DISCHARGE OF NEGATIVE ELECTRICITY FROM HOT PLATINUM. 
261 
heating, at first rapidly and then much more slowly, and I suggested that this was 
due to a change in the state of molecular aggregation of the platinum. The platinum 
in these experiments was heated by a coal-gas flame and so no doubt contained 
hydrogen, so that this effect was probably due to the same cause as that just 
described. 
3. Variation of the Current with the Temperature. 
The leak from wires which had been heated in hydrogen at high pressures and were 
giving a leak independent of the pressure was measured at a series of temperatures on 
several occasions. The following table contains a set of results obtained in this way 
at a pressure of 0'003 millim. 
Temperature. 
Current. 
°C. 
ampere per sq. centim. 
1578 
9-51 x 10~ 5 
1613 
19-26 xIO - 5 
1648 
38-7 x HT 5 
1683 
72-3 x10" 5 
The mean values of A and Q in the formula x = A6 i e~ ql ’ 29 calculated from the above 
results are A = l'67xl0 10 and Q = 135300, and with these values the formula 
represents the observations very well. 
The value of Q for a new wire in air is probably about 135000, so that Q for an old 
wire in hydrogen is nearly the same as for a new wire in air. 
With the same wire, about a month later, the following results were obtained. The 
wire was heated in air, which was then pumped out and hydrogen was let in to 
a small pressure. After a time a leak developed which increased with the pressure. 
After heating for several hours at low pressures, hydrogen was let in to 0 - 084 millim. 
pressure. The leak then rose, but became constant after one hour. At each tempera¬ 
ture the leak was then steady. 
Pressure 0'084 millim. 
Temperature. 
Current. 
°C. 
ampere per sq. centim. 
1265 
6-8 x 1CT 7 
1377 
51-Ox 10- 7 
1496 
317-0x 1(V 7 
These numbers give Q = 89000 and A = 6x 10 4 . The formulae A = Kp~ c and 
Q=P —2a log p obtained in Section 1 give A= 5'5 x 10 4 and Q= 83000 at this 
