Mr. Tovey 07i the Optical Theory of Crystals, 259 



of the cold body. It appears therefore that this substance, 

 from tlie peculiar suddenness with which it changes colour, is 

 well adapted for experiments on the radiation and conduction 

 of heat. 



If now we throw some drops of ammonia on the paper it 

 turns white, and if we hold it to the fire we find that it has 

 lost the power of changing colour. Gradually however the 

 ammonia evaporates, and then the alternations of colour occur 

 as before. This seems to prove that the alkali enters into 

 chemical combination with the iodide, and possibly the white 

 substance is the double iodide of silver and ammonia. This 

 opinion is confirmed by observing that potash and soda act 

 in a similar manner, giving rise to permanent white com- 

 pounds unchangeable by heat, which are probably the double 

 iodides of silver and potassium, and silver and sodium. 



This is the reason why the paper was directed to be washed 

 with a rather dilute solution of hydriodate of potash ; for if 

 we use a concentrated solution the resulting tint is white, and 

 the colour of the paper is not changeable by heat. 



I have kept some pieces of this prepared paper for a year 

 or two, and find that it still remains as sensitive to heat as 

 ever. 



XLI. On Prof essor Sylvester* s Analytical Development ofFres-- 

 neVs Optical Theory of Crystals. By John Tgvey, Esq, 



To the Editors of the Philosophical Magazine and Journal, 

 Gentlemen, 



A S Mr. Sylvester's valuable analytical development of 

 -^ Fresnel's optical theory of crystals, published in your 

 last and present volumes, is based upon those ideas of Fresnel 

 which I conceive to be erroneous, I think I ought to show how 

 it may be founded upon the laws which I have deduced in 

 your 9th volume, (third series) p. 420, et seq. 



For this purpose, then, let a/^ y, s' be the coordinates of a 

 point in the line Oy (fig. 3, p. 428, vol. ix.), the distance 

 of this point from O being unity; then cos x^ Oy = .r', cos 

 y O y = i/') cos z' O z = zK 



a:/2 .^ y2 ^^ 2;'* = 1 . 



Let Ix' + my" -{• nz' = (a.) 



be the equation to the wave-surface, Oy and O z being the 



lines of vil)ration therein ; and suppose the axes of .%\y\ z' 



to be the axes of elasticity. Now, since the expression for 



2D2 



