﻿to maintain a Current in a Gas at Low Pressures. 211 



potential seems to require. It was thought worth while 

 accordingly to extend the determination of the variation of 

 the current with the potential at various pressures, and 

 particularly below the critical pressure. 



The general phenomena of the spark-discharge are now 

 well known. The so-called Paschen's* Law, confirmed pri- 

 marily by the determinations of Peace f and the careful 

 measurements of Carr J, is now well established within 

 certain limits. The sparking potential for parallel plate 

 electrodes may confidently be said to depend simply on the 

 amount of gas between the electrodes, and not to vary with 

 the material of the plate-. 



A number of the phenomena in connexion with these poten- 

 tials, and other effects noticed in the spark-discharge, have been 

 explained by the theory of ionization by collision. There are 

 peculiarities, however, which still need investigation. Under 

 certain conditions the electrodes produce effects which cannot 

 be explained by the simple theory mentioned above, which 

 has so successfully accounted for the general action. "When 

 precautions are not taken to eliminate these effects, the 

 peculiarities in the character of the discharge are very com- 

 plicated ; and some account of these is given in the second 

 part of this paper. 



Part I. 



The direct object of the investigation, then, was the mea- 

 surement of the current and at the same time the potential 

 required to maintain it between parallel plate electrodes at 

 comparatively small pressures. 



Several similar determinations of this have been made, 

 notably by Riecke§. All this work has. however, been 

 undertaken with ordinary vacuum-tubes, with electrodes pro- 

 jecting into the gas and with no attempt to secure a uniform 

 field. In Riecke's work, for instance, the tube was a sphere, 

 with electrodes near the ends of diameters at right angles to 

 each other, their distance apart being approximately 6*5 cms. 

 Sparking accordingly frequently started from the back of the 

 cathode when the pressure was decreased below a certain 

 value. So the statement he makes, that the value of the 

 current and the potential to maintain it is a function depending 

 on the tube, cannot, be wondered at. 



Carr has shown the necessity of trying to secure uniformity 



* Pasclien, Wied. Ann. xxxvii. p. 79 C18S9). 

 T Peace, Proc. Roy. Soc. lii. p. 99 (1892). 

 X Carr, Proc. Roy. Soc. lxxi. p. 374 (1903). 

 § Riecke, Ann. der Phys. iv. p. 592 (1901). 

 P2 



