Rontgen Radiation from Gases and Vapours. 673 



then the amount o£ secondary radiation given off by the 

 substance under the action of a primary pulse whose energy 

 per unit area is E, is given by the expression 



6 mr 



It is thus proportional to the number of corpuscles per unit 

 volume, and independent of the hardness of the primary rays. 



Since the number of corpuscles in an atom is proportional 

 to the atomic weight, it follows that on this theory the ratio 

 of the secondary radiation per atom to the atomic weight 

 should be constant. The last column of Table III. shows how 

 far this law holds good. For helium, oxygen, nitrogen, and 

 carbon, and approximately for hydrogen, sulphur, and chlorine 

 also, it gives a good agreement with the experimental results. 

 For elements of higher atomic weight and especially for 

 arsenic and bromine, it absolutely fails. 



The simple theory briefly sketched in the preceding para- 

 graphs assumes that the corpuscles contained in the atom have 

 all, at any rate, some degree of freedom, and are able to move 

 independently of the rest of the atom. Any corpuscles 

 which were so rigidly attached as to be incapable of inde- 

 pendent motion, would be practically unaffected by the 

 primary rays, and add little or nothing to the secondary 

 beam . 



It does not seem possible, however, to explain the experi- 

 mental variations in this way. The good agreement given 

 by all the lighter elements with the simple theory seems to 

 show that the secondary radiation from them is produced in 

 accordance with the theory, and that to explain the radiation 

 from such substances as arsenic and bromine, which is of 

 quite a different order from that of the lighter elements, we 

 must look for some other source. 



Rontgen rays, as is well known, ionize the gases through 

 which they pass, and it is possible that in the ionization we 

 have a source of radiation which may assist in accounting for 

 the large amounts of secondary rays given off by the elements 

 of high atomic weight. The disturbances taking place within 

 the atom, which lead to its ionization, and the expulsion of 

 a /3- corpuscle, might well be the source of a considerable 

 amount of Rontgen radiation. The mere expulsion of the 

 /3-corpuscle from the atom would of course give a Rontgen 

 pulse. Unfortunately, however, sufficient data do not exist 

 to test this theory completely. The relative ionizations have 

 only been determined for a few of the gases contained in Table I., 

 and bromine and arsenic are not among them. So far as the 

 figures go, however, there seems to be at any rate a rough 



