20 



MR. O. \V. RICHARDSON ON THE TONISATION 



very little heat is carried from the hot wire to the surrounding electrode and the 

 walls of the tube, which consequently remain quite cold. As soon, however, as the 

 pressure of the gas is increased the temperature of the walls increases too, and if the 

 above effect were coming into play, the increase in the leak due to the heating of the 

 walls might easily appear to be a steady function of the pressure of the gas in the 

 tube. 



PUMP ETC. 



EL. 



EARTH 



Fig. 8. 



It was considered advisable to settle this question definitely by examining the leak 

 in a vessel in which the only thing which could possibly get hot was the platinum 

 wire itself. This was done by means of the apparatus shown in fig. 8. The metal 

 tube A served as the electrode to which the leak was measured and thus replaced 

 both the glass tube and the platinum electrode in the previous apparatus. This outer 

 tube was kept cold by means of water placed in the inverted wide-mouthed bottle B. 

 A heavy brass tube C, permanently connected to earth and insulated from A by 

 a clean rubber stopper D, served as a guard ring to prevent leakage from the high 

 potential wire E across the supports to the testing electrode system AB. The guard 

 tube C had a heavy flange G soldered to it which formed a base for the apparatus 

 and rested on the bottom of an earthed biscuit tin F. The lower end of the tube C 

 projected through a hole in the tin. The hot platinum wire was bound to the thick 

 copper leads H by means of fine copper wire, and the leads were supported and 

 insulated from the guard tube C by means of the rubber stopper K. Both the hot 

 wire leads and the guard tube were kept cold by a stream of water flowing through 

 the composition spiral PLMQ. The part of this which was laced round the leads H 

 was insulated from them by thin rubber tube. This was found to conduct heat well 

 enough to keep the leads cold. 



