DOUBLE REFRACTION IN ISOMORPHOUS SUBSTANCES. 289 



a great extent, a consequence of the dispersion of the optical 

 axes for different colours in different planes. 



In fact, when a thread of white light, which has traversed the 

 crystal either as an ordinary or extraordinary ray, and under a 

 certain inclination, is observed, each of the elementary colours 

 of which it consists is found to be polarized at its emergence in 

 a different plane. 



According to the rule laid down by M. Biot, and confirmed 

 by the theory of Fresnel, this plane of polarization divides the 

 angle contained between the planes which pass through the ray 

 itself and through each of the two optical axes. But in the 

 present case these latter have always a different direction for 

 different colours ; consequently when the analyser is rotated, it 

 arrests or permits, according to the adjustment, the passage of 

 different colours from the light emerging from each point ; and 

 these phaenomena necessarily call to mind the optical properties 

 of quartz, because all the elementar}^ colours of a white ray pre- 

 viously polarized in a similar plane are each polarized in a dif- 

 ferent plane at their emergence from that substance. 



Nevertheless, if the chromatic phaenomena just mentioned are 

 in the case of the tartrates solely the consequence of the disper- 

 sion of the optical axes, they ought to disappear entirely when 

 the plate is illuminated normally by a bundle of perfectly parallel 

 rays polarized in any one of the rectangular diametric planes in 

 which the optical axes are situated. The incident ought then 

 to pass completely, whether as, ordinary or extraordinary ray, 

 and to pass in a direction parallel to the common bisector of all 

 these axes, and its plane of polarization could not be modified 

 by the crystalline plate. This, however, is never the case com- 

 pletely. It therefore becomes a question, whether the develop- 

 ment of the variable colours, which present themselves even 

 when the analyser is rotated, may not be simply a consequence 

 of irregularities in the crystalline structure, rendered sensible by 

 the extreme diminution of the birefractive power in the direction 

 of the bisector, and analogous to those phaenomena which have 

 long been known to occur in crystals with one axis, as for in- 

 stance beryl ; and, on the other hand, whether it may not be 

 owing to the reappearance, likewise by reason of the diminution 

 in the double refraction, of the molecular rotatory power proper 



