886 Prof. W. H. Bragg on the Intensity of 



We may now compare the method of the revolving crystal 

 with the methods by which the earlier measurements of 

 intensity were made. The method then used was to move 

 crystal and ionization chamber together, the latter at twice 

 the rate of the former. In this way a " spectrum " was 

 mapped out, in which the occurrence of " peaks " marked 

 the existence of homogeneous pencils of greater or less 

 strength ; and the height of the peak was taken to repre- 

 sent the strength of the reflexion, and used as a means of 

 comparing the relative strengths of various orders of re- 

 flexion. This method would be as good as the other if 

 crystals were perfect. But they are not. Rocksalt, for 

 example, is very far from being uniform : it consists rather 

 of an agglomeration of smaller and more perfect crystals put 

 together in imperfect alignment. In consequence, a pencil 

 of X rays passing through a fine vertical slit and falling on 

 a vertical crystal face, is not reflected simultaneously at all 

 points of the narrow vertical band along which it meets the 

 crystal. It is reflected first in one part, then in another as 

 the crystal turns round, and one piece of it after another 

 presents itself to the rays at the proper angle for reflexion. 

 Since reflexion does not occur within close limits but is- 

 spread over a wide range, the highest point of the peak 

 is much lower than it would be if the crystal were perfect,, 

 and is no true measure of the intensity of the reflexion. 



For example, the curves of fig. 2 show the result of 

 measuring the intensity of the reflexion of the first order 

 spectrum in the (100) face of rock- 

 salt, the incident rays being limited to Fig. 2. 

 a fine pencil. The abscissae represent 

 the angular positions of the crystal, the 

 ordinates the measured ionization cur- 

 rents. It will be observed how very 

 irregular the curves are ; and still more 

 remarkable is the difference between 

 them, for they were obtained from 

 neighbouring portions of the same face. 

 Clearly no measure of intensity is to be 

 obtained from the maximum ordinate 



of either of these curves. The areas of 5° 6° 



the curves are nearly the same, how- 

 ever. The specimen of rocksalt which was used was very 

 irregular. It is easy to find much better. 



If a wide pencil of rays is used the results of irregularities 

 are far less obvious. Portions of the crystal lying on the 

 narrow vertical band above mentioned may not be reflecting 



