Disintegration in a Vacuum Tube. 425> 



cathode-fall (abscissae) for a current of 3'08 milliamps. The 

 values in hydrogen are shown by crosses, and those in air 

 by circles. It will be seen that the effect observed by 

 previous experimenters — namely, that the rate of disintegra- 

 tion in hydrogen is less than that in air — is only true at 

 higher cathode-falls. At low falls a number of measure- 

 ments failed to show any difference between the rates in air 

 and hydrogen. 



This result, if it should prove to hold generally at all 

 currents, is one of considerable interest. It may be noted 

 that the normal cathode-fall is practically the same in air 

 and hydrogen : hence it would be expected that the rates 

 of disintegration would vanish in both gases at the same 

 point. Further, there is reason to believe that the flattening 

 of the hydrogen curve at higher falls is a secondary effect 

 and not necessarily a density effect, because it has been 

 observed by Kohlschutter in nitrogen and oxygen at still 

 higher falls. The experiments are being extended to other 

 gases, particularly argon, and until these have been carried 

 out, anv suggestions which are made must be regarded as 

 purely tentative. But the results, so far as they go, suggest 

 that the influence of the nature of the gas may not be a 

 density effect, but may depend on the value of the normal, 

 fall coupled with a secondary effect setting in at higher values 

 of cathode-fall. On this view, the favourable influence of 

 argon on disintegration would be due to the low value of 

 normal fall, and the fact that the secondary effect does not 

 occur within the limits of cathode-fall employed. This is* 

 further consistent with the fact observed by Kohlschutter, 

 that small traces of impurity added to the pure gases have a 

 marked effect on the rate of disintegration : it is well known 

 that the normal cathode-fall in a gas is influenced in the 

 same way. 



In order to test whether the above results in hydrogen 

 were caused by an impurity acquired during the passage of 

 the hydrogen from the palladium to the discharge-tube, 

 some sodium amalgam was placed between them and the 

 hydrogen bubbled through it. No difference could be 

 detected between the measured rate of disintegration in 

 those? and previous experiments. 



General Considerations. 



The experiments on the early stages of discharge and on 

 the effects of pressure and duration of discharge, show that 

 when a cathode is used which, does not oxidize and which is 



