204 Prof. J. A. Cunningham on the 



discharges is that they are both accompanied by a low value 



dY 

 of -7p (corresponding to u resistance" in metallic conduction, 



where V = P.D. and C = current), or a high conductivity of 

 the gas. They are separated by a region of luminous dis- 

 charges accompanied by low conductivity of the gas. In 

 other words, we have here got to a second saturated current 

 succeeding a region of much ionization by collision, and suc- 

 ceeded in turn by another stage of further rapidly increasing 

 ionization. 



There seem to be three possible explanations of thi^ 

 phenomenon : — 



(1) That a new sort of negative ions come into the field 

 which produce fresh ionization by collision at this particular 

 stage. 



(2) That we have here reached a value for the energy of 

 the colliding corpuscle such that it knocks more than one 

 new corpuscle off the molecule which it strikes. Or 



(3) That this second diminution of the Aalue of 7.- marks 



the point at which the positive ions also begin to produce 

 fresh ions by collision. 



The objection to the first of these hypotheses is that, at the 

 low pressures here dealt with, it is not likely that the cor- 

 puscles coming from the cathode can get loaded up by 

 sticking to molecules of the gas, and so produce heavy 

 negative ions. And we have hitherto no evidence of two 

 kinds of corpuscles. 



As regards the second hypothesis, J. J. Thomson* has 

 shown that the energy available for ionization by a moving 

 corpuscle attains a maximum for a certain value of its 

 velocity, and after that diminishes inversely as the square of 

 the velocity, without any possibility of a further increase, at 

 least if the corpuscles be assumed to repel each other with a 

 force inversely proportional to the square of their distance 

 apart. 



The third seems, on the whole, the most likely hypothesis, 

 if we distinguish between the action of the positive ions, on 

 the one hand, at the surface of the cathode in helping out 

 the corpuscles which may be regarded as going on ordinarily t 

 with small currents not exceeding this second saturation 

 value ; and, on the other hand, their action in ionizing the gas 

 in the Crookes's dark space when the P. D. increases still further, 

 and the current in consequence rises again above its saturation 

 value. But the fact that this " supersaturation current ? f 



* ' Conduction of Electricity through Gases/ pp. 344-345. 

 t J. J. Thomson, ibid. p. 480. 



