C,0 Mil. 0. \V. RICHARDSON ON THE tONISATlON 



The fact that hydrogen diffusing out of a platinum wire produces a positive 

 ionisation proportional to the amount of gas diffusing, taken in conjunction with the 

 fact that the additional ionisation so produced, even when the amount of gas 

 diffusing per minute is equal to 1 cub. centim. at C. and 760 millims. per sqiiare 

 centimetre, is only about equal to the positive leak when no hydrogen is apparently 

 present, proves that the positive ionisation in oxygen and other gases as well as the 

 negative is not due to residual traces of absorbed hydrogen. This position has 

 already been shown to be highly probable by other considerations, the chief of which 

 are: (1) The constancy of the positive ionisation in oxygen with long-continued 

 heating, (2) the agreement between different wires, (3) the fact that heating a wire 

 in hydrogen seemed to produce a permanent decrease, and not an increase, when the 

 steady ionisation it produced in oxygen was measured subsequently. 



IX. 16. SOME THEORETICAL CONSIDERATIONS. 



The above experiments show that the steady positive ionisation produced by hot 

 platinum in different gases, so far as its variation with temperature is concerned, 

 obeys a formula first deduced by the writer* and shown to represent the negative 

 corpuscular ionisation from hot platinum. That this would be the case was rendered 

 highly probable by the fact established by the writerf some time ago that the 

 temperature relations of the positive ionisation, when it is changing with time, were 

 adequately expressed by the formula C = A^e" 9 ' 2 *, A and Q being constants. The 

 only theoretical conclusion which this temperature relation seems to warrant is that 

 the liberation of an ion occurs when the dynamical system from which it is produced 

 acquires a certain amount of energy, which is furnished, it may be indirectly, by the 

 energy of thermal agitation of surrounding systems. It does not really afford any 

 evidence as to whether the production of ions is, or is not, accompanied by chemical 

 action. 



It is interesting to compare the values of Q, which represent, on the assumption 

 that equilibrium is possible, the amount of energy in calories associated with the 

 production of 1 gramme equivalent of ions. The numbers which were obtained are 

 given in the following table : 



* 'Camb. Phil. Proc.,' vol. 2, p. 286. 



t 'B.A. Reports, Cambridge,' 1904, p. 472. 



