106 



Proceedings of the Royal Irish Academy. 



the apparatus, by heating the platinum spiral X to incandescence by the 

 current from a few cells with the gas turned on. The gas pressure is 

 then slowly reduced until the flame is about 4 or 5 mm. high. It may be 

 noted that the flame burning in this way does not heat the wire X to even a 

 dull red heat, so that there is no question of the wire acting as a source of 

 ions. The air carrying the large ions and nuclei passes through the water- 

 jacket E, and then tlirough the larger vessel S, which has a layer of a few 

 centimetres of water at the bottom. The rate of flow of the air and the volume 

 of S are so chosen that the ions passing out of S and into the other part of 

 the apparatus have attained their stable states. 



.T o mea suring tube 



n/ter 



Atr frcn Geta gmeJ^^— 



Fig. h. 



In order to test the apparatus we made a determination of the coefficient 

 of recombination of small ions, nuclei of all sorts being absent. With no 

 flame burning, filtered air was passed through. Samples of air were drawn 

 off and tested by Aitken's apparatus. When the air was absolutely free from 

 nuclei, tlie X-rays were turned on and the currents measured at each of the 

 terminals in C. Since the rate of flow of air is known, the time interval 

 corresponding to the passage of the air from one terminal to the next is 

 known. Experiments were carried out with different rates of flow. In tliis 

 and in subsequent experiments the zero of time is taken as the instant when 



the gas passes a point 10 cms. in front of the first terminal. Plotting — 



against time, we attained, as usual, a straiglit line, N being the number of 



