40 



MR. O. W. RICHARDSON ON THE IONISATION 



avoid the occurrence of ionisation by collision, was frequently adopted in this 

 investigation. 



1501 



800 



10O 



20O 



300 400 



Volts 



Fig. 12. 



500 



600 7OO 800 



The temperature variation of the approximately saturated leak in air at 

 atmospheric pressure was also examined with the wire 0'2 millim. diameter in the 

 glass tube apparatus. The numbers obtained are given in the following table. The 

 effective-area of this wire was 0'66 sq. centim. : 



These numbers may be compared with the values for the leak in oxygen at different 

 temperatures given on p. 23. In doing this it is necessary to multiply the latter by 

 a factor of about 3 '3, which varies slightly with the temperature, to bring the 

 positive leak to its value at atmospheric pressure. Botli sets of numbers have also to 

 be divided by the effective areas of the respective wires. When this is done, it will 

 be seen that the maximum leak in oxygen is greater than that in air at all 

 temperatures, though the latter increases more rapidly with the temperature. Both 

 the positive and negative leaks in air obey the formula A0 ! r~" /2 *, the values of the 

 constants which determine the energy change associated with the formation of one 



