Propagation of Waves in a Resisted Fluid. 525 



ciilnr point of the transparent substance, and u^ the intensity 

 when the wave has traversed any thickness z of the substance, 



we have 



■hz 



'-"..c,-c.(?^)Vc,(?;)'...' 



which formulae show the absorbing power of the transparent 

 substance, and its dependence upon the colour. 



22. It would be useless to attempt any experimental confir- 

 mation of these formulae, because they must be necessarily im- 

 perfect until we find in what manner they are modified by the 

 conditions which must be satisfied at the bounding surface of 

 the transparent substance. The paper in the Cambridge 

 Transactions, vol. viii. p. 3, already quoted, shows the great 

 importance of taking these conditions into account. If my 

 reasoning in that paper be correct, it appears that, what other- 

 wise might pass for a good explanation of absorption, is shown 

 to be really fallacious when the conditions at the surface of 

 separation are taken into account. I have fully investigated 

 the effect of these conditions in the present case, and I find 

 that they introduce some remarkable modifications into the 

 above formulae, which appear to lead to very important re- 

 sults. These I hope to give in my next communication. 



23. The peculiar kind of absorption exercised by a plate of 

 tourmaline may be very easily explained upon the hypothesis 

 of resistance to the vibrations of the aether, as we shall very 

 briefly show. 



If we put for X and Y their values obtained in article 19, 

 the equations of motion are 



, rf^ M din A d^ri 



Now if we suppose the transparent substance to be no longer 

 uniform in all directions, but crystallized, and if we take the 

 axes of coordinates, so that they shall coincide with the axes 

 of symmetry, these equations of motion will still hold; only 

 we must suppose that the /;, q and A in the second equation 

 are different from those in the first, as is quite manifest. Con- 

 sequently, when we integrate each of these equations (just as 

 in article 10), we shall find that the h (which may be called 

 the index of absorption) is not the same quantity for ^ that it 



