910 
MR. A. E. TUTTON ON AN INSTRUMENT FOR GRINDING SECTION- 
tightly down. Care should be taken that the slight space between the two flanges is 
equal all round. The whole arrangement is then again suspended from the axis of 
the instrument, and the grinding proceeded with until the section is sufficiently thin 
to exhibit the interference figure in convergent light. This may be ascertained without 
any disturbance of the adjustment by removing the holder from time to time from the 
axis, unscrewing the little nuts, taking out the glass disc carrying the section, and 
examining it upon the stage of the polariscope, or better, between the polarizing and 
analyzing tubes of the axial angle goniometer which is to be actually employed in 
measuring the axial angle. Gi'inding should cease when small rings are clearly visible 
round the hyperbolic brushes. When this is ascertained to be the case, the apparatus 
enclosing the section is again put together and replaced at the end of the axis, the 
grinding disc is removed, and the parallel surface well polished by use of the polishing 
disc. Provided care had been taken while cementing the crystal that the surface of the 
glass disc and the polished artificial surface of the crystal were truly j)arallel, that 
is, only separated by a very thin film of cement of equal thickness, the second ground 
surface will be truly parallel to the first. If the crystal is one of the first type, the 
parallelism can be verified while the holder is in position (after removal of the oil by 
a silk handkerchief) by observing whether the images of the slit of the collimator 
reflected from the faces on the edge of the section are symmetrical to the horizontal 
cross-wire of the telescope. 
The thinness of the sections which can be thus prepared is, of course, limited by 
the tenuity of the annulus which supports the glass disc in the holder ; as the lattei 
is made so thin it will rarely happen that the double refraction is so powerful that a 
section cannot be ground sufficiently thin to exhibit small rings in convergent light. 
Whenever such is the case, however, the difficulty can be overcome by cementing the 
glass disc directly on to the truly plane surface of the thick disc. 
Sections prepared in the manner which has no\v been described will never fail to 
exhibit the interference figures precisely in the centre of the field of the polariscope 
For the purposes of the measurement of the separation of the optic axes the crystal 
plate may be unmounted from the glass disc, if desired, by dissolving off the cement 
with benzene or other solvent which does not attack the crystal. The highly- 
polished section may be conveniently cemented, at a suitable point about its edge, by 
means of a little marine glue (or other cement which resists the action of the highly- 
refracting liquid, a-monobromnaphthalene, in which the crystal is to be immersed for 
the measurement of the angle 2Ha or 2Ho), to the end of a small rectangular strip of 
th.in glass, wdiich can be held in the spring holder of the axial angle goniometer. 
