Ionization Potential of Mercury Vapour. 755 



were distant apart 1*7 cm., while D was placed 1'2 cm. 

 from C. All the joints employed were of wax, care being 

 taken that they were situated at such a distance from the 

 heater that they were not affected by it. The heater itself 

 was a double-walled electrical one, variable resistances in 

 the circuit allowing the temperature of the vessel to be 

 maintained constant. 



§ 3. Experiment. 



The vessel was heated to a temperature of 79° C. for 

 several hours before evacuation. By this means most of the 

 gas occluded in the walls of the vessel was driven off. If 

 the vessel were exhausted before, or while its temperature 

 was rising, it was found that the brass electrodes attained 

 their iinal temperature much more slowly than did the 

 mercury, owing to the fact that with high exhaustion the 

 heating is not performed by means of convection currents, 

 so the metal would be heated by radiation only, whereas the 

 mercury would be heated by conduction. This caused the 

 mercury vapour to condense on the electrodes. The evacuation 

 was performed by means of a Gaede pump, and afterwards 

 continued by means of charcoal and liquid air. The contents 

 of the vessel were examined bypassing a discharge through E 

 both before and after measurements had been taken to ensure 

 that mercury vapour was the chief gas present, although as 

 all other gases are more difficult to ionize than mercury 

 vapour, traces of them do not affect the result. 



The electrons produced by the incandescent wire were 

 accelerated by a potential difference applied between B 

 and C, and, passing through the gauze C, encountered the 

 reverse field between C and D, so that they were prevented 

 from impinging on I). When, however, the velocity of the 

 electrons reached such a value that they were able to ionize 

 the mercury molecules by collision in the neighbourhood of C, 

 then, owing to the field existing between C and D, the positive 

 ions produced by such ionization were attracted to the elec- 

 trode D ; and the collection of these positive charges was 

 at once shown by a movement of the electrometer-needle. 

 If the potential difference applied between B and C were not 

 as great as that required for ionization, then no such positive 

 ions were produced, which fact was indicated by there being 

 practically no leakage with the electrometer. Thus it was 

 found that at a temperature of 79° C, when the mercury 

 vapour has a pressure of about 0*1 mm., the smallest velocity 

 which the electrons must possess in order that they shall 



