Nolan and Enright — Experiments on Large Ions in Air. 97 



reached the testing vessel. The effects produced by SO2 and NH3 are shown 

 in the graphs of fig. 1. The graph A is the current-voltage graph obtained 

 from an ordinary Bunsen flame. The saturation value of the current is 

 attained at a voltage corresponding to a mobility of '00033. The curve is con- 

 vex upwards, indicating the presence of ions of higher mobilities. It has been 

 well established that various discrete groups of ions are generally present.' 

 These groups have been investigated in the ionisation produced by spraying, 

 bubbling, oxidation of phosphorus, etc. Corresponding results with the flame 

 ions have been obtained by Dr. H. Kennedy in this laboratory. He found 

 that the ion of mobility cmy« '0012 was usually present in the flame-gas along 

 with the well-known ion of mobility '00033. Similar results for flame-gas have 

 been found by Lauster.- To show up the groups separately it is necessary to 

 take a great number of close-lying readings of current and voltage. As we 

 are not specially interested in showing up the groups in this work, we have 

 taken more widely spaced observations, and have drawn the general outlines 

 of the curves without attempting to show details. 



Curve B shows tlie effect of supplying SOo plentifully to the flame. The 

 number of ions is increased very considerably, and there is also a decided 

 change in the mobility. The bulk of the ions now have a mobility of about 

 •00016, about half the normal value. The effect of mixing SO2 with the 

 products of combustion at a short distance from the flame is shown by 

 curve C. The most marked effect is a considerable reduction in the number 

 of ions. There is also a change in mobility, some of the ions now having a 

 mobility lower than the normal value. 



Curves D and E show the effect of ammonia. This gas when applied at 

 the seat of combustion has no effect on the number of ions, but when present 

 in sufficient quantity it changes the mobility from '00033 to 'OOOIS in the 

 same fashion as SOj. The curves obtained for smaller admixtures of 

 ammonia lie between A and D, and seem to indicate a mixture of the two 

 ions just mentioned. When ammonia is mixed with the products of com- 

 bustion (Curve E) it produces a considerable reduction in the number of 

 ions and also reduces the mobility to half. 



From these curves certain conclusions can be drawn. Ammonia, as already 

 found in the case of the hydrogen flame, cannot act as a starting-point 

 for nucleation. Sulphur dioxide and ammonia, if supplied to the flame or sub- 

 sequently to the flame-gas, cause the ions to grow beyond their normal size. 

 The transition in size is not gradual but abrupt. When the alteration is 



' P. ,r. Nolan, Physical Review, Sept., 1921. 

 2 Lauster, Zeit fur Physik, 3, 5, p. 396, 1920. 



