166 DR. FRANK HORTON ON THE DISCHARGE OF 



gradually increasing the temperature the luminous discharge began without any 

 sudden jump in negative leak taking place. I again found that after a certain 

 temperature had been attained and a large current was passing, I could decrease the 

 temperature of the cathode and still get the luminosity to continue and the current 

 to pass. 



These experiments show that the negative leak from lime is enormously increased 

 by replacing the gas in the apparatus by hydrogen. With a lime cathode in hydrogen 

 at a pressure of O'Ol millim. I obtained a current of 0'15 ampere per square 

 centimetre at about 900 C., with a difference of potential of 40 volts between the 

 electrodes. This is the largest negative leak I have measured under this potential 

 difference. 



The appearance of the luminous discharge is of great interest. Generally the 

 luminosity began round the cathode, and was of a very faint blue colour, getting 

 whiter and more extensive as the temperature of the cathode was increased. When 

 the cathode was unequally covered with lime, the discharge could be seen to radiate 

 out from a few points only. The appearance of the luminosity at any point depends 

 on the current density at that point, and with a very evenly covered cathode large 

 currents could be made to pass through the tube without any signs of a luminous 

 discharge appearing. The appearance of the luminosity also depends on the potential 

 difference between the electrodes. The luminosity could not be obtained with a 

 potential difference of less than 18 volts ; and it seems probable that this is the value 

 of the anode fall of potential, for the cathode fall is reduced to a very small amount 

 by the enormous emission of negatively charged corpuscles from the cathode.* As a 

 rule, the luminous discharge gradually became visible, and increased in brightness as 

 the temperature of the cathode was slowly raised. When the luminosity appeared 

 gradually there was no sudden jump in the current passing. This is well illustrated 

 in the curves of fig. 5. On the other hand, when the potential difference between 

 the electrodes was much greater than 40 volts, the luminosity usually appeared quite 

 suddenly and was accompanied by a sudden increase in the negative leak. 



The current density obtained with a calcium-covered cathode was, generally, not 

 sufficient to produce a luminous discharge, but on one occasion, on heating the 

 cathode to a much higher temperature than usual, a faint luminosity was observed. 

 This was at about 1520 C., and the current passing through the tube was 4 milli- 

 amperes. Some interesting experiments were made with this cathode. The tempera- 

 ture was kept constant, and the potential difference between the electrodes was 

 increased from zero by two volts at a time. No luminosity was obtained until a 

 potential difference of 20 volts was reached. With this voltage a pale glow was seen 

 round the anode. This glow increased in brightness as the voltage was increased. 

 With 28 volts the glow left the anode and a pale luminosity appeared round the 

 cathode. At the same time the current passing increased from being too small to 



* WEHNELT, 'Phil. Mag.,' 6, vol. 10, 1905. 



