158 



DR. FRANK HORTON ON THE DISCHARGE OF 



possible, because the calcium gradually sublimed off the platinum strip if left heating 

 for a long time. It is somewhat surprising that it remained on long enough for the 

 above readings to be taken. The explanation is, probably, that the calcium melts 



o i 



fgd -Kg 



800^ goo 5 



Temperature centigrade 



Fig- 



Relation between negative leak and temperature of calcium cathode in helium 

 gas at 3 236 millims. pressure. 



and combines with the platinum on the surface of the strip, to form an alloy from 

 which the calcium only slowly vapourises away. After long heating the whole of 

 the calcium could be driven off' the platinum, and the negative leak was reduced to 

 the value it had before the calcium was sublimed. 



From the numbers given it will be seen that the leak from calcium is enormously 

 greater than that from platinum at the same temperature. For example, at 1385 C. 

 the leak from calcium is about 5000 times the leak from platinum, which means that 

 there are 5000 times as many corpuscles liberated per second per square centimetre 

 from calcium as from platinum at this temperature. 



The values of the constant Q, deduced from successive pairs of these results, are 

 given in the last column of Table IV. 



The large variations in Q are most probably due to the experimental difficulties of 

 measuring the negative leak from calcium. It was generally difficult to get a steady 

 reading of the current at any temperature, for the leak increased and decreased in a 

 capricious manner. Moreover, the series of observations had to be taken very quickly, 

 because the leak gradually decreased with continual heating, owing to the calcium 

 subliming away. 



The mean value of the constant A, calculated from the temperatures and currents 



