﻿390 -Dr. C. V. Burton on 



of the plate C in the manner indicated ; it passes partly 

 around two fixed upstanding pins, and is clamped under the 

 heads of two screws. (\ is a draw-tube in which an eye- 

 piece (not shown) can be inserted. To adjust the vertically 

 of the wire "W, the plate C 2 can be slightly turned in its own 

 plane about the screw 2 as required, after which C 2 , C 3 are 

 screwed home. It will be understood that the plate A is 

 pierced with holes where required for the transmission of 

 light. In the apparatus as actually used, the plate A was 

 clamped to a cast-iron bedplate, at the further end of which 

 the object-glass O and mirror M (fig. 1) were supported. 



9. The micrometer calliper F could be read by estimation 

 to 0*0001 inch ; but for the tests which had to be made, 

 much closer readings were required. These were obtained 

 by means of a radial arm (not shown in fig. 3) which could 

 be clamped at will upon the rotatable barrel of the micro- 

 meter, and the outer end of which rested always against the 

 point of an auxiliary micrometer screw, the arm being 

 approximately horizontal and the auxiliary screw vertical. 

 The range of this extra-fine motion was about eight whole 

 turns o£ the auxiliary screw, the head of which was divided 

 into 100 parts. It was found that 530 of these parts 

 were equivalent to O'OOl inch traverse of the main micro- 

 meter F. 



10. Increased sensitiveness may be secured by means of 

 any device which will increase the brightness of the diffrac- 

 tion pattern as seen under given lateral magnification* This 

 is- always true for homogeneous light, and in the case 

 of illumination by white light the limit where increase of 

 brightness ceases to be helpful has not yet been reached. 

 Accordingly great advantage has been found in using as 

 eyepiece a plano-convex cylindrical lens of crown glass, 

 with generating lines vertical. The (vertical) length of slit 

 illuminated then affects the brightness of the pattern seen, 

 but not its apparent height, which is simply the apparent 

 height of the mirror as regarded from the same view-point 

 without the interposition of an eyepiece. There is, of course, 

 a limit to the increase of brightness which can be effected 

 by merely lengthening the slit ; the limit is attained when 

 the pencil of rays reaching the ere from each point of the 

 mirror extends over the whole height of the pupil, and when 

 a short-focus cylindrical eyepiece is used, the eye being 

 thus brought fairly close to the primary focal plane, the 

 condition is realized ly a slit whose length is equal to the 

 diameter of the pupil. If we take this to be only 3 mm., 

 there should be a 25-fold gain of brightness when an ordinary 



