134 THE ROYAL SOCIETY OF CANADA 



In the third column, corrections have been made for absorption 

 of both primary and secondary rays in the radiator. From the 

 previous results, we know that the greater part of the secondary rays 

 have been softened. The relative intensities per unit mass of the 

 secondary radiations should, therefore, bear some relation to the 

 number of secondary j8-rays formed per unit mass of the radiator. 

 As a first approximation, we may take the number of jS-rays to be 

 proportional to the mass absorption coefficients. We find, however, 

 that the relative intensities per unit mass of the secondary radiation 

 bear no relation whatever with these absorption coefficients but merely 

 show a gradual increase in intensity as the atomic number of the radi- 

 ator is increased. 



This gradual increase has been explained on the ordinary theory 

 of scattering, electrons being closer together in an atom the higher 

 the atomic number. As the scattered waves from neighbouring elec- 

 trons re-enforce each other, we should get a greater intensity per unit 

 mass of scattered radiation from elements of high atomic number than 

 we do from those of low atomic number. This point of view is also 

 borne out by the fact that, as we decrease the angle between primary 

 and secondary rays, the relative intensities per unit mass of the 

 secondary X-radiation show a larger rate of increase with atomic 

 number than that given in column 3 of the above tables. For example, 

 for an angle of 45°, the approximate numbers are paper 100, aluminium 

 160, copper 380 and tin 800. 



It is, therefore, concluded that the observed change in frequency 

 and consequent softening of secondary X-rays is not due to their 

 being formed by secondary (S-rays, and that every electron which 

 the primary beam may influence in passing, plays a part in this 

 phenomenon. A greater knowledge of the experimental facts relating 

 to this very important question seems to be required before a satis- 

 factory explanation of it can be given. 



For the work carried out at University College, London, a grant 

 was received from the British Scientific and Industrial Research 

 Council, and for that carried out at McGill University, one from the 

 Honorary Advisory Council for Scientific and Industrial Research. 



The writer has much pleasure in expressing his thanks to Sir 

 William Bragg for placing all the laboratory facilities at his disposal. 



McGill University, 

 Montreal. 

 June 10, 1922. 



