1903.] On the Discharge of Electricity from Hot Platinum. 275 



It is found that this formula accurately represents the variation of 

 the negative leak with the temperature. Q is found to be 130,000, 

 and it is shown that it cannot vary more than one part in one hundred 

 thousand per degree Centigrade. The value of the constant A is 

 shown to be diminished by cleaning the wire with nitric acid. With 

 a wire boiled for a few minutes in nitric acid A = 7 x 10 r , while with 

 a wire very thoroughly cleaned A = 6 x 10 6 . 



The negative leak in hydrogen was measured and found to be very 

 much greater than in air. At low pressures, using a P.D. too small 

 to produce ionisation by collisions, the current is proportional to the 

 pressure of the hydrogen. The leak from a wire in hydrogen at 

 0*1 mm. pressure is several thousand times that from a clean wire in 

 air or in a vacuum. 



It is shown that the negative leak in hydrogen depends on the 

 amount of hydrogen absorbed by the wire. When the temperature or 

 pressure is suddenly varied, it takes a considerable time for equilibrium 

 to be established between the hydrogen in the wire and that outside, 

 and the leak varies in consequence with the time. The following 

 table gives the negative leaks at 1400 C. at several pressures in 

 hydrogen. 



Pressure. Current per sq. centimetres. 



133-0 mms. 1 -0 x 10~ 3 ampere. 



0-112 l-2xlO~ 5 



0-0013 2-0 xlO~ 7 



0-0 l-2xlO- 10 



The following table gives the values of the constants Q and A found 

 for wires in air and hydrogen at several pressures : 



Pressure. Q. A. 



(1) Thoroughly cleaned wire in air 



or vacuum 155,000 6'OxlO 6 



(2) Cleaned wire in air or vacuum 131,100 6'9 x 10 7 



(3) 0-0013 mm. H 2 120,000 10 7 



(4) 0-112 H 2 85,900 5'3xl0 4 



(5) 133-0 H 2 36,000 0-1 



If we regard the leak in air, or a vacuum, as due to traces of hydrogen 

 occluded in the wire, then we see from the above table that Q steadily 

 increases as the amount of hydrogen in the wire diminishes. The 

 constant A, however, increases with increasing quantity of occluded 

 hydrogen, when very little hydrogen is present, but when the amount 

 of hydrogen is further increased it attains a maximum, and then 

 diminishes. The fact that the negative leak in air at low pressures 

 always falls off with long continued heating, confirms the view that it 

 is due to occluded hydrogen. Treating the wire with nitric acid 



