586 Research Staff of the G. E. C, London, on the 



gases, may be made to disappear, although some gases — 

 ■e.g., hydrogen and nitrogen — disappear more readily than 

 others — e. q., helium. Again, the material of the walls and 

 the electrodes affects somewhat the rate of disappearance*; 

 but whatever the material, some disappearance will occur. 

 On the other hand, there are instances in which the dis- 

 appearance is determined directly by the chemical nature 

 of the gases and the electrodes. Thus, a cathode of one of 

 the alkali metals will produce a disappearance of most gases 

 in circumstances in which a cathode of a less active metal 

 would not ; but it has no effect on the inactive gases. 

 Again, it is well known that the introduction of the vapour 

 of phosphorus accelerates very greatly the disappearance 

 of all active gases. The conclusion is therefore generally 

 drawn that chemical action may play a part in the dis- 

 appearance of gas, probably by the formation of chemical 

 compounds of the elements of the gas and of the electrodes 

 which have no appreciable vapour-pressure ; but that in 

 some cases there must be at work agencies that are not 

 chemical f . There is much evidence that the gas can 

 be caused by the discharge to adhere to the solid parts of 

 the discharge-tube in some manner which still remains 

 obscure, and that it can be liberated again from them by a 

 rise in temperature. 



Concerning the electrical conditions which determine the 

 rate of disappearance of the gas, the most definite evidence 

 seems to be given by the experiments of Vegard % and of 

 Brodetsky and Hodgson §. These workers have shown that 

 the disappearance of gas is closely connected with the 

 cathode fall of potential in the discharge ; it is therefore 

 connected with the cathodic spluttering, which also increases 



* For the action of the glass walls of the vessel, see G. E. Hill, Phys. 

 Soe. Proc. xxv. p. 35 (1912), and R. S. Willows and H. T. George, ibid. 

 xxviii. p. 124 (1916). No experiments seem to have "been made without 

 glass or silica. 



t The term " chemical " has been used by some writers to include so 

 much that it may be well to state precisely what is here meant by the 

 term. An action is not considered chemical unless it results in the 

 formation of new combinations which follow the law of "constant 

 proportions/' which can be separated to some extent from the oiiginal 

 combinations, and which can be shown to differ from them in physical 

 characteristics — boiling and melting points, and so on. Of course, there 

 may be reason for believing that an action is chemical in this sense 

 without a definite proof that the law of constant proportions is obeyed ; 

 but unless there is reason for such a belief, the action will not be termed 

 chemical. 



t L. Vegard, Ann de Phys. 1. p. 769 (1916). 



§ S. Brodetsky and B. Hodgson, Phil/ Mag. xxxi. p. 478 (1916). 



