362 POLARISATION. 



other. The conditions of neutrality are, in general, given in both 

 cases, if the radial axis of the middle elements is taken as equal, 

 and hence their uniaxial constitution assumed. 



The application of these relations to combinations 3 and 4 in 

 the above table, which alone suffices for the prescribed conditions, 

 affords us a means of at once striking out either the first and third, 

 or the second of the possible cases there adduced. 



2. SPHERICAL AND OVAL OBJECTS. 



We must at once remark that these terms refer only to objects 

 whose double-refracting elements are grouped round one centre in 

 more or less concentric layers, where therefore a certain congruity 

 exists between the outer form and the inner structure. Spheres 

 cut from a crystal or from a cylinder would of course act just as 

 these objects themselves, and therefore do not require a special 

 treatment. 



The possibilities which are imaginable in stratified structures 

 with regard to the optical action of the objects fall, on nearer 

 examination of the relations, into two categories, which may 

 always be readily distinguished on revolution of the object during 

 observation. In all possible views that we get on rotation, either 

 the polarisation colours retain the same character and the same 

 order i.e., we always observe the same position of the neutral 

 lines of the acceleration and retardation quadrants or this agree- 

 ment of the different views does not take place, the image 

 exhibits, on rotation of the object, a very marked change of 

 phenomena, which fails to take place only when perchance the 

 axis of rotation coincides with a certain diameter of the object 

 which is distinguished from the other. 



It is apparent that with objects of the first class the double- 

 refracting elements must be situated symmetrically with regard 

 to the position of their axes of elasticity round each radius, and 

 thence round each point of the surface ; while in the latter case 

 such symmetry, just as with a cylinder, is indicated only with 

 regard to that particular direction which may be denoted 

 as the axis. Rotation of the objects under the Microscope 

 therefore discloses to us whether they are to be considered 



