1(54 DR. FRANK HORTON ON THE DISCHARGE OF 



with different cathodes, are thus seen to vary very considerably ; the values of Q, too, 

 were not in very good agreement. It seems likely that this may be due to the 

 platinum being more completely covered with lime in some cases than in others, for it 

 will be readily understood that the cathode could not be quite uniformly covered with 

 calcium at each attempt by the method of sublimation. In order to obtain more 

 accurate knowledge of the values of these constants, experiments must be made with 

 very carefully prepared lime cathodes. Meanwhile, the fact that a large emission of 

 negative corpuscles takes place from a Nernst filament at high temperatures seems 

 to indicate that in experiments with lime-covered cathodes the corpuscles proceed 

 from the oxide, and not from the platinum underneath. 



In view of the experiments of Professor WILSON, which have shown that the 

 presence of hydrogen enormously increases the negative leak from platinum, it was 

 thought to be interesting to see if the leak from lime was increased by admitting 

 hydrogen into the apparatus. It was found that this is the case, the leak in 

 hydrogen being many times greater than in helium or oxygen. The effect of 

 introducing a little hydrogen into the apparatus is well shown by some observations 

 taken with a lime cathode which had been used for some days, and the negative leak 

 reduced to even a smaller amount than the values given in Table VI. The following 

 are the values of the negative leak from such a cathode. The gas present was a 

 mixture of helium and oxygen at a pressure of 3 '91 millims. The voltage used was 

 -40 volts. 



TABLE VII. 



Temperature, C. 1038 1382 1520 



Negative leak in amperes 6 x lO" 9 1 32 x lO" 5 1 13 x 10~ 4 



The above numbers are the smallest values of the negative leak at the temperatures 

 given that I ever obtained from a lime cathode. There was no sign of a luminous 

 discharge, even at the highest temperature. After taking these observations, some 

 pure dry hydrogen was let into the apparatus, and the cathode was gently warmed 

 until no further diminution of pressure took place. The resulting gas was a mixture 

 of helium and hydrogen at a pressure of 3 '81 millims. A luminous discharge was 

 now noticed when the temperature of the cathode was 1220 C. and the current 

 passing 3'12xlO~* ampere. The glow in the gas was of a pale blue colour, and 

 appeared only round the edges of the anodes. As the temperature of the cathode 

 was gradually raised and the current passing increased, the glow became more extensive 

 and brighter. At 1465 C. it was very white, and had gathered up into little balls 

 about points on the rim of the anodes. Although the temperature was raised as high 

 as was compatible with the safety of the cathode, and the current passing rose to one- 

 twentieth of an ampere, no cathode glow was obtained. The following is a selection 

 from a list of readings obtained in this series of experiments. The readings of the 

 negative leak, at temperatures over 1250 C., were taken with a milliainmeter. The 



