On the Spontaneous lonisation of Gases. 



281 



It will be noticed that except at the higher pressures the leak is 

 proportional to the pressure. The falling off from this law at the 

 higher pressures might be taken as indicating that the ionisation is 

 due to radiation from the walls of only moderate penetrating power ; 

 it is possible, however, that at the higher pressures a sensible pro- 

 portion of the ions recombine, so that the current is not a measure 

 of the ionisation. There can be no doubt that, for the range of 

 pressures for which the leakage is proportional to the pressure, we 

 may consider that the saturation current was attained, and the rate 

 of leak may be taken as a measure of the ionisation. It is for 

 such pressures that the relative ionisations have been calculated. 

 They are collected together in the table which follows. 



If we except hydrogen, the ionisation is seen to be very nearly 

 proportional to the density of the gas. The deviations in the case of 

 the other gases in fact hardly exceed what might be attributed to 

 experimental errors, for variations amounting to as much as 10 per 

 cent, occurred in the determination of the rate of leak for a given gas. 

 The departure from the law in the case of hydrogen cannot be attri- 

 buted to experimental error, nor is it due to want of purity of the gas, 

 as is shown by the agreement in the results obtained with the gas 

 prepared by different methods. 



The results obtained for the relative ionisation occurring spon- 

 taneously in all the gases tried, including hydrogen, show a remarkable 

 similarity to those obtained by Strutt* in the case of the same gases 

 when under the action of the Becquerel rays, as the following table 

 shows (p. 282). 



It was pointed out by Strutt that the conductivity under the action 

 of the rays is nearly proportional to the density of the gas, hydrogen, 

 however, being an exception. As will be seen, the results I have 

 obtained for the spontaneous ionisation agree more nearly with those 

 obtained by Strutt for gases exposed to radium radiation than for the 

 same gases under the action of polonium rays. 



An account of measurements of the loss of electricity from a 

 charged conductor suspended in a closed vessel containing various 

 * ' Phil. Trans.,' A, TO!. 196, pp. 507-527. 



