Discharge from an Electrified Point. 427 



galvanometer, suitably shunted, was connected to each 

 of the zones A and 0, the zones B and D being earthed 

 directly. In some of the work a galvanometer was con- 

 nected to each zone, so that simultaneous observations might 

 be made on the current at different parts of the plate ; and 

 the total current arriving at the plate was also measured by 

 a galvanometer placed between the other galvanometers and 

 the earth. The bell-jar was filled with dry air at atmospheric 

 pressure. 



The electrometer used to measure the potential of the point 

 was of an attracted disk type, used originally by Chattock *. 

 In it the attracted disk is suspended, and the force of attrac- 

 tion measured by tilting the whole instrument until the disk 

 falls back by its own weight to a fixed point ; a fine glass 

 pointer attached to the back of the disk and viewed through 

 a microscope serves as an indicator. By altering the distance 

 between the attracting plate and the suspended disk the 

 range of the instrument can be varied from about 100 volts 

 to some tens of thousands. 



The spots of light of the two galvanometers connected to 

 A and C were made coincident upon the same scale, and the 

 Wimshurst was turned at such a rate that the deflexions of the 

 two galvanometers were equal. Similar readings were then 

 taken for the other sign of discharge, though the value of 

 the deflexion common to both instruments was now different. 

 Since the ratio of the currents at the two areas A and C was 

 the same in both cases, the necessary condition of constant 

 spread was thus attained. From the current and potential 

 readings the ratio of the specific velocities of the positive 

 and negative ions in air at atmospheric pressure was then 

 calculated. 



Some typical results are given in the following table in 

 which d is the distance between point and plate, and r the 

 ratio of the specific velocities of negative and positive ions 

 in air. 



Table I. 



d (in cms.). 



r. 



1-4 



(1-49) 



2-4 



1-38 



4*4 



1-38 



6-4 



1-41 



8-4 



1-39 



9-4 



1-40 



Mean 1-39 

 * Phil. Mag. [5] vol. xxxii. p. 302 (1891). 



