DISCHARGE OF NEGATIVE ELECTRICITY FROM HOT PLATINUM. 
249 
It was shown in the previous paper that Q and A depend on the pressure of the 
hydrogen. I propose now to consider this variation of A and Q, and it will be shown 
that it can be deduced from the assumption that x = Bp", where p denotes the pressure 
of the hydrogen, and B and n depend only on the temperature. 
The values of A and Q given in the previous paper were calculated from some of 
the observed currents, and when put in the formula x — A9 i e~ Ql,2e give nearly the 
observed values of x. I have recalculated them, using all the observations in turn, 
and taking the mean of the results. The new values of Q so obtained from the 
currents and temperatures given in the previous paper differ to some extent from the 
old values, and since a small error in Q produces a large error in A, the new values of 
A are considerably different. The following table contains the new values :— 
Gas. 
Pressure. 
Q. 
A. 
For a Wire Treated 
with HN0 3 for 24 H 
ours. 
Air 
Small 
145000 
1•14 x 10 8 
Wire Boiled in HN0 3 for 1 Hour. 
Air 
Small 
131000 
6-9x10 7 
Ho 
0 ’0013 mm. 
110000 
10 6 
Ho 
0-112 „ 
90000 
5 x 10 4 
Ho 
133 
56000 
2 x 10 2 . 
The new values, of course, represent the results a little better than the old ones. 
Any one measurement of the negative leak is subject to a large possible error, but 
since the values of Q are obtained from a considerable number of observations, it is 
probable that the effects of errors are to some extent eliminated in the values of Q. 
For this reason 1 have attempted to establish the relation between the leak and the 
pressure by means of the values found for Q and A. 
The results given in the previous paper {loc. cit.) on the variation of the leak with 
the pressure of the hydrogen show that the leak is proportional to a power of the 
pressure less than unity. 
In fig. 1 the logs of the pressures and currents given on p. 265 are plotted, and 
the points fall nearly on a straight line, the slope of which shows that the current 
varies as _p 0 ' 75 nearly. 
Let x denote the current per square centimetre of platinum, then at each constant 
temperature assume x = Bp", where p denotes the pressure of the hydrogen and B 
and n depend only on the temperature. This is to be consistent with x = A 9*e~ Q2e , 
where 9 denotes the absolute temperature and A and Q depend only on the pressure. 
VOL. CCVIII.-A. 2 K 
