90 Proceedings of the Roynl Irish Acailemy. 



16,000, we get for a the value ">. In this calculation we have assumed that 

 all the nuclei of Barus are of the nature of those necessary for the pro- 

 duction of large ions. "We shall return to this point later. The values of q, 

 therefore, derived from the resiiltS given by Langevin, Pollock, and by 

 our own observations, seem to agree much better with the results of 

 direct measurements than those deduced from the values of n given by 

 observers working with Ebert's apparatus, when we take into account the com- 

 bination of small ions with large ions of opposite signs and with neutral nuclei. 

 Piutherford* has drawn attention to the effect of finely di\-ided particles 

 suspended in the air in hastening the loss of conductivity of the air, when the 

 source of ionisation has been withdrawn, and, as has been pointed out, tliis 

 effect will be relatively of very great importance in the case of air exposed 

 only to natural sources of ionisation. 



The point referred to above — namely, whether only some or all of the 

 condensation nuclei of Aitken and Bams are identical in character with the 

 nuclei required for the formation of large ions — is one of very gi'eat interest. 

 In a series of his experiments Pollock failed to notice any condensation on 

 large ions by expansion. This, however, is explained by the fact that the 

 " dust " was removed by a preliminary process, and very probably with it 

 were removed the large ions and the nuclei from which they are formed. In 

 fact, these experiments are in agreement with oiir observations that no large 

 ions and no fresh nuclei are produced in a closed vessel containing air from 

 which these nuclei have been removed. Barusf gives a very detailed account 

 of daily values of the number of nuclei present per c.c. during the greater 

 part of the years 1903 and 1904. These values were obtained by condensation 

 methods.- They range from 10,000 to 100,000, and are, therefore, of the same 

 order of magnitude as we should expect if they were all of the same nature 

 as the nuclei from which large ions are formed. Similar values have been 

 obtained by Aitken. 



All the observations on the number of large ions per c.c. in the 

 atmosphere recorded in this paper were taken at the College buildings situated 

 in the city, and the question aiises to what extent the numbers observed may 

 be pecuhar to the air of a city, and what results might be expected in purer aiiv 

 Other work at present being carried on in this laboratory has shown that, in 

 the gases drawn from flames of various sorts, the ions decrease in mobility 

 with time until the same value is reached as we have found for the large 

 atmospheric ions, and their mobility then remains constant. This is in 



*PM1. Mag., 1897, pp. 422-440. 



t A continuous record of Atmospheric Ifucleation. Smithsonian Contributions to Knowledge,, 

 vol. xxsir. 190.5. 



