24 



MR. 0. W. RICHARDSON ON THE IOX1SATION 



40 



30 



20 



10 



(2) 



700 



1300 



> 



lv*-" 



_. I > ^r^ . &- .l.-o-- 



800 900 1000 1100 1200 



Temperature : Degrees Centigrade. 

 Fig. 9. 



deduced by the author for the negative ionisation, and can be expressed by means of 

 the formula A0*e~*'", where A and 6 are constants and is the absolute temperature. 

 The constant b which measures the work done in setting free an ion is, in general, 

 much smaller in the case of the positive than the negative ionisation. This may be 

 tested, as the writer has explained in previous papers, by taking logarithms, when, 

 if L is the current and A' a new constant, 



log, L/0* = A'-6/(2'300). 



The value of the logarithm should therefore be a linear function of 1/0. Values of 

 log 10 L/0*, where L is the current per unit area of surface for two wires of different 

 lengths and diameters, have been plotted against 1/6 in the accompanying diagram. 



The extent to which the points fall on two straight lines furnishes a test both of 

 the applicability of the alxrve formula and also of the nature of the agreement between 

 different wires. The data for the wires tested are (l) diameter = O'Ol centim., 

 effective superficial area = 0'223 sq. centim., positive ionisation-points thus x, 

 negative ionisation-points thus (x); (2) diameter = 0'02 centim., effective area = O'GG 

 sq. centim., positive ionisation-points thus , negative ionisation-points thus O. 



It will be observed that the points for the positive leak, and also for the negative 

 leak, fall very nearly on the same straight line for the two wires. This shows not only 

 that the leak may be expressed by a formula of the above type, but also that the 

 constants A and b which enter into the formula are the same for both wires. The 

 tangent of the angle the above lines make with the axis of 1(0 is a measure of the 

 work done in setting free an ion. This quantity is evidently much less for the 

 positive than for the negative ions. The value of this work is conveniently expressed 

 in calories per gramme molecule of ions, a gramme molecule of ions being the amount 

 which would occupy 22 '4 litres if in the state of gas at C, and 760 millims. 



