﻿450 



The Hon. J. W. Strutt on the Scattering 



These are the component rotations. The resultant in the 

 general case would be rather complicated, and is not wanted for 

 our purpose. It is easily seen to be about an axis perpendicular 

 to the scattered ray, inasmuch as 



CC'CT l + ?/S7 2 + 2^ 3 = 0. 



Let us consider the particular case of a ray scattered normally 

 to the incident light, so that x=0. Denoting for brevity the 

 common factor by p } we have 



An yz 



r n r 2 



AD y 

 V r 



Anz* 

 n r* 



(6) 



®o = 



whence 



-G7 



2,2,2 S/^V ** , 9^\'V 



=«?+«?+ «i=-p\tJ ?+ p xw) 



2 „,2 



Here we have reached a result of some importance and one 

 which can be confronted with fact. For from the value of -cr it 

 appears that there is no direction in the plane perpendicular to 

 an incident ray of polarized light in which the scattered light 

 vanishes, if An and AD be both finite. Now experiment tells 

 us plainly that there is such a direction, and therefore we may 

 infer with certainty that either An or AD vanishes. So far we 

 have a choice between two suppositions; either we may assume, 

 as in my former paper, that there is no difference of rigidity be- 

 tween one medium and another, and that the vibrations of light 

 are normal to the plane of polarization, or, on the other hand, 

 that there is no difference of density between media, and then 

 the vibrations must be supposed to be in the plane of polarization. 

 The former view is the one adopted by Green and (virtually) by 

 Cauchy in their theories of reflection ; while the latter is that of 

 MacCullagh and Neumann, which I now proceed to show is 

 untenable. 



Suppose then that AD = 0. Reverting to the general values 

 of Wj, OTg, «r 8 in (5) , we have 



An yz 

 •* n r z 

 An xy 



An z 2 -x 3 

 P n ~1*~ 



(7) 



