the Absorption of Rontgen Rays. 757 



without even considering the secondary rays which are 

 superposed on the primary. 



A lew words are also necessary regarding the rays emitted 

 by an X-ray tube. As the substance of the anticathode of 

 an X-ray tube is subject to a very intense X-radiation, it is 

 the source o£ an intense secondary radiation. The primary 

 radiation proceeding towards the surface of incidence of the 

 cathode rays, unless very " soft," excites the secondary radia- 

 tion, which is usually more penetrating to the anticathode 

 and produces greater ionization in the air outside. The 

 primary radiation penetrating further into the anticathode 

 also excites an intense secondary radiation which proceeds in 

 all directions. Half of this is therefore directed towards the 

 surface, and as it usually suffers much less absorption than 

 the primary radiation producing it, it emerges with little loss 

 of intensity and is superposed on the primary beam. 



Even if the primary rays were all produced in the surface- 

 layer of molecules, the secondary radiation would produce 

 ionizations comparable with those produced by the true 

 primary rays : as the primary rays cannot be produced 

 absolutely at the surface, the proportion of secondary rays 

 must increase rapidly with an increase at the depth at which 

 they are produced. Thus the " soft " radiations experimented 

 upon by Mr. Kaye (Phil. Trans. A. vol. 209, pp. 123-151) 

 are what might have been expected from our previous 

 experiments on secondary rays, and many of the properties 

 which he records are the properties of the secondary 

 rays previously published by us. A comparison of some 

 of the absorption coefficients reveals their unmistakable 

 identity. 



In the majority of experiments on X-rays, however, the 

 radiation has already passed through a comparatively thick 

 sheet of glass and has been robbed of a large proportion of 

 the secondary rays. 



These considerations also afford an explanation of the fact 

 that the beam of X-rays proceeding from a " soft " tube is 

 much more completely polarized than one from the same 

 tube when harder. In the former case there are less secon- 

 dary X-rays, as well as less X-rays produced by secondary 

 corpuscular rays. 



They also show why the more penetrating portion of a 

 heterogeneous beam from an X-ray tube is the more polarized. 

 The penetrating portion contains the true primary radiation 

 in larger proportion. 



[Note, — All our experiments on the behaviour of nickel have 

 confirmed our previous conclusion — that in all phenomena 



