M'Clelland and Nolan — Ions produced by Bubbling Air. 57 



the motion of the gasometer being too slow for accuracy. Further observa- 

 tions with more rapid air streams also gave a larger value for the mobility of 

 the last ion in the table ; -00015 is the mean value of all the observations. 



There are two ways in which we may explain the disappearance of the 

 faster ions and the appearance of the slower ions with time. We might 

 suppose that all the ions are present at the very beginning just as in the case 

 of the spraying of water. The slow ions would be present in very small 

 quantity compared with the faster ions. At short time-intervals the small 

 ions would give so much ionisation that the large ions would be obscured. 

 As time increased the small ions would disappear more quickly than the 

 large ions. The consequence would be that the large ions would come 

 into prominence. 



The second way of explaining these results is to suppose that the large 

 ions are not present at the beginning, and that they are formed, as time goes 

 on, from the smaller ions. 



For reasons similar to those given when dealing with the effect of pressure 

 on the mobilities, it is difficult to decide between these two theories. The 

 balance of evidence is in favour of the latter view. The most decisive evidence 

 in support of the second theory is obtained from a consideration of the experi- 

 ments with the very long time-intervals, especially in examining the appearance 

 of the ion of mobility '00015 cm. /sec. With these long intervals the rate of 

 decay of the ionisation is slow, and the total quantity of ionisation does not 

 vary much with the different times. Consequently the percentages of ions 

 present in the different experiments can be compared more definitely. 



On this view we see that the ion of mobility '00034 cm./sec. is formed 

 between 43 and 85 sees, after bubbling. In the case of air bubbled through 

 mercury the ion of mobility '00034 cm./sec. was present 30 sees, after 

 the ionisation had taken place. The formation of the slower ions does 

 not take place so rapidly with the alcohol ions as with the ions derived 

 from mercury. 



The ion of mobility "00015 cm./sec. was not present 433 sees, after 

 bubbling, but it was observed after an interval of 540 sees. So it can be 

 stated that the ion of mobility '00015 cm./sec. is not formed until about 

 500 sees, have elapsed from the formation of the ions. 



It is obvious, from the results obtained in the variation of pressures, 

 that the time which elapses before the appearance of any ion will depend on 

 the pressure of the bubbling. The times given above for the appearance of 

 the two slowest ions relate to the pressure at which most of the experiments 

 were conducted — a pressure of about 8 cms. of mercury. In work to be 



