224 Prof. Barkla and Miss J. G. Dunlop on the 



The first two conclusions find support from other reasoning, 

 and are becoming generally accepted. There is, however, 

 no evidence in support o£ the tentative conclusion regarding 

 the heavier elements. Indeed, we have never considered 

 the validity of the theory as applied to these elements to 

 have been established ; for the results were based upon 

 isolated observations which did not show that the necessary 

 conditions were fulfilled. It seemed possible that the simple 

 theory of scattering was not applicable to the passage of 

 radiation of the particular wave-length used in these experi- 

 ments through the heavier elements. It was therefore 

 decided to study the effect of a considerable variation of 

 wave-length on the scattering. The following investigation 

 was undertaken in order to obtain further experimental 

 data on the subject. 



There is much greater experimental difficulty in determining 

 the scattering from these heavier elements than from the 

 lighter, because under ordinary experimental conditions 

 the scattered radiation from each of the heavier is accom- 

 panied by a much more intense fluorescent (characteristic) 

 X-radiation which is liable to mask the true scattering effect. 

 The idea of using homogeneous beams of X-rays as primary 

 radiations was soon abandoned, as after the elimination of 

 all fluorescent X-radiation for the complex secondary set up, 

 the intensity of the purely scattered radiation was too small 

 to measure with any accuracy. Instead of this, a primary 

 beam from an X-ray tube was used under conditions which 

 limited the range of w x ave-length employed. The beam of 

 radiation was one giving a continuous spectrum between 

 narrow limits of wave-length, — at least the effective portion 

 of the beam was that between such limits. 



Two methods were employed in examining the radiation 

 scattered from copper, silver, tin, and lead. 



Method (a). 

 In one, the scattered radiation from a thin sheet of 

 substance was compared with that from a thin sheet of 

 aluminium subsequently placed in the same position in the 

 primary beam. The intensity of the primary beam was 

 standardized by one electroscope and that of the scattered 

 radiation measured in a second electroscope. After correction 

 for the absorption of the primary and scattered radiations in 

 these scattering sheets — corrections which in this case were 

 not laro-e, — the relative intensities of radiation scattered in a 

 direction perpendicular to that of propagation of the primary 

 radiation from equal masses of, say, copper and aluminium 



