262 
PROFESSOR H. A. WILSON ON THE EFFECT OF HYDROGEN ON THE 
pressure. The leak obtained in this experiment was nearly the same as with a new 
wire at the same pressure. This confirms the conclusion that an old wire after 
heating in air gives the same leak in hydrogen as a new wire, although the current 
takes much longer to get to its final value at any pressure. 
The results obtained with the wire after heating in hydrogen at a high pressure 
are quite different from those given by a new wire. The values of A and Q, 
1 ‘67 x 10 lu and 135300, obtained do not satisfy the relation A = A 0 e c(Q ~ Qo)/2 *, which 
has been shown to hold for all the other values of A and Q. 
According; to the view suggested in the last section, the wire after heating for some 
time in hydrogen at a high pressure contains a stable compound of platinum and 
hydrogen having a dissociation pressure less than 0‘002 millim. The amount of 
combined hydrogen is consequently independent of the pressure. 
The amount of combined hydrogen must be independent of the temperature as well 
as of the pressure. For if on raising the temperature some of it escaped, this would 
show that its dissociation pressure was greater than that of the hydrogen outside, and 
so the whole of the compound would decompose. But a wire giving a large leak 
independent of the pressure can be heated in a good vacuum to 1400° C. without the 
leak falling ; hence, at any rate below this temperature, the amount of combined 
hydrogen must be independent of the temperature. 
Since Q for a wire containing combined hydrogen is nearly the same as Q for a new 
wire in air, it follows that the ratio of the leaks should be independent of the 
temperature. The following table contains the values of the leak given above and 
those for a new wire in air taken from my previous paper. 
, T ... 
JNew wire m air. 
Old wire. 
Temperature. 
Tj6Ii1v. 
Temperature. 
Leak. 
Ratio. 
° c. 
° C. 
1686 
4 x 10-s 
1683 
7•23 x10 -4 
1-81x10 4 
1651 
2 x 10 _s 
1648 
3-87 x10“ 4 
1•93 x10 4 
1616 
1 x IQ” 8 
1613 
1 -93 x 1CT 4 
1-93 x10 4 
It will be seen that the ratio is nearly constant. 
The leak from wires which had been heated in hydrogen was measured in air at 
various temperatures many times during the course of this investigation. The leaks 
obtained were of the same order of magnitude as those given in the previous paper 
for new wires in air. They were from 2 to 10 times greater than the leak given in 
the previous paper for a wire which had been treated with nitric acid for 24 hours. 
One wire, which had been used a long time in hydrogen, was made the positive 
electrode in IIN0 3 for several hours, and was then heated in air at a small pressure. 
It was found to give nearly the same leak as that from the wire just mentioned. 
