48 MR. O. W. RICHAR1>SON ON THE TON1SATION 



All the above facts point to the view that the reduction of the negative leak in 

 hydrogen produced by the application of a high potential is due to a change in the 

 surface caused by the bombardment of the surface by the positive ions produced in 

 the gas by ionising collisions. This view is supported by the fact that after the high 

 potential had been applied a greater heating current was always required to maintain 

 the wire at its original temperature. This shows that the amount of heat radiated 

 from the surface at a given temperature was greater than before, so that the nature 

 of the surface must have become changed in some way. 



Two views which are not mutually exclusive may be taken as to the nature of the 

 action by which the bombardment of the positive ions, which are really weak canal 

 rays, reduces the negative leak. They may act either by destroying a layer of 

 positively-charged hydrogen which helps the corpuscles out of the metal, or they may 

 merely allow the absorbed hydrogen to escape from the wire. The last suggestion 

 receives strong support from the recent experiments of SKINNER on the evolution of 

 hydrogen from metallic cathodes under the influence of the luminous discharge.* On 

 the other hand, it is difficult to conceive how bombardment by positive ions for two 

 minutes can allow so much hydrogen to escape from a wire that it takes several 

 hours for it to diffuse back again. On the whole, the evidence, though inconclusive, 

 is in favour of the double layer theory. 



The writer has examined the effect of changing the temperature on the value of 

 the steady negative ionisation in hydrogen, and has confirmed WiLSON'st result that 

 increasing the temperature gives a leak which is too big initially, whilst decreasing 

 the temperature has the converse effect. The curve showing the recovery with time, 

 after heating to a high temperature, is similar in form to that obtained after exposure 

 to a high potential, although the ratio of final to initial value of the leak was smaller 

 in the cases examined. 



So far little has been said about the steady positive ionisation in hydrogen. We 

 saw on p. 44 that a wire when first heated in hydrogen gives a considerable positive 

 leak which gradually decays with time. It is an interesting question whether this 

 decay would go on indefinitely, or if the positive leak has a minimum value depending 

 in pressure, temperature, &c. The following experiments show that the steady 

 positive leak due to hydrogen at 3' 8 millims. pressure and a temperature of 900 C. is 

 very small, even if it exists. 



The wire (diameter = 0"2 millim. and effective area 0'67 sq. centim.) was 

 maintained at a constant temperature of 900 C. in a hydrogen vacuum for about 

 3 hours. By pumping from time to time, the steady pressure was kept below 

 O'OOl millim., though hydrogen was being given off by the wire. The values of the 

 leak with + 40 volts at various times were as follows : 



* SKINNKK, ' Phys. Rev., 1 vol. 21, p. 1 (1905). 



t H. A. WILSON, 'Phil. Trans.,' A, vol. 202, p. 265. 



