308 WRIGHT: INTERFERENCE FIGURES IN OBJECTIVES 



From equation 2 we find by forming the first and second 

 differential quotients after <p, that for a given § the intensity 

 of illumination is a minimum when ^ = 90° + §. This is also 

 evident from figure 1 in which the relations appear so clearly 

 marked that their application to the case of a system of the steeply 

 inclined lens surfaces in a high power objective involves no diffi- 

 culty except that of tedious computation in following each ray 

 through the lens system. Such a computation is, moreover, 

 unnecessary because we know the approximate inclinations of the 

 different lens surfaces for the different parts of the field, and 

 can estimate the approximate rotation of the plane of vibration 

 of the transmitted light wave, and can deduce therefrom the 

 resultant color. The results of such an estimation agree well 

 with those of observation. It seems, however, unnecessary to 

 present the details of this part of the solution of the problem as no 

 new principle is involved and the final result can be inferred at 

 once by analogy from the results already given. 



From the foregoing it is evident that, in weakly birefracting 

 minerals, the color phenomena produced in the objective alone 

 on insertion of the sensitive tint plate tend to veil and to render 

 uncertain, to a greater or less extent, the interference colors due 

 to the mineral plate under observation; special care should be 

 taken in such instances to test by other methods any inference 

 regarding the optical character of the mineral under test. 



It is also evident that the rotation, by the lens system, of the 

 plane of vibration of transmitted light waves has an effect on 

 the measurement of the optic axial angle of a crystal plate but 

 this effect is usually not sufficiently large to interfere seriously 

 with the accuracy of the results obtained, which at best are not of 

 a high order of precision because of a number of factors which enter 

 into the problem and over which the observer has little control.'* 

 These have been discussed elsewhere by the writer and need not be 

 repeated here. Suffice it to state that the weak uniaxial inter- 

 ference cross observed between crossed nicols in high power 

 objectives is inevitable and also the apparent optically positive 



* Carnegie Inst, of Washington, Pub. 158: 147^200. 1911. 



