308 POLARISATION. 



the end faces. If we now draw through the terminal points /> 

 and d, of the layer of Canada balsam, two lines parallel to 

 these limiting rays and produce them upwards to their point of 

 intersection s, it is evident that if s lies in the field of view 

 A B, this is the only point which is illuminated by the full 

 cone of light. A portion only of this cone of light readies the 

 other points, and this diminishes in inverse ratio to the distance 

 of s. It is also clear that if the plane of adjustment of the 

 Microscope is situated at a somewhat higher level, as is usually 

 the case, the aperture of the cone must be the more reduced for 

 the middle of the field of view the greater its distance from A B. 



The practical result involved in our explanation is that Nicol's 

 prism should be brought as near as possible to the plane of 

 adjustment, since its withdrawal to a distance from that plane 

 reduces the cone of light. But as the optician has also to bear 

 in mind that sufficient space must be left for the insertion of 

 the selenite plates, he cannot exceed a certain limit of proximity. 

 The optically effective portion of the cone of light is consequently 

 reduced to about 20 aperture. A polariser of Beneche gave at 

 the level of the stage in the middle of the field of view 21 30'. 

 Where an aperture of this size is not sufficient, the convergence of 

 the rays must be increased by a condensing lens applied above 

 the Nicol. 



If the polariser is required also to illuminate sufficiently the 

 marginal points of the field of view, the diameter of the end face 

 must not be too small. Since the length of a Nicol prism is 

 almost three times this diameter, such a prism is necessarily 

 costly. 1 Foucault endeavoured to overcome the difficulty by 

 replacing the Canada balsam by a stratum of air. The limiting 

 angle of the total reflexion is thereby reduced to 37 5' for the 

 ordinary ray, and the refractive indices are the most favourable 

 when a (in Fig. 173) = 51, and /3 = 70 52', which latter 



1 Hasert (in Poggendorff's " Annalen," Bd. cxiii. p. 188) states that he has 

 devised an improved and less expensive construction of the Nicol prism ; instead 

 of Canada balsam, he selects a medium whose refractive index is equal to 

 that of the extraordinary ray in the calc-spar. He affirms that the sectional 

 angle is thereby reduced from 90 to 81, which, of course, amounts, with 

 equal bases, to a corresponding shortening of the prism. He also states that 

 the edge of tie polarised zone appears neither blue nor red, but shows only a 

 faint secondary tint, just like a well-corrected objective which points we are 

 unable to follow. 



