292 SCIENCE PROGRESS 



he became Wheatstone Professor of Physics at King's College, 

 London. This he resigned in 191 3 to take up his present 

 position as Professor of Natural Philosophy at Edinburgh. 

 During the whole of the time, wherever he has been, there has 

 issued a steady stream of X-ray research from him, working 

 alone or in conjunction with his students. Some account of 

 his work can best be given by a short description of the pro- 

 perties of the various kinds of secondary radiation excited by 

 X-rays. 



This is of three types — scattered, characteristic or fluorescent, 

 and corpuscular. The scattered radiation is the main pro- 

 duct formed when X-rays fall on substances of low atomic 

 weight (up to sulphur). In quality it resembles very closely 

 the primary radiation, and its quantity is proportional to the 

 mass of the radiator and independent of its nature. On the 

 ether pulse theory of X-rays it is due to radiation from the 

 accelerated electrons set in motion by the passage of the 

 primary pulse over them. Accepting this view, Barkla calcu- 

 lated the number of radiating sources (electrons) per atom in 

 the radiator (1904). Owing to the inaccuracy of the values 

 of the charge and mass of an electron then accepted, the num- 

 bers obtained were too high ; but using modern values for e 

 and e/m, the original X-ray measurements show that the 

 number of electrons per atom is about half the atomic weight, 

 except in the case of hydrogen, where it is equal to the atomic 

 weight (191 1 ). 



By investigating the amount of the scattering in various 

 directions in the plane perpendicular to the primary ray, 

 Barkla was able to show (1904) that the X-rays from a cathode 

 tube were partially polarised, and later (1906) that the scat- 

 tered rays from carbon were much more nearly completely 

 polarised. This was of great importance in confirming the 

 ether pulse theory of X-rays and showing their similarity to 

 light. The question of the nature of the rays was, however, 

 by no means settled, for many of the relations between the 

 energy of the original cathode ray, the X-ray produced by it, 

 and the corpuscular ray to which it in turn gives rise, are most 

 easily explained by assuming that the X-rays are corpuscular 

 in nature. Bragg suggested (1907) a special form of particle 

 theory which was able to explain the polarisation of the rays, 

 but Barkla proved that, if care is taken to work with purely 



