with a Liquid Mercury Anticathode. 



425 



The accuracy is still smaller for the fainter lines. In the 

 case of the fainter lines the error of the microscope reading- 

 was sometimes 10 times more than for the strong lines. 



It has been mentioned that formula (1) had to be corrected 

 owing to the fact that the crystal surface was not exactly in 

 the axis of revolution. The complete equation is 



I 



tan 2x= j- tun 



I J Asm a. bin a fl J , 



L = distance between the axis and the plate, 



d = distance between the crystal surface and the axis. 



The ratio of d/L was always kept small ; the maximum of 

 its value was about '005, but in most of the cases d/L was ten 

 times smaller. 



L was calculated from (1), i. e. by measuring the distance 

 between the symmetrical positions of the standard lines and 

 putting the numerical value of a in the equation, 



d was found by the following procedure. The crystal- 

 holder was rotated until the direct beam of X-rays passed 

 the edge of the crystal. This could easily be observed on a 

 fluorescent screen. Starting from this position, a small to 

 and fro movement was given to the crystal-holder by a 

 revolving cam. The crystal-holder was then rotated through 

 180 and the same operation performed. 



Fiff. 3. 



Ant/cathoo'e 



W<3t<? 



Suppose the axis of rotation lies in the crystal (fig. 3). 

 On each side of the centre of the plate a dark band appeared 

 after the plate had been developed. Between these two 



