Electricity through Argon and fJelh 



303 



The following is an account of some experiments to deter- 

 mine the ratio of saturation currents in helium and in air : — 



The gas was contained in an 

 aluminium cylinder, 2 centim. in 

 diameter (a, fig. 5), closed at one 

 end by a sheet of aluminium b 

 cemented on. The rays passed in 

 through b. The other end was 

 closed by a tightly-fitting ce- 

 mented plug of ebonite c, through 

 which passed a smaller aluminium 

 tube d. This last reached nearly 

 to the bottom of the outer cylin- 

 der. It served as an inlet for 

 the gas, and also as one of the 



Fig. 5. 



^U^ 



a 



electrodes. The outer cylinder 

 a was used as the other electrode. 

 A glass tube e was cemented into 

 d, so as to enable a connexion 

 to be made with the mercury- 

 pump, by means of which the air 

 was removed before admitting 

 the helium. 



It was found that if the rays 

 fell directly on the ebonite plug c 

 the results were very irregular. 

 To guard against this a metal disk 

 was carried on d so as nearly to 

 fill up the whole cross-section of a, 

 and thus to screen the ebonite from 



the rays. This screen was of brass, but was faced with 

 aluminium. This was to guard against the abnormal ioniza- 

 tion which, as Perrin has shown, occurs at the surface of most 

 metals, but not appreciably at an aluminium surface. 



A second precisely similar vessel was provided, which was 

 left full of air, for comparison. 



A lead-covered box was used to contain the induction- 

 coil and focus-tube which generated the rays. The two 

 aluminium vessels were arranged side by side so that the 

 rays emerging from holes in the lead box could fall on their 

 ends b. 



The wires leading to the electrometer from the gas cells 

 were screened by being led through metal tubes connected to 

 earth. 



The method of determining the conductivity was to find 

 the rate at which an electric charge could leak through the 



