248 Mr. R. T. Glazebrook on Wool's Prism. 



Join B D, and lot C bo the angle of the rhomb through which 

 the optic axis passes. 



A Gr E is a principal plane; and B D is perpendicular to it. 

 Hence B D is a possible direction of vibration in the rhomb. 



Again, B D is clearly parallel to the intersection of the inci- 

 dent wave and the nice A B D. Both the refracted waves 

 therefore cut ABCDina line parallel to B D. Thus B D is 

 the direction of vibration in one of the refracted waves, while 

 that in the other is perpendicular to B D. 



Let us consider that wave in which the direction of vibra- 

 tion is parallel to B D. It cuts the face EFGH in a line 

 parallel to FH or BD; and this line is also parallel to the 

 direction of vibration. 



The emergent wave cuts the face E F G H in the same line; 

 and the direction of vibration in it will be also parallel to this 

 line. The emergent light is polarized, therefore, in a plane 

 parallel to the principal plane of the prism. Had we taken 

 the other wave, the emergent light would have been plane- 

 polarized in a perpendicular plane. 



But now suppose that the incident light is not parallel to 

 A E. The incident wave will no longer cut A B C D in a line 

 parallel to B D. 



Let the wave-front in the crystal cut the face E F G H in 

 a line P Q. 



P Q is not generally a possible direction of vibration. 

 The direction of vibration in the crystal is inclined to P Q at 

 an angle (6 r say); that in the emergent wave will be inclined 

 at another angle (6 suppose). Let </>' be the angle of inci- 

 dence, </> that of emergence ; then, if the medium were iso- 

 tropic, we should have the equation 



tan 0'= tan 6 cos (<£-<£') (1) 



This follows from the electromagnetic theory of the reflexion 

 and refraction of light (Phil. Mag. April 1880, page 290), as 

 well as from Fresnel's. 



Now experiments made recently by myself appear to show 

 that the same law holds very approximately for Iceland-spar 

 when there is only one wave transmitted in the spar*. Thus, 

 assuming this formula to hold, the emergent light is plane- 

 polarized ; but its plane of polarization no longer coincides 

 with that of the light in the prism. The two are inclined at 

 an angle equal to that between the directions of vibration, 



* Since writing the above, I have shown that, on the electromagnetic 

 theory, this equation expresses the condition that one wave only should 

 be propagated in the crystal. 



