PROCEEDINGS OF THE SOCIETY. 127 



any useful size. They belong to the oblique system, and have a plane 

 of very easy cleavage parallel to the flat sides of the crystal. There are 

 other planes of cleavage, or at least of easy fracture, which are probably 

 connected with the formation of the features which are about to be 

 described. 



" If a normal twin-plate — showing, say red — is laid on a rotating stage, 

 between crossed nicols, there will be four positions in which one area (A) 

 will be red and the other (B) black, and at nearly 30° from each posi- 

 tion these colours will be reversed. ( ? 28° 16'.) Between the two 

 areas A and B there will be merely a boundary. This boundary may 

 be quite irregular, perfectly straight, or zigzag with beautiful regular 

 angles. Bat, in addition, this boundary, which is called the twin-plane, 

 may not be perpendicular to the cleavage-plane. The result then will 

 be that the twin-plate is practically made up of two opposed wedges, 

 and if the plate is rotated so that one of these is dark the other will 

 shine out in bands of colours, showing Newton's orders of thin film 

 colours. The angle of this twin-plane may be more or less acute, giving 

 the orders in thin lines or spread into broad bands. Instead of an angle 

 the junction may be a series of steps, or alternate steps and angles. Also 

 these variations may occur lengthwise, and all these in every possible 

 combination. The most striking forms may be classified into broad or 

 narrow wedges, parallel bands, bands cut up into rhomboids, mitred 

 angles, and a very beautiful zigzag form. Specimens of each are shown 

 under the Microscopes. Some beautiful effects result from crossing two 

 wedge twins on each other. Another interesting feature is that whereas 

 the two halves of some twins are at about 30°, all that I have myself 

 obtained from the London clay are at about 75°, so within 7|° either 

 way of cancelling each other, and therefore when wedged they give a 

 nearly dark band with Newton's orders running both ways. Most of 

 the features are best seen by from 1-in. to 2-in. objectives, but some 

 are large enough to show well on the table polariscopes, or projected 

 on the screen. 



" This form of reflecting polariscope, as constructed for me by Messrs. 

 Baker, is a most useful appliance for workers in thin films. The 

 analysing reflector can be used for general observation of large surfaces 

 and for display of finished work, while a Nicol, with low-power lens, is 

 easily substituted for actual exact marking and cutting on the large 

 horizontal glass stage. Also, a mirror and single glass plate converts it, 

 in a moment, into a Norremburg doubler, so useful for gauging the 

 thickness of \- and -i-wave films in mica work. A revolving |-wave 

 mica plate under the stage gives change of colour (plus and minus), or 

 a pair of ^-wave plates would give change by actual rotation of the 

 polarised beam. 



" I have also a small appliance, consisting of a fragment of Iceland 

 spar, mounted on the nose of objective. This gives a double image, 

 and if diagonal cross lines are ruled on a blackened slip on the stage, 

 with a selenite plate, two complementary coloured images appear of the 

 network superposed, and wherever the lines cross, the coloured lights 

 re-combine into white light. 



" Small clear pieces of Iceland spar about ^-in. thick can be selected, 



