332 POLARISATION. 



view remains dark, then the axis of rotation itself is an axis of the 

 ellipsoid. 



(4.) If, on the other hand, the changes which are observed on 

 revolution in one direction or the opposite are unequal, none of 

 the axes of the ellipsoid lie in the plane of the 

 field of view. The same test is then repeated at 

 the other sectional surfaces, and observation is 

 continued until a surface is found which satisfies 

 the above-mentioned conditions. 



(5.) If one of the axes of the ellipsoid is deter- 

 mined, then the two others evidently lie in a 

 sectional surface perpendicular to it. It is there- 

 fore only necessary to bring this sectional surface 

 into action, and to determine by the known method 

 FIG 189 *ks axes ^ e ^ as ^ c ^7> which are at the same time 



the axes of the ellipsoid. 



We give examples explanatory of the above. A tabular crystal 

 of selenite (Fig. 189), in which the truncation face of the acute edges 

 of the prisms is the predominant one, appears dark in the polarising 

 Microscope with crossed Nicols, if the directions a a and c c (the 

 former making an angle of 50 with the side lines) are parallel with 

 the polarising planes of the Nicols, and consequently brightest if 

 they are brought into the diagonal position. These directions are 

 consequently the axes of the ellipse of elasticity. On rotation 

 round the axis a a, the changes are the same in either direction ; 

 similarly round the axis c c. Both are therefore simultaneously 

 axes of the ellipsoid ; hence the third axis is at right angles to the 

 tabular surface. If we turn the crystal round the left or the right 

 edge, so that the third axis is brought into the plane of the field of 

 view, then this axis can be recognized in a similar manner ; the 

 direction in the field of view at right angles to it is not in this case 

 an axis of the ellipsoid, as is at once manifest on turning the crystal 

 round its transverse axis. 1 



Flat membranes of free vegetable cells act in a precisely similar 

 way. The axes of the ellipses of elasticity which are effective in 

 surface views, always show themselves on rotation to be axes of the 



1 For such observations only small needle-shaped crystals should be used. 

 Larger plates, in vertical position, appear white, or give too high interference 

 colours to enable one to observe the changes caused by the rotation round a 

 horizontal axis. 



