Compounds by homogeneous X-radiation. 187 



Conclusions. 



It has been shown (Barkla & Simons, Phil. Mag., Feb. 

 1912) that the relative ionizations produced by equal ab- 

 sorptions of Rontgen rays in various gases are the same 

 (within limits of experimental error) as those produced by 

 total absorption of corpuscular radiation excited by X-rays. 

 This important result indicated the possibility of ionization 

 by X-rays being due entirely to the resulting corpuscular 

 radiation — a conclusion which was indeed stated more de- 

 finitely by Bragg. Subsequent work has supported this 

 indication, and the condensation experiments of (J. T. R. 

 Wilson (Roy. Soc. Proc, June 1912) seem to have put the 

 matter beyond doubt. If this is the case, and it is now 

 supported by ionization experiments, it would appear from 

 the present experiments that the real atomic effect is the 

 liberation of corpuscles from the atoms, accompanied, of 

 course, by an absorption of the X-ray beam, which would 

 also l)e " atomic," i. e., it would appear that the number of 

 corpuscles liberated from an atom by a beam of X-rays is 

 the same whether the atom is in combination or not, the 

 ionization produced being a secondary effect, the result of 

 the absorption of this liberated corpuscular radiation by 

 the gas. 



This conclusion is in agreement with the conclusions 

 arrived at as a result of previous work on this subject, viz., 

 " Ionization by Rontgen radiation is approximately atomic, 

 but depends somewhat on the chemical combination of the 

 elements." 



Apparently the corpuscular radiation obeys an " atomic J " 

 law, and the ionization is approximately "atomic" because 

 the " corpuscular factor " is rarely very different from unity. 

 Of course, the present experiments have been limited to 

 elements of low atomic weight, which according to estab- 

 lished Jaws give off only a very weak secondary characteristic 

 X-radiation when exposed to X-rays of ordinary penetrating- 

 power. There is, however, no apparent reason for doubting 

 their complete generality. 



I should like to express my great indebtedness to Professor 

 Barkla whose suggestions and advice have been invaluable 

 throughout the whole of these experiments. 



