350 Prof. S. P. Thompson on a New Polarizing Prism. 



of balsam, and that the extraordinary ray is transmitted 

 because the extraordinary index of refraction is less than that 

 of balsam. Neither of these statements is completely true. 

 All that its inventor claimed for the Nicol prism*, and all that 

 it actually performs, is as follows : — The critical angle of total 

 reflexion being different for ordinary and extraordinary rays, 

 the ordinary ray is totally reflected and thrown out of the field 

 at an incidence at which the extraordinary ray is still trans- 

 mitted, the available field of polarized light being the region 

 between the points where the extraordinary ray itself vanishes 

 by total reflexion and the ordinary ray enters by lack of total 

 reflexion. The former limit is in all ordinary Nicol prisms 

 marked by a broad blue iris or band of colour ; the latter is 

 delimited by a curved band at the opposite side of the field, in 

 which, amidst a prevailing line of red and orange, a system of 

 interference-bands can be seen. The existence of these inter- 

 ference-fringes was examined by the author in 1877, in a 

 paper which appeared in the Proceedings' of the Physical 

 Society of London, vol. ii. p. 157. In the Foucault prism a 

 similar limitation of the field occurs, interference-fringes being 

 visible at both limits. 



3. The refractive index of balsam for light of mean refran- 

 gibility may be taken as 1*54, that of the ordinary ray in 

 calc-spar as 1'66, that of the extraordinary ray as 1*487. The 

 reciprocals of these are very nearly in the respective propor- 

 tions of 65, 67, 60. The extraordinary index, however, is 

 1*487 only for rays at right angles to the crystallographic axis, 

 having there a minimum, and increasing up to 1*66 for rays 

 whose direction coincides with that of the axis. The ellip- 

 soidal wave-surface of the sheet of extraordinary waves lies 

 partly without and partly within the spherical wave-surface 

 for Canada balsam, while the spherical wave-surface of the sheet 

 of ordinary waves lies wholly within. Hence total reflexion 

 may occur for the extraordinary as well as for the ordinary 

 rays ; but of the extraordinary rays only those can suffer total 

 reflexion which are situated in such a direction with respect to 

 the optic axis that their corresponding portion of the ellipsoidal 

 wave-surface lies within the spherical wave-surface for balsam. 

 As the Nicol prism is usually constructed, this limit of possible 

 extraordinary total reflexion occurs for ravs (in a principal 

 plane of section) inclined at about 10° to the balsam film, giving 

 rise to the limit of the polarized field marked off by the blue 

 iris before mentioned. 



* See Edinburgh New Philosophical Journal, 1828, p. 23, W. Nicol, 

 "Ona Method of so far increasing the Divergency of the two Rays in 

 Calcareous Spar that only one Image may be seen at a time." 



