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



contain alternately sodium atoms only and chlorine atoms 

 only. The result of this unequal loading is that the re- 

 flexions of even order are strongly enhanced in comparison 

 with those of odd order : we are taking d to be the distance 

 from one chlorine plane to the next, in accordance with the 

 definition given above, and the glancing angle of the first 

 order to be sin -1 \/2d. 



When both spacings and loadings are irregular the rela- 

 tive intensities are more complicated still ; as for example 

 in the case of reflexion by the (100) or (111) planes of iron 

 pyrites *. 



Since the relative intensities depend so directly upon the 

 spacings and loadings of the planes, and since the manner of 

 dependence can be easily explained, it follows that experi- 

 mental determinations of the intensities may be used to draw 

 conclusions as to the structure of the crystal in question. 

 This method has been largely used by W. Lawrence 

 Bragg f. 



It is of importance, therefore, that the intensities should 

 be measured with as much accuracy as is possible. It happens 

 that this method of analysis is extraordinarily powerful, and 

 that very rough measurements coupled with general assump- 

 tions have been sufficient to lead, in the paper quoted, to 

 definite conclusions. The measurements which were used, 

 and for which I was myself mainly responsible, could only be 

 considered as provisional, under the circumstances. They 

 were sufficient, however, for the determination of several 

 complicated crystals, such as iron pyrites, the calcite series, 

 and so on. 



For example, in the case of iron pyrites it appeared that the 

 iron atoms were arranged on a face-centred cube, that is to 

 say, at the corners of a cube and at the centres of each face. 

 A sulphur atom lay in each of the eight small cubes into 

 which the large cube could be divided ; it was situated on a 

 diagonal of the small cube which diagonal was to be chosen 

 in a special way, and calculation determined the position 

 as dividing the diagonal in th3 ratio 1 : 4. Ewald has 

 now shown (Phys. Zeit. April 15, 1914) that by the 

 aid of this determination of the structure of iron pyrites, 

 the Laue photographs of the crystal can be completely 

 unravelled, and the structure is therefore confirmed by an 

 independent method. One small change is all that is 

 required. In order to explain the relative intensities of 

 certain of the less prominent spots in the photograph, it is 



* W. L. Bragg, Proc. Roy. Soc. lxxxix. p. 476. 

 t Proc. Roy. Soc. lxxxix. p, 468. 



