Joty—Identification of Crystals by the use of Birefringence. 493 
erystal originally showed a white, suppose; it now exhibits a clear 
and pronounced yellow. If the question was between orthoclase 
and quartz, the decision is that the mineral is quartz. Measure- 
ment of the thickness of the section is, in such a case, hardly 
required, for a section of orthoclase giving a distinct yellow must 
possess a thickness of over 0:05 mms., and half of this thickness 
is more than is usually conferred upon good rock-sections. Gene- 
rally, too, undoubted felspars or other minerals will be present, 
observation of which in the same manner will settle the question. 
A more accurate mode of using this method is to measure the 
retardation in each case with the comparatewr. 
The illuminator may be left permanently in position upon the 
microscope if provision is made for the withdrawal of the reflecting 
prism when its use is not required. The mirror, which need not 
‘be more than 5 or 6 millimetres in diameter, might conveniently 
be carried in a sliding piece fitted in the stage of the telescope. 
Or a half mirror only is used (or one perforated centrally), so that 
part of the field is illuminated by once-transmitted, part by twice- 
transmitted light. This may be left in position during most 
examinations, the tints due to double thickness being compared 
with those of single thickness by a small movement of the slide; 
or even when obtaining simultaneously in the one crystal, one half 
of the crystal being brought over the mirror, while the other half 
is illuminated by light transmitted from beneath. 
Alternative dispositions of the polarized illuminator of course 
also suggest themselves. Thus the small loss of phase due to ° 
elliptic polarization upon reflection of the beam may be obviated 
by placing the reflector above or in advance of the polarizer, 
locating both in the wider body of the microscope above the objec- 
tive. This, however, is hardly required. A construction which 
would confer more perfect verticality upon the incident ray would, 
however, be advantageous as enabling examination by this method 
to be extended to more minute crystals. Perfect verticality would 
be obtained by the use of a semi-transparent mirror placed in the 
axis of the microscope and at 45° with this axis; this mirror re- 
flecting the polarized beam, and transmitting the ascending beam. 
However, a disposition which secures a degree of verticality 
sufficient for the examination of very fine-grained rocks consists in 
applying a construction similar to that figured in the text at a 
