OPTICAL EXAMINATION OF MINERALS 1079 



Inferences may be drawn from the presence or absence of this on the 

 surface of easiest cleavage in a flake. In a slice from a rock the 

 minerals may be cut in any direction, and are often too small for 

 proper study ; nevertheless important inferences may be drawn from 

 the shadows seen to sweep over them as the stage is rotated between 

 crossed nicols. 1 Even if only parallel rays be used, with the ordinary 

 apparatus, minerals often may be identified with practical certainty 

 from their optical characters. Minerals of the regular system, like 

 colloids, being isotropic, 2 produce 110 effect on the polarised rays, and 

 thus remain dark between crossed nicols. So do all slices cut from 

 a uniaxial mineral perpendicular to the principal axis (that of 

 symmetry), for they are isotropic to light passing in that direction. 

 The same property exists in all biaxial minerals in two directions 

 (called the optic axes). But in passing through slices cut in any 

 other directions from doubly refracting minerals, the polarised ray is 

 divided into two rays, vibrating in directions perpendicular to each 

 other and coincident with three lines called the axes of elasticity, 

 i.e. the directions of greatest, least, and mean elasticity. When the 

 slice is turned into such a position that two of these correspond 

 with the vibration planes of the crossed nicols, it becomes dark. If 

 extinction (of light) occurs parallel with the trace of a pinacoid or 

 prism face (or with a corresponding cleavage plane) in a section 

 through the vertical axis, or with the trace of the former in a section 

 perpendicular to it, this is called ' straight extinction,' but if not, it 

 is said to be oblique. Thus in a uniaxial crystal every slice cut 

 parallel with the principal axis gives straight extinction. In the 

 orthorhombic system, the axes of elasticity correspond with the 

 crystallographic axes, so minerals belonging to it also extinguish 

 straight. In the moiioclinic system the orthodiagonal axis is an axis 

 of elasticity, hence the extinction angle is at a maximum in clino- 

 diagonal sections, and is zero in the zone containing the ortho- and 

 liasal pinacoids. In the triclinic system there is no relation between 

 the two sets of axes. Of this system, however, oscillatory twinning, 

 producing alternately banded colours, is a frequent characteristic. 

 Measurements of the extinction angle are of much value for dis 

 tinctive purposes. Thus a rhombic pyroxene can at once be dis- 

 tinguished from a moiioclinic by its straight extinction. 3 Again the 

 maximum extinction angle in a hornblende falls short of 20 ; in an 

 augite it may exceed 40. The magnitude of this angle is affected 

 by changes in the chemical composition of a mineral : for instance, it 

 is very small in soda-hornblendes, such as glaucophane and riebeckite. 

 It varies in the felspar group, and is very useful in distinguishing 

 the several species. 4 But as the minerals in a rock-section seldom 

 chance to lie in the right positions for accurate measurement, better 



1 See for a full account of this, with illustrations, F. Fouque and M. Levy, Minera- 

 logic Micrograph i-qiie, 1879, pp. 101-3. Also F. Rutley, Rock-forming Minerals, p. 84. 

 ; That is, having the ether equally elastic in all directions. 



3 Obviously, more than one observation is needed, because, as intimated above, a 

 moiioclinic mineral, if cut in certain directions, also gives straight extinction. 



4 Levy, Determination drs Felspatlis (1894) p. 31. Summaries of results will be 

 round in Rutley, Hock-forming Minerals, pp. 204, 221, and Cole, Aids in Practical 

 Geology (see ' Felspar' for the references). 



