52 



MR. O. \V KlfllAKPSOX ON THE IONISATION 



about nine-tenths of the hydrogen originally present in it at the end of the 

 experiment, so that the experiment appears to warrant the conclusion that the 

 amount of the negative ionisation depends very little on the amount of hydrogen in 

 the wire. In fact, the hydrogen appears to act by altering the condition of the 

 surface of the wire, and once this change has taken place it is very little affected by 

 changes in the amount of hydrogen either outside or inside. The most reasonable 

 view appears to be to suppose that the positively charged hydrogen atoms form an 

 electrical double layer, which helps the corpuscles out of the metal. 



A few measurements of the variations of the ionisation, both positive and negative, 

 with the temperature were made in hydrogen. Experiments at pressures of the 

 order of 1 millim. were found to be particularly difficult to carry out on account of 

 the length of time required for equilibrium to be established and the difficulty 

 of being certain that it was established. The following values, with a pressure of 

 1'90 millims. represent the best series of measurements at this kind of pressure. 



These numbers for the positive ionisation will be seen to be considerably greater 

 when the difference of temperature is allowed for than the minimum value given on 

 p. 49. This indicates that the steady condition had not really been attained when 

 the measurements were made, although the final reading was never recorded until 

 the leak appeared to be varying very slowly, if at all, with the time. [Another 

 possibility, which must be kept in view, is that these inconsistencies are due to some 

 other undiscovered factor, which is not taken account of.] If we calculate from the 

 above numbers the energy change associated with the liberation of a gramme 

 molecule of ions of each sign we find iv + = 3'58xl0 4 calories and IV- = 12'0xl0 4 

 calories. 



A series of measurements was also made at a much higher pressure (22G millims.). 

 The numbers obtained are given in the table following. 



No regular change could be detected in the value of the positive ionisation at the 

 lowest temperature over a space of half-an-hour. This tends to confirm the con- 

 clusion from the experiments on p. 49 that there is a positive leak in hydrogen 

 which is a function of the pressure. At low pressures this is much smaller than the 

 positive leak in oxygen, but it increases more rapidly with the pressure. It also 



