﻿Effect of Clouds on Ionization. 301 



tube being — 22 for Curve I. and —4 for Curve IT. It is 

 interesting to note the striking difference between the two 

 curves. 



A closer inspection of the curves shows that the results 

 of the expansions present interesting features at expansions 

 of magnitudes 1*25, 1'2S, 1*31, and 138. Now these are 

 the identical values of the expansions which C. T. R. Wilson 

 found 



(a) to start catching the negative ions, 



(b) to catch all the negative ions, 



(c) to start catching positive ions, 



(d) to catch the nuclei of cloudy condensation. 



The most remarkable feature, however, is that whereas the 

 cloud formed on the negative ions under a potential of 22 volts 

 diminishes the current by 26 per cent., and under 4 volts by 

 74 per cent., yet the supposed increase in the number of cloud 

 particles when the positive ions are caught at an expansion 

 of 1*31 produces a further decrease of only some 3 per cent. 

 This result may be due to either of the following causes : — 



(a) the number of positive ions caught is, owing to the 

 greater efficiency of the negative ions, enormously less than 

 the number of negative ions caught when both are present in 

 the same space, and when the expansion is known to be big 

 enough to catch both kinds of ions, or 



(b) when the density of a cloud exceeds a certain value 

 a further moderate increase in the density produces very 

 little further diminution of the ionization current through 

 the gas. 



There is no doubt that cause (a) may come in as J. J. 

 Thomson found in determining the ionic charge *, unless 

 care be taken to make the expansion extremely rapid. 



Since in our experiments the expansion was very rapid, 

 and since, in addition, the further diminution in the current 

 due to the positive ions at their maximum effect (expansion 

 1'36) is only some 3 per cent., we incline to the second of 

 the two explanations. It may be noticed, however, that 

 under a voltage o£ — 22 the increase in the density of the 

 cloud at 1*38 due to the catching of the nuclei of cloudy 

 condensation is well marked, whereas it is almost inappre- 

 ciable under —4 volts, when the current had already suffered 

 the large reduction of nearly 80 per cent, before these 

 nuclei started to come in. Thus the greater the percentage 

 reduction of the current already produced by the cloud, the 



* 'Conduction of Electricity through Gases,' 2nd ed. pp. l~>7. 168 

 (1906). 



