Prof. Tyndall on Ice and Glaciers. 91 



rent, a cloud of smoke may be carried off without quickly losing 

 its form. Both the other facts adduced by Buijs-Ballot also 

 admit of simple explanation. The remark made by him at- 

 tached to one of his objections, that the molecules of gas in a 

 room must traverse the room many hundred times in one second, 

 is completely foi*eign to the theory. Perhaps it might be said 

 of a remark which occurs in the mathematical development at- 

 tached to my previous paper, that it afforded reason for such an 

 idea. I assumed there, namely, that the gas was in a very flat ves- 

 sel, and I then assumed that the molecules of gas without disturb- 

 ing one another, sped backwards and forwards between the two 

 great parallel sides. Nevertheless, this assumption was there 

 introduced with the following words : " In estimating the pres- 

 sure, instead of regarding the motion as it really occurs, we may 

 introduce certain simplifications." I believe I thereby called 

 sufficient attention to the fact that this assumption should not 

 serve to furnish an image of the real process, but only to sim- 

 plify the calculation there intended, the result of which could not 

 be thereby changed. 



XI. Remarks on Ice and Glaciers. 

 By John Tyndall, F.R.S. <S)'c.* 



THERE are two or three points connected with ice and gla- 

 ciers to which I intended to refer in a paper now ready 

 to be presented to the Royal Society, but which, on reflection, 

 I think may be more fitly treated in the pages of the Philo- 

 sophical Magazine. 



In the December Number of that Journal thei-e is a reprint 

 of a paper by Prof. W. Tliomson, which first appeared in the 

 Proceedings of the Royal Society : the following points in this 

 paper need clearing up. 



In the last paragraph Professor Thomson expresses the diffi- 

 culty which he experiences in accounting for the resolution of a 

 mass of ice into six-petaled liquid flowers f- That such a diffi- 

 culty should present itself is, I acknowledge, a matter of some 

 surprise to me. The effect, as stated in my paper, is manifestly 

 due to the crystallization of the substance. Water, when cry- 

 stallizing, builds itself up into flowers of this kind. Last August 

 I observed a magnificent example of this six-rayed architecture 

 upon the summit of Monte Rosa; Dr. Scorcsby has given us 

 numerous drawings of polar snow-crystals illustrating the same 

 point; and the drawings of Mr. Glaishcr also illustrate it. Not 

 only do snow-crystals exhibit this structure, but in freezing water 



* Communicated by the Author. 



t Phil. Traus. part 1, 1858, \y\}. 212-213. Phil. Mag. Nov. 1868, p. 334. 



