Reflexion of X Rays by Crystals. 883 



necessary to suppose that the sulphur atom divides the 

 diagonal, not in the ratio 0*2 to 0*8, as W. L Bragg had 

 found, but more nearly as 0*224 to 0*776. 



If, therefore, it has been possible to find so much from the 

 early and rough measurements of intensity, it ought to be 

 possible to do far more if the measurements are made more 

 accurate. 



There are other reasons for attempting to increase the 

 accuracy of the intensity determinations. In the first place, 

 it will be possible in this way to test theories which profess 

 to explain the general process of reflexion of X rays by 

 crystals. And again, it is important to be able to determine 

 to what extent the intensities are dependent on the tempera- 

 ture of the crystal. The thermal movements of the atoms 

 should affect the reflexion ; indeed, Debije has calculated 

 the extent to which they should do so *. Experiment should 

 therefore be of great use in this direction also. 



For these various reasons I have endeavoured to increase 

 the degree of accuracy with which intensities can be com- 

 pared by means of the spectrometer. This paper contains a 

 brief account of the method which appears to be the most 

 suitable, and gives some results of its application to rock- 

 salt and to the diamond. Although much greater accuracy 

 is no doubt still obtainable, the determinations already made 

 seem capable of being applied usefully. 



Let us first attempt to give a definite meaning to the term 

 intensity of reflexion. 



Imagine a fine pencil of homogeneous f X rays to proceed 

 from a source at A, to be limited by a slit at B, and to be 

 incident upon a crystal C, so placed that a reflected ray is 

 formed which passes into the ionization chamber D. Let 

 the crystal be so adjusted as to give the maximum value 

 of the reflected ray ; and let the ionization current in D 

 be measured in the usual way, the current being allowed to 

 run for a stated time, say t seconds. 



We may now remove the crystal and turn the ionization 

 chamber round until the primary beam enters it directlv : 

 and we may again measure the ionization current for a 



given time 



The comparison of these two quantities so obtained does 



* Verh. cl Deutsch. Phys. Ges. xv. pp. 678, 738, 857 (1913) ; also 

 Ann.d. Phys. 1914, p. 49. 



f Homogeneous pencils have been used in these experiments because 

 they are definite, and can be readily obtained intense and pure. 



