﻿232 Potential required to maintain a Current in a Gas. 



tabulation, as above, to indicate some of the difficulties to 

 be met with in work of this kind. 



Work is now going on to determine some of these things 

 more definitely, as well as possibly to repeat all the obser- 

 vations with another gas, perhaps nitrogen. 



Conclusion. 



To sum up : — 



Various peculiarities in the behaviour of a spark-discharge 

 must be noted, which tend to complicate an apparently simple 

 investigation. 



(a) A large number of cells is necessary to secure stable 



conditions. 

 (h) Occluded gases, or other causes, may tend to introduce 

 subsidiary results depending on the material of the 

 electrodes. 



(c) The capacity of the electrodes affects the current when 



the total external ohmic resistance is small. 



(d) The sparking potential is slightly affected by the pre- 



vious direction of the current, this not being a gas 

 effect, and due either to the sides of the chamber or 

 the electrodes. 



(e) The current may be intermittent, even if the appearance 



of the spark or the galvanometer (not sensitive) shows 

 no unsteadiness. 



None of these things, however, seemed to vitiate the fol- 

 lowing main results of the investigation so far established. 



(1) The potential required to maintain a current in general 

 varies as the current is increased. 



(2) At pressures above the critical pressure this value is 

 lower than the sparking potential, tending toward a 

 limiting value. 



(3) At pressures below the critical pressure it at first rises 

 very rapidly, as the current is increased, the smaller 

 the pressure the more rapid being the rise. 



In conclusion, I must express my deep appreciation of 

 the unfailing kindness of Professor J. S. Townsend, under 

 whose direction and constant guidance the present work was 

 undertaken and carried out. 



University Museum, Oxford. 



