196 Prof. J. A. Cunningham on the 



words, some of the energy supplied to the cathode as a 

 heating current is consumed in aiding the cathode to dis- 

 charge negative electricity. Mr. 0. W. Richardson* has 

 shown that this is a very considerable quantity, increasing 

 rapidly with increasing temperature. This view is also sup- 

 ported by the results shown in fig. 4, where it is evident 

 that it requires far less electrical potential to extract a large 

 supply of corpuscles out of the cathode at a high than at a 

 low temperature. I hope to put this matter to the test of 

 further experiments specially arranged for that purpose. 



It was noticed that at a low temperature, on the other hand, 

 the starting of the discharge through the gas at once heated 

 up the cathode to a temperature estimated at from 100° to 

 250° — the resistance of the carbon diminishing to as low as 

 0*92 of its value when cold — being markedly warmer with 

 increased current. It was, of course, in this way that Hittorf t 

 heated up an iridium electrode to a bright yellow heat. He 

 also (loc. cit. p. 128) showed that there was enough heat 

 developed in different parts of the discbarge to heat up 

 mercury thermometers placed in its path, and this heating- 

 increased as the cathode was approached. E. Wiedemann i 

 estimated the temperature of the gas molecules carrying a 

 current at from about 70° to about 86,000°, according to 

 the diameter of the discharge-tube, being inversely pro- 

 portional to the cross- sectional area for a constant current. 

 The temperature at different points along the discharge 

 was very carefully measured by R. W. Wood §, who 

 also found the temperature developed proportional to the 

 current, ceteris paribus. For a current of *001 ampere at a 

 gas-pressure of 1*5 mm. it was about 32° in the positive 

 column. And in all cases he too found the temperature 

 increasing very rapidly towards the cathode ; so that it may 

 be estimated from his curves at anywhere between 100° and 

 300° at the surface of the cathode. The distribution is 

 exactly what was to be expected from the curves of potential 

 gradient, to which the heat developed should evidently be 

 proportional, since the same current flows across every section 

 of the discharge-tube. 



The temperature of the hot electrode was determined by 

 including it as one arm of a Wheatstone's bridge, and so 

 measuring its resistance. The second arm of the Wheatstone's 

 bridge which had to carry the heating current consisted of a 



* Phil. Trans. A, vol. cci. p. 500 (1903). 



t Wied. Ann. vol. xxi.p. 90 (1884). 



t Ibid. vol. vi. p. 298 (1879). 



§ Ibid. vol. lix. p. 238 (1896.. 



