268 



PKOF. W. H. BRAGG ON X-RAYS AND CRYSTAL STRUCTURE. 



in the (ill) series by 12 ; the second 34 by 36, the 37 by 12, and the 21 by 100, and 

 so on. The results are indicated by the dots, which, it will be observed, follow closely 

 a regular curve.* We may be said to have allowed for differences in the masses 

 contributing to the various spectra, and to be free to attack the question of the cause 

 of the rapid change in intensity as we proceed to higher orders. 



As far as regards the investigation of crystal structure an explanation is not 

 immediately pressing. We know that it exists, and the law of the decline is not 

 very different in different cases. We can assume it to be expressed by the 

 series 100 : 20 : 7 : 3 ; and we can interpret the variations from the normal values by 

 means of the principle we have discussed above. Finding in this way the distribution 

 of the atomic masses in the various planes, we have data sufficient for the determina- 

 tion of crystalline structure. But it is of course unsatisfactory from a physical point 

 of view to leave the question at this point. 



no 



Fk. 16. 



I think that an ample explanation of the rapid diminution of intensities is to be 

 found in the highly probable hypothesis that the scattering power of the atom is not 

 localised at one central point in each, but is distributed through the volume of 

 the atom. 



We may take the analogy of a diffraction grating. If the transparent portions are 

 very small compared to the opaque, the different orders of diffracted spectra are all 

 equal in intensity. (KAYLEIGH, 'Phil. Mag.,' 1874, XLVII, pp. 81 and 193.) But 

 if the transparent portions are not relatively small, the higher orders tend to be 

 weaker than the lower. And again, if the transparent portions are not sharply 

 defined, but have " fuzzy" edges, the higher orders are much diminished in intensity, 

 as A. B. PORTER has shown by actual experiment (' Phil. Mag.,' Jan., 1905). 



* Note, Jurw 7. I have not been able to account by this principle alone for all the relations between 

 the intensities of the spectra of every crystal hitherto examined : for instance, iron pyrites. But it is 

 most likely that different atoms have different distributions in space and we have not yet taken account 

 of that. Moreover, no case has yet occurred in which the opportunities for comparison are so direct as in 

 calcite, with its different sets of equally spaced planes. 



