 Atmospheric Tides and Meteorology of the Deccan,' where he says : ' Under a com- 

 bination of favourable circumstances, it (the circular rainbow) appeared quite perfect 

 of the most vivid colours, one half above the level on which I stood, the other half 

 below it. * * * The brilliant circle was accompanied by the usual outer bow 

 in fainter colours ' (' Philosophical Transactions for 1835/ p. 193). In this instance 

 it appears possible that the inner bow which he describes as a " brilliant circle " 

 was exceeded in brilliancy by the outer bow, notwithstanding the statement that 

 the latter appeared " in fainter colours ; " for this may have reference to the mixture 

 and consequent indistinctness of the colours, and not to the intensity of the light. 

 Bearing in mind this acceptation, and leaving out the central figures, Col. Sykes' 

 description of the phenomena would seem to agree with that which I have above 

 given. 



As to the cause of the additional rings, it would seem reasonable to regard them 

 as analogous to the ' interference bows,' which so frequently appear in showers of 

 rain, were it not that on such a supposition (according to received principles) the 

 red colour ought to be the first to re-appear after evanescence, and should therefore 

 be outermost. How the green comes to have that place instead, or why the colours 

 appear separately in the additional bow though blended together in the primary 

 and more brilliant one, is more than I have been able to determine. 



19. On the Temperature of the Interior of a Block of Melting Ice. 

 By James B. Francis. 



The paper by Professor James D. Forbes in the Proceedings of the Royal 

 Society of Edinburgh for April 19, 1858, contains the following description of an 

 experiment by him : 



' Water being carefully frozen into a cylinder several inches long, with the bulb 

 of thermometer in its axis, and the cylinder being then gradually thawed, or 

 allowed to lie for a considerable time in pounded ice at a thawing temperature, 

 showed also a temperature decidedly inferior to 32°, not less, I think, than 0-35° 

 Fahrenheit.' 



In this experiment it appears that the formation of ice on the bulb of the ther- 

 mometer may have acted upon it mechanically and affected the accuracy of the 

 results. To avoid error from this cause I procured a block of clear ice from 

 Merrimack River, and cut it to fit loosely into a cask, the dimensions of the block 

 when fitted being seventeen inches in diameter and sixteen inches high, the dia- 

 meter of the block of ice being about two inches less than that of the cask. The 

 space not occupied by the ice was filled with water from the same river, allowing 

 the ice to float. A hole, one and three-fourths inches in diameter, and ten inches 

 deep, was bored in the axis of the block, and a cover of ice closely fitted to it. 



Two thermometers, No. 1 and No. 2, of similar form, by Seifort of Boston, 

 were used. They had two bulbs with stems, about 3*25 inches long between them, 

 graduated from 29° to 35° Fahrenheit ; the length of a degree in No. 1 being 0'405 



1884. u u 



