252 
PROFESSOR H. A. WILSON ON THE EFFECT OF HYDROGEN ON THE 
Q = P — 2a log p, which has been shown to agree with the observed values of Q at 
different pressures. 
Since n — a0~ 1 —c at the temperature 6 = aJc, the leak should be independent of 
the pressure. 
If we substitute the expressions found for A and Q in the equation x = Affie~ Q2fl , 
we get 
x — A 0 (1 + ap c ) (alec ~ l) ffie~ Qr/2e . 
If 
R' = Qo + 2 (0—ac~ l ) log (1 +ap c ) 
and 
D' = A 0 (l +apj alec ~ 1 \ 
this takes the forms 
x = A o dA _R,/20 and x = D'0*e -Qo/2fl . 
In the equation x — A0 i e~ QI2e , which represents the variation of the leak with the 
temperature, Q is usually taken to be the energy required for the liberation of one 
gram-molecule of electrons, while A is proportional to the total number of free 
electrons present per cubic centimetre in the surface layer of the platinum. The 
experiments show that Q and A are both diminished by hydrogen. If we suppose 
that R denotes the true value of the energy required for the liberation of one gram- 
molecule of electrons, and that this quantity is a function of both 0 and p, then, for 
example, if R = R', we have x = A (j 0 i e~ n ' 120 , and this equation shows that all the 
results with new wires can be represented by supposing that the true value of A is 
A 0 , whatever the temperature and pressure. On the other hand, if we denote the 
true value of A by D, then, for example, if D = D', we have x — T>'0 i e~ Qol2e , which 
shows that all the results can be represented by supposing that D is a function of 0 
and p, while R is always equal to Q 0 . 
In general, if R = f{@,p), where f(0,p ) denotes a function of 0 and p, then I) 
will be given by the equation 
lo ^ = p (R -Q) + (|- 1 ) Io s( 1+ ^). 
so that, in general, both D and R may be any functions of 0 and p which satisfy this 
equation, and the observed results will still be represented equally well by the 
equation x = D0-e~ me at any pressure and temperature. 
It appears, therefore, that on the view that R and D may be functions of 0 as well 
as of p the results considered so far are not sufficient to determine the values of either 
of these quantities. 
Let x denote the leak in air and x! that in hydrogen at the same temperature ; 
then we have 
x'jx = (1 + ajf) {a,0c ^ l) . 
