DISCHARGE OF NEGATIVE ELECTRICITY FROM HOT PLATINUM. 
257 
hydrogen at 0'005 millim. pressure. The leak remained unchanged for two hours 
and then rose to about 4xl0 -6 ampere per square centimetre. After this, on 
increasing the pressure, the leak rose at once and remained steady. 
Temperature 1490° C. 
Pressure. 
Current. 
millim. 
amperes 
0-0135 
6-56 x10-6 
0-0168 
7-80 x 10-e 
0-0281 
10-50x10- 6 
0-0399 
12■70 x10^« 
0-0579 
l7-20x 10“ fi 
0-0776 
20•50 x10 -6 
0-1151 
24-60 x 10“ 6 
This experiment was repeated several times with similar results. On diminishing 
the pressure the leak fell off very slowly with the time, so that the pressure could be 
reduced to nothing without much diminishing the leak. This shows that the increase 
in the leak is not due to ionisation by collisions. 
If the logs of the numbers just given are plotted, the points fall nearly on a straight 
line, the slope of which shows that the current is proportional to p 0 ' 57 . At 1490° C. 
aO~ l — c is equal to 0 - 63. Thus the variation of the leak with the pressure is nearly 
that calculated from the old results with new wires. The value of the leak at 
1490° C., with p — 0T151, given by the formula obtained in Section 1, is l - 3xl0 5 , 
which does not differ much from the leak found. This wire, therefore, which had been 
heated in hydrogen at high pressures for a long time and then in air, gave the same 
leak in hydrogen as a new ware. It differed from a new wire only in that the leak 
fell very slowly with the time when the pressure was diminished, and that on first 
admitting the hydrogen the leak did not rise for a considerable time. 
If we regard the leak from a wire which has not been heated in hydrogen at a high 
pressure as due to dissolved hydrogen, then, since in the case just described the leak 
was the same as that given by a new wire, we must regard the leak in this case also 
as due to dissolved hydrogen. Thus the difference between an old wire and a new 
one appears to be that the old wire absorbs and evolves hydrogen much more slowly 
than the new wire. 
A wire which has been heated in hydrogen at high pressures differs from a new 
wire and from an old wire that has been heated in air. It always gives a very large 
leak which is nearly independent of the pressure. Even when it is heated in a good 
vacuum for many hours the leak does not fall. 
Professor O. W. Richardson (loc. cit .) found that the leak in hydrogen was 
independent of the pressure when this was reduced from a few millimetres to nearly zero. 
VOL. CCVIII.—A. 2 L 
