Disappearance of Gas 'in the Electric Discharge. 227 



For much greater accuracy the moments of inertia would 

 have to be checked by experiment. 



The experiments with the small ring are interesting in 

 showing how much further we can go. The deflexion, 

 with an ordinary electrometer suspension, was approximately 

 23 cm. on the scale, for a current of 0*06 ampere. With 

 geometrical proportionality the deflexion varies as the cube 

 of the linear dimensions. Thus with a ten times weaker fibre 

 and a current of 2 amperes, a circle of inductance 1 would 

 give a deflexion of 1 cm. It is clear, therefore, that the- 

 limit would be reached by mechanical difficulties long before 

 the deflexion became inappreciable. The resistance correc- 

 tion could be kept down by cooling in liquid hydrogen. 



5. Summary. 



The inductance of a number of small loops is measured by 

 suspending them in an alternating magnetic field. The low- 

 frequency experiments are inaccurate : with high-frequency 

 current inductances from 20 cm. to 50,000 cm. are measured 

 with an average error of 1^ per cent. 



These experiments were performed in the Electrical Labo- 

 ratory, Oxford, by kind permission of Prof. Townsend. 

 I am indebted to Mr. C. A. Bowen, Exhibitioner of Corpus. 

 Christi College, for valuable help in taking the observations.. 



XXVII. The Disappearance of Gas in the Electric Discharge.. 

 — III. By The Research Staff of the General Electric 

 Company, Ltd., London *. ( Work conducted by N. R. 

 Campbell.) 



(Continued from vol. xl. p. 585 and vol. xli. p. 685.) 

 Summary. 



1-5. rj^HE previous work has shown that, except in 



A hydrogen, the absorption of gas occurs onlv if 



there is a glow discharge. If absorption is determined by 



electrical conditions at all, it is to be expected that it 



is closely associated with ionization of the gas absorbed. 



Some previous experiments might appear inconsistent with 



this hypothesis. This difficulty is now removed by showing 



that, in a discharge vessel with electrodes of which the area 



is small compared with that of the walls, those walls can act 



as a third electrode, receiving almost all the positive ions. 



the charge on which is neutralised by electrons from the 



cathode. Moreover, in such a vessel, the free path oi the 



* Communicated bv the Director. 



