200 J. A. POLLOCK. 



as to the character of the foundation of the complex struc- 

 ture which is indicated. 



It is well known, since Aitken's notable work on the 

 subject, that under ordinary circumstances the air is 

 crowded with particles in suspension on which the water 

 vapour condenses into visible drops if the air becomes 

 slightly supersaturated. These particles, whose number 

 varies greatly from time to time, can be removed by filtra- 

 tion of the air through cotton wool, or by settlement with 

 the drops formed during repeated expansions, and as shown 

 by Mr. 0. T. R. Wilson 1 are, in fact, necessary as nuclei for 

 visible condensation with supersaturations less than four- 

 fold. In general these nuclei are electrically uncharged, 

 and whatever their exact nature are conveniently known 

 as dust particles. 



In describing an atmospheric ionisation recorder, which 

 in their hands has already yielded important results, MM. 

 Langevin and Moulin 2 mention that the simultaneous vari- 

 ations in the numbers of large and small ions in the air are 

 opposed in direction, a fact which is also shown in the 

 measures made on the few occasions when we have taken 

 continuous observations of both classes of ions in Sydney. 

 MM. Langevin and Moulin further state that the number 

 of large ions is the greater the more numerous the [dust] 

 particles in the air, and they consider the large ions as 

 created by the attachment to these neutral particles of 

 small ions in the process of their diffusion. Further evidence 

 in support of such a view is afforded by the results of the 

 following experiments. 



With the testing pipe described in the previous paper, all 

 the large ions as well as the small ones can be withdrawn 

 from a stream of air ; in this way the pipe may be used as 



1 Wilson, Phil. Trans. A, 189, p. 265, 1897. 

 2 Langevin and Moulin, Le Radium, 4, p. 218, 1907. 



