382 Sir David Brewster on the Compensations of Polarized Light. 



zation towards a plane perpendicular to that of refraction, and the subsequent 

 reflexion at n has exactly counteracted that rotation, by turning back the planes 

 as many degrees towards the plane of reflexion. The reflexion at n has, there- 

 fore, brought back the ray ra into the same state as the original ray Rr, that is, 

 the ray nm' is common light. 



In order to ascertain if this principle is general, and to determine the laws 

 of the compensation of partially polarized light, I shall now describe the instru- 

 ment by which I have ascertained the physical condition on which compensation 

 depends, and the leading facts on which the doctrine rests. From its property 

 of measuring degrees of polarization, I have called this instrument a Polarimeter. 

 It is represented in Fig. 2, and consists of two parts, one of which is intended to 

 produce a ray of compensation, having a physical character susceptible of nume- 

 rical expression, and the other to produce polarized bands, or rectilinear isochroma- 

 tic lines, the extinction of which indicates that the compensation is effected. The 

 first part of the instrument consists of a goniometer ab, carrying on its axis mn, a 

 frame cd containing six or seven plates of glass, about the 70th of an inch thick, 

 such as are now used for holding microscopic objects. This frame can be taken 

 off and replaced by a black glass reflector highly polished, and free of all oxidation 

 on its surface, or it may be fixed permanently at ef, alongside of the frame cd.* 



The second part of the Polarimeter is a combination of two plates of rock 

 crystal, or any other transparent doubly refracting mineral, such as I described 

 in 1819, in my paper on the Properties of Amethyst A The object which I had 

 in view by this combination was to exhibit the colours of polarized light in recti- 

 linear bands, and this is effected in the following manner. A plate of rock 

 crystal, ab. Fig. 3, from the fiftieth to the tenth of an inch thick, Is cut so that 

 its faces are inclined 45° to the axis of the prism, which is the axis of double re- 

 fraction. When the plate has been divided into two equal parts at the line cd, 

 the one is placed transversely above the other, and cemented to it by Canada 

 balsam, so that the two plates act in opposition to each other upon polarized 

 light. When this plate is fixed at the end of a Nicol's prism (or a rhomb of cal- 

 careous spar, with a circular aperture just sufficient to separate the two images), 

 as shewn in Fig. 4, the depolarizing axis of the plate being parallel to the prin- 



* When much light is desired, a plate of a highly refracting substance, whose index of refrac- 

 tion is known, may be substituted for the glass, 

 t See Edinburgh Transactions, vol. ix. p. 148. 



