DISCHARGE OF NEGATIVE ELECTRICITY FROM HOT PLATINUM. 
259 
leak 4x10 8 times greater than the leak at the same temperature in oxygen. On 
heating in oxygen the pressure fell from 1'067 millims. to 1'02G millims. I have 
tried a similar experiment with a wire giving a large leak independent of the pressure 
which was heated in hydrogen at a small pressure until it ceased to evolve gas. A 
new wire was put up and heated for a few hours in hydrogen at atmospheric pressure. 
It was then left cold in hydrogen all night, and in the morning gave a large leak 
nearly independent of the pressure. The gas was then pumped out till the pressure 
was O'OOll millim. The wire was then heated to a temperature of about 1200° C. 
for four hours, during which time the leak was about 3 x 10~ 5 ampere, but rose 
slightly towards the end of the experiment. After one and a-half hours the pressure 
had risen to 0'01 millim., but it remained constant during the next two and a-half 
hours. The wire was then allowed to cool, and oxygen was let in to a high pressure 
and pumped out several times. The oxygen was then pumped out to O'0015 millim. 
The wire was then heated for a few seconds to about 1100° C., and after ten minutes 
the pressure was 0'0012 millim. The wire was then heated for two minutes and after 
a time the pressure fell to 0'0009 millim. The wire was now heated for two minutes 
to about 1500 C., and after a time the pressure was again 0'0009 millim. 
On first heating the wire in the oxygen a large leak was observed which very 
quickly fell to a very small value. The capacity of the apparatus in the above 
experiment was 500 cub. centims., so that the volume of hydrogen (measured at 
760 millims.) contained in the wire was 8 x 10~ 3 cub. centim. The volume of the wire 
was 10 x - 7 - x (Y^qp = 3xl0 -3 cub. centim., so that the wire contained about three 
times its volume of hydrogen, although it had been heated for two and a-half hours 
without evolving any gas. The gas evolved during the first one and a-half hours 
may have come from the wire, but some of it probably came from the surrounding 
electrode, which, of course, got warm during the long continued heating. When the 
wire was heated in oxygen, the electrode was not heated appreciably, because the 
current was only kept on a short time. I think this experiment shows that a wire 
which has ceased to evolve hydrogen when heated in hydrogen at a small pressure 
may still contain several times its volume of hydrogen. 
If the wire absorbed oxygen when heated in it this would account for the fall in 
the pressure. The oxide of platinum, PtO, which is formed at 510° C., decomposes 
completely at 560° C. The temperature in the above experiment was much higher 
than 560° C., so that it is not likely that any oxide was formed. According to 
Bodlander, platinum absorbs oxygen energetically at from 700° C. to 900° C., but 
these temperatures are also lower than the temperature used. 
If we admit that a wire giving a large leak independent of the pressure contains 
hydrogen, to which the leak is due, then it is necessary to suppose that the hydrogen 
is not all merely dissolved in the platinum. Since the leak is independent of the 
pressure, the amount of hydrogen in the wire must be independent of the pressure. 
If the temperature of the wire is raised while it is in a good vacuum, the leak on 
2 l 2 
