112 Proceedings of the Royal Irish Academy. 



point where there is no longer available a charge apiece for half of the 

 nuclei. Further reduction in the ionisation leads to a reduction in the 

 number charged. 



These results enable us to conclude with some certainty that the large 

 ion in the atmosphere carries a single electronic charge. The bulk of the 

 large atmospheric ions in regions free from contamination are probably, 

 as we have previously suggested, due to the action of ultra-violet light 

 on moist air. They gain their charge from the weak natural ionisation 

 of the atmosphere, and, therefore, in view of the results given above, 

 carry pi'obably only a single charge. But it may be said that we have 

 no evidence that the greater part of the atmospheric ions at low levels 

 are not due to flames and such sources which, as we have seen, tend to 

 produce trebly charged ions. Our results show that a connexion exists 

 between the charge per ion and the distribution of the nuclei between 

 charged and uncharged of such a kind that if the uncharged are strongly 

 in excess no multiple charges are likely to exist. There are no direct 

 observations available, as far as we know, of the number of nuclei and 

 large ions in the atmosphere at any one time, but values of the order of 

 20,000 fo 30,000 electronic charges per c.c. are obtained for large ions in 

 localities where the observations of Aitken and Barus would suggest the 

 presence of nuclei in concentrations greater than 50,000. Again, McClelland 

 and Kennedy give a curve' showing the production of large ions (by reaction 

 between the nuclei and small ions due to the natural ionisation) in atmo- 

 spheric air freed from its large ions. In four hours over 80 per cent, of the 

 original ionisation was reproduced. Seeing that decay had been going on 

 during this time, it is clear that the large ions must have been considerably 

 less than half of the total nuclei originally present. On these grounds, 

 therefore, in view of the results given above, we conclude that it is most 

 unlikely that atmospheric ions carry other than single electronic charges. 



There is another way in which we may test our results in application to 

 ionisation in the atmosphere. When equilibrium is reached the following 

 relation should hold : — 



q = aN- + iiNn. 



Where q is the number of small ions of one sign pi'oduced per second, N is 

 the concentration of small ions of one sign and n the concentration of nuclei 

 of all kinds. When N is small the first term on the right is negligible, and, 

 therefore, q = i}Nn. We have no simultaneous observations of N and n, 

 but we have the observations of M'Clelland and Kennedy on small ions and 



' McClelland and Kennedy, Proc. Roy. Irish Acad., vol. xxx. A, 1912. 



