DISCHARGE OF NEGATIVE ELECTRICITY FROM HOT PLATINUM. 
273 
Thus —/3 o p o 0 — |E, where E is the kinetic energy of the electrons. Hence the energy 
required to raise the temperature of the electrons in 1 cub. centim. of platinum one 
degree is dE/dO = — fp 0 /3 ( , = f x 2'73 x 10 u x 2 - 8 x 10 7 ergs. This gives dE/dO = 2‘7 
calories, which is about three times the heat required to raise the temperature of 
1 cub. centim. of platinum one degree at high temperatures. 
It is possible that the number of free electrons per cubic centimetre is greater close 
to the surface of the platinum than inside. The value of p„ deduced from the 
negative leak is, of course, the value close to the surface. Another very possible 
explanation of the difficulty is that D and p 0 have been over-estimated. If D is 
taken ten times smaller, or 1'44 x 10 9 , the energy required per cubic centimetre is only 
0 - 27 calories per cubic centimetre, which is about one-third of the heat capacity per 
cubic centimetre. This value of D makes the thickness of the double layer 
2x 1CT 8 centim. instead of 2‘6x 10“ 8 , and with it Q 2 log (D/A) is nearly as constant 
as with the value used. In fact, the possible error in D is at least a factor of 10, so 
that the difficulty about the specific heat is not very serious in the case of platinum 
if the value of p 0 , deduced from the negative leak, is taken. 
10. Conclusion. 
The view taken in the above paper is that the effect of the hydrogen on the 
negative leak is due to its presence in the surface layer of the platinum. To explain 
this it is supposed that the hydrogen atoms are positively charged, so that they 
diminish the charge per unit area in the electrical double layer covering the surface 
of the platinum. The hydrogen appears to dissolve in the platinum at first, but at 
high pressures in time forms a stable combination, having a very small dissociation 
pressure. With new wires, before the stable compound has been formed, the leak is 
proportional to a power of the pressure of the hydrogen. 
My best thanks are due to Mr. G. H. Ma rtyx, B.Sc., for his assistance in carrying 
out the experiments described. 
PRESENTED 
io JUiiiooa 
VOL. CCVIII.—A 
