l'l>|>rci.l> BY HOT I'LATINTM IX DIFFKHKXT CASES. 



11 



Similar results were obtained at a pressure of 0'0015 millim.. so that this effect 

 occurs at all pressures. 



The numbers quoted do not give the value of the current immediately the potential 

 was put on, but what it settled down to later. In all these cases it was found that 

 on raising the potential the current was too big at first, and only settled down to its 

 steady value after some minutes. Similarly, on changing to a low voltage the 

 current was too small at first and subsequently increased somewhat. Effects of this 

 kind were much more marked with wires which had not been heated very much, and 

 will be considered more fully under 13. They can obviously be explained if we 

 assume that the electric field displaces the equilibrium condition in such a way as to 

 reduce the number of ionising systems. 



Effects of this kind could not be detected at higher temperatures, owing to the 

 greater difficulty of attaining saturation already alluded to. In these cases it was found 

 that, though no certain increase in the current could lie detected on changing the 

 potential from 40 to 80 volts, yet it increased by about 60 per cent, with a potential 

 of 760 volts. This increase was approximately proportional to the voltage. When 

 it occurred, 600 volts was used to measure the saturation current ; it was not safe to 

 go up to much higher voltages, owing to the possibility of ionisatiou by collisions 

 occurring even at high pressures. 



5. Current and Pressure. 



We come now to what has been regarded throughout as the main object of this 

 investigation, namely, the relation between the ionisation produced by the wire and 

 the preasure of the surrounding gas. In interpreting the results, it is important to 

 remember (1) that the current was always saturated, and (2) that, at low temperatures 

 at any rate, the negative leak in oxygen was always found to be small compared with 

 the positive. These two results conjoined prove that volume ionisation of the gas 

 round the wire is negligible, at any rate at low temperatures. Moreover, there is every 

 reason to believe that the negative ionisation, which makes itself felt even in oxygen 

 at higher temperatures, is the ordinary corpuscular emission from hot metals, so that 

 everything points to the positive ionisation being due to an action which takes place 

 at the surface of the metal. 



c 2 



