DETERMINATION OF THE AXES OF ELASTICITY. 337 



or the mean axis. But since sections cannot be made through the 

 crystal in the direction of the principal planes, rotation upon the 

 two horizontal axes is the only method of testing applicable here. 

 The observer must bear in mind the following theoretical deduc- 

 tions : (1.) If the third is the major axis, then by rotation upon 

 a a a position must be reached in which an optic axis coincides 

 with the direction of the transmitted light. The interference 

 colours must therefore descend rapidly, becoming black in the 

 given position, and, on further rotation, again rise. (2.) If the 

 third is the minor axis, the same changes must take place by 

 rotation upon c c. (3.) If the third axis is the mean one, the 

 colour cannot in any case descend to black. The effective ellipse 

 of elasticity becomes less excentric by rotation upon a a as well 

 as upon c c, though without ever becoming a circle. But since 

 the longer path which the rays of light describe in the inclined 

 plate of crystal necessarily increases the difference, this influence 

 can become preponderant under certain circumstances, and can 

 produce a rise of the colours notwithstanding the slight excentricity 

 of the ellipse. 



We learn from observation that the colour of a plate of selenite 

 rises by rotation upon a a, and falls by rotation upon c c. This 

 latter circumstance proves that the third cannot in any case be 

 the major axis ; but whether it is the minor or the mean one can 

 hardly be determined with certainty from its total action, since 

 the inclination of the transmitted rays can only be increased to 

 a certain limit, owing to the refraction at the surface, the direction 

 of the optic axis therefore being possibly not reached. We should 

 have to carry out the rotation in an approximately homogeneous 

 medium, such as oil, and moreover investigate the rise and fall 

 of the colour more exactly in order to arrive at certainty upon this 

 point. 



1 If we rotate a uniaxial crystal of suitable shape upon the horizontally 

 placed optic axis, the ellipse of elasticity remains unchanged, and the rising of 

 the colour is due merely to the increased length of the path in consequence of 

 the inclination. The comparison of this change in the colour with that of an 

 unknown crystal must therefore show whether the difference of path in the 

 latter increases in higher or lower ratio, from whence it is at the same time 

 decided whether the excentricity of the ellipse of elasticity increases or 

 decreases during rotation. In this manner we learn with tolerable certainty 

 that the third axis in the selenite is the mean one. In general, however, the 



Z 



