﻿210 Mr. J. A. Brown on the Potential required 



position. Putting resistance in series with the condenser 

 diminishes the magnitude of these throws very considerably. 

 With d* Arson val galvanometers having unshielded coils 

 similar effects are produced. The throws, however, are now 

 towards the symmetrical position, and thus if the circuit be 

 made and broken rapidly the spot of light moves to the 

 centre of the scale whatever may be its initial position. 



In conclusion the author wishes to emphasize the impor- 

 tance of the Rayleigh correction for the galvanometer equation. 

 It explains many of the troublesome phenomena often noticed 

 by those engaged in practical tests. When this correction 

 is taken into account, the usefulness of the galvanometer is 

 considerably increased, and the ease with which the dead- 

 points can be accurately determined enables this correcting 

 factor to be readily found. The positions of the dead-points 

 also enable us to find the effective internal resistance of 

 condensers. In several of the ordinary tests used in practice 

 a knowledge of this resistance is necessary. 



XXX. An Investigation of the Potential required to maintain 

 a Current hetween Parallel Plates in a Gas at Low Pressures, 

 By J. A. Brown, M. A* 



The Potential required to maintain a Current in a 

 Spark-Discharge. 



IN some recent papers Professor J. S. Townsendf called 

 attention to the fact that the potential necessary to 

 maintain a current in a gas is in general less than the 

 potential required to cause a spark to pass, and proceeded to 

 explain this phenomenon on the theory of collision or the 

 production of fresh ions by the impact of positive and negative 

 ions on the molecules of the gas. Several determinations of 

 the current and the potential necessary to maintain it were 

 made, using a high-resistance voltmeter and a known external 

 resistance, which clearly showed the reality of this fact and 

 suggested the necessity of further and more accurate mea- 

 surements along these lines. Further observations were also 

 needed with parallel plate electrodes to show definitely how 

 the potential required to maintain a current varfes with the 

 intensity of the current, in addition to the ordinary dependence 

 on spark-length and pressure which the law for the sparking 



* Communicated by Prof. J. S. Townsend, F.RS. 

 t J. S. Townsend, Phii. Mag. viii. p. 750 (Dec. 1904) ; Phil. Mag. ix. 

 p. 289 (March 1905; , 



