1 '.'•_' F. E. Wright — Tranxmixsion of Light through 



of the waves. If the plate he placed in the position of total 

 extinction for one of the edges in the image under crossed 

 nicols, the edge produced by the second wave, and with it the 

 entire plate, can then be extinguished by rotating either the 

 analvzer, in case the second wave enters at uniradial azimuth, 

 or the polarizer, in case the line of vibration of the emergent 

 first wave is contained in the extinguishing plane of the 

 analyzer. This method of first determining the position of 

 extinction for the one wave and then rotating the polarizer 

 was adopted in many of the above measurements, as the edge 

 produced by the second wave generally extinguished imperfectly 

 because of internal reflexions. Usually, however, internal 

 reflexion did not extend across the entire plate ; the area free 

 from it was then observed to became completely dark for a 

 certain position of the polarizer. In this case the nicols were 

 not crossed but included an angle which differed from 90° by 

 exactly the angle of rotation which the entering, plane-polarized 

 waves suffered on transmission through the two boundary 

 surfaces of the crystal plate. The waves entered at their uni- 

 radial azimuth and emerged vibrating in the extinguishing 

 plane of the analyzer, and were therefore totally reflected in 

 the nicol. From lines 3 and 4, Table II, it is evident that the 

 total amount of rotation (e u — 8') of the plane of polarization 

 is the same for both the ordinary and the extraordinary wave. 



In case the crystal plate is tilted so that the optic axis makes 

 only a small angle with the line of propagation of the refracted 

 waves, it is not possible to obtain satisfactory total extinction. 

 The entering waves, although approximately parallel, are never 

 precisely so, and for a slight difference in direction of pro- 

 pagation, a considerable variation in the azimuth of the 

 planes of polarization of the transmitted waves results. This 

 effect is most pronounced near the optic axis 1 but it is still 

 perceptible on refracted waves, making angles of 20° or even 

 25° with the optic axis. In the table these positions of im- 

 perfect extinction are indicated by an asterisk. On line No. 6 

 of the table a series of readings on a second cleavage plate is 

 recorded, each angle listed being the average of 20 readings, 

 10 for the position +i, + co, and 10 for the position +i, — w. 



In the measurement of these uniradial azimuths, the observer 

 was unfortunately unable, because of other duties, to finish 

 each series of readings as rapidly as could be desired, a month 

 or more often intervening before a series, once begun, was 

 finally completed. This gave opportunity for the effects of 

 surface films to enter the problem, which, as noted by Drude, 

 may seriously affect the accuracy of the values obtained. It is 



1 Compare W. Voigt, Wied. Ann. d. Phys. u. Chemie (4), xviii, 676, 1905. 



