Nolan and Enright — Experiments on Large Ions in Air. 99 



available voltages. The number of ions is increased, the increase being 

 greater in the case of the positive ions. Saturation is attained with a voltage 

 approximately four times that sufficient Lo saturate the ordinary ions, thus 

 indicating the existence of ions of mobility about -00008. In addition to the 

 diffei'ence in quantity, there is a difference in the character of the positive 

 and negative ionisation, the negative containing the greater proportion of 

 ions of higher mobility. Whether the increase in the ionisation is due to 

 the NaCl supplying extra nuclei or extra ions to the flame is not clear, but 

 the unsymmetrie character of the increase suggests the latter alternative. 

 The stable nuclear size has undergone a further increase. The action 

 of NaCl in this respect is not, however, as remarkable as that of phos- 

 phorus, as in the ionisation from phosphorus McClelland and P. J. Nolan' 

 found separate groups of ions not only of mobilities ■00031, -OOOlo, and 

 •000085, but as low as -OOOOS:! 



Effect of Variation of Temperature on Large Ions. 



Since McClelland's original experiments it has been known that the 

 mobility of the ions produced in a flame diminishes very rapidly as the gas 

 is removed from the flame. We have seen that if sufficient time is allowed, 

 the bulk of the ions normally attain a mobility of about "OOOSS, but that 

 some other groups are generally present. If this be regarded as the equili- 

 brium size peeiiliar to normal temperatures, it is important to know how the 

 equilibrium size varies when the temperature is varied. There are two ways 

 of approaching this experimentally. We could measure the mobilities of 

 the ions coming immediately from the flame at different temperatures, with- 

 out allowing any cooling. On the other hand, we could allow the flame-gas 

 to cool so that the ions attain their normal size, and then measure their 

 mobility when the gas is heated. We tried the latter method in the first 

 instance. The gas was drawn off from the flame and kept for a sufficient 

 time so that almost all the ions had reached the -00033 mobility. The gas 

 was then passed through the testing apparatus, which was heated to a steady 

 known temperature. The heating was produced electrically, and the heating 

 coils surrounded not only the measuring vessel but the tube leading up to it 

 for a distance of one metre. The temperature was measured by two indepen- 

 dent platinum-iron thermo-couples of small heat capacity. The heating 

 coils were wound in sections, and by controlling the current it could easily 

 be arranged that both thermo-couples indicated the same steady value of the 

 temperature over a long interval without a fluctuation of more than 1° C. 



' McClelland and P. J. Nolan, Proc. Roy. Irish Acad., vol. xxxv, A, p. 1, 1919. 



