602 J. W. EVANS ON THE DETERMINATION OF MINERALS UNDER 



are parallel to those of the nicols and therefore to the cross 

 wires.* 



These positions of darkness are known as extinctions, and they 

 are distant 90 from one another. If the section is exactly at 

 right angles to a plane of crystal symmetry or parallel to an 

 axis of crystal symmetry, in which cases it is at right angles to 

 a plane of optical symmetry, the position of extinction will be 

 identical for all colours and will be characterised by complete 

 darkness. At the same time the crystal outline will usually 

 be symmetrical to the cross wires, which will now indicate the 

 directions of vibration, both in the nicols and the crystal, for all 

 colours. In such cases, the extinction is said to be symmetrical. 



If, on the other hand, the section does not occupy such a position 

 the extinction will be different for different colours, or, as it is 

 usually expressed, is dispersed. Unless the dispersion be very 

 small, there will never be complete darkness with white or other 

 composite light, but it may always be obtained by employing 

 monochromatic light. 



There is usually some difficulty in determining the position 

 of maximum darkness corresponding to the true position of 

 extinction, even where there is no dispersion, or where mono- 

 chromatic light is employed, and resort has been had to various 

 methods of obtaining an exact result. 



One of the simplest of these is to rotate the stage towards 

 the position of extinction alternately from opposite cyclical 

 directions and to note the readings on each side where the same 

 degree of darkness has been obtained. The mean of several 

 pairs of careful observations will approximate closely to the 

 index reading corresponding to the true position of extinction. 



In another method, which has been investigated in detail by 

 F. E. Wright, | the crystal is first placed in the approximate 

 position of extinction obtained in the manner already described, 

 and then one of the nicols is rotated through a small but definite 

 angle and the degree of illumination that results is carefully noted. 

 The nicol is next rotated in the opposite direction through 

 exactly the same angle on the other side of its normal position. 



* According to one view of the direction of the vibration of light 

 in crystals this is not strictly true ; it is, however, in any case sufficiently 

 accurate for practical purposes. 



t Am. Journ. Sci., Series IV., vol. xxvi., 1908, pp. 349-3G8, 379. 



