On the Reflexion of Light by Incandescent Surfaces. 177 



Also BF =90° = CI; therefore IF = BC = «; and 



EGD = 180°-IGF = 180°-«. 

 Hence 



cos DE = cos DG . cos EG + sin DG . sin EG . cos EGD 



= cos (90° -B) . cos (90°- C) 



+ sin(90°-B).sin(90°-C).cos(180°-a) 



= sin B . sin C — cos B . cos C . cos a. 



Consequently 



sin 6 . sinc + cosfi . cose, cos A = sin B . sin C — cos B. cos Ccoscr, 



which is Mr. Cayley's theorem. 



I take this opportunity of offering a remark on Professor 

 Challis's " Proof that every Equation has as many Roots/' &c. 

 in your last Number. 



The sentence which commences on page 113 and terminates 

 on page 114, stands thus: — "Hence after eliminating z from 

 the above two equations, we may obtain a possible value of y^ 

 from the resulting equation by methods of approximation." I 

 submit that at present we are not entitled to assume that we can 

 obtain a possible value of y^ from the resulting equation. If it 

 can be shown that the resulting equation is of an odd order ; or 

 if it can be shown that, if of an even order, its last term is ne- 

 gative ; then it will be certain that a value of y'^ can be found. 

 But till one of these cases is proved, I submit that the possibility 

 of finding a value for y^ is not proved. 

 I am, Gentlemen, 



Your obedient Servant, 

 Royal Observatory, Greenwich, G. B. AlRY. 



February 4, 1859. 



XXVII. On the Reflexion and Inflexion of Light by Incandescent 

 Surfaces. By W. R. Grove, Esq., F.R.S. ^c. 



To W, Francis, Esq. 

 My dear Sir, 



ON putting in order some old papei's, I found a manu- 

 script in my own handwriting, and the subject of which 

 I had entirely forgotten ; and it was not until some time had 

 ela])8cd that I could recollect anything about the experiments 

 contained in it. 1 now remember that they were made at the 

 London Institution; and it must have been from ten to fifteen 

 years ago. I have no recollection of the reason why I did not 



