Residual Ionization in Gases. 425 



the respective gases, i. e., the drops in ionization and the 

 residual ionization for these gases are about in the same 

 ratio. As the drops in ionization are due to the absence of 

 the earth's penetrating radiation, which is probably of the 

 gamma type, one would have expected the numbers in 

 column 6 to be approximately the same as those in column 3 

 (or column 2, since the ionization under gamma rays is very 

 probably due to secondary beta rays emitted by the walls o£ 

 the electrometer). The differences, however, are consider- 

 able, which would point to the probability of the earth's 

 penetrating radiation being somewhat different in quality 

 from the gamma rays used in obtaining the numbers given 

 in column 3. 



The natural ionizations, with the exception of that for 

 acetylene given in column 5, it will be seen, approximate 

 to the numbers expressing the relative ionizations under 

 beta or gamma r;iys rather than to those obtained with 

 alpha rays. As materials such as zinc, ordinarily classed 

 as non-radio-active, are not known to emit a gamma radia- 

 tion, it would seem therefore that the residual ionization 

 observed with the different gases, with the exception of 

 acetylene, could be ascribed to a beta radiation probably 

 emitted by the walls of the zinc electrometer. 



It will be noted, however, that in the case of each gas, 

 where the number expressing the residual ionization is 

 larger than the number expressing the beta ionization, the 

 number expressing the alpha ionization is also larger. Also 

 when the number expressing the residual ionization is smaller 

 than the number ex] >ressing the beta ionization, then the 

 number expressing the alpha ionization is also smaller than 

 the number expressing the beta ionization. This indicates 

 that though the greater part of the residual ionization in the 

 various gases may be due to beta rays a considerable part of 

 it could be ascribed to a radiation of the alpha type. If, for 

 example, we suppose the residual ionization to be due to 

 alpha and beta rays emitted by the zinc walls of the 

 electrometer, and suppose X per cent, of the ionization 

 to be due to alpha rays, we have for carbon dioxide the 

 relation 



X80-I+ (100- X) L5(3 = 100 x 110-3, 



which gives X = 60 - I. 



Applying this to the other gases we find the values for X 

 which are given in Table IV. 



