100 Proceedings of the Eo>jai Irish Academy. 



As the mean velocity of the air thix)ugh the tubiug was 35 cm/sees., it will 

 be seen that the ions were exposed to the temperatme for nearly three 

 seconds before they eame under observation. 



Even if the constitution and dimensions of the ion are unchanged by 

 heating, its mobility in the hot gas will be increased. This is due to t\FO 

 causes : the change in density of the gas and the change in the velocity of 

 thermal agitation of the ion. Since in these experiments the velocity of the 

 air-stream is measured by the displacement of cool gas, the first of these 

 changes is automatically allowed for. If the ion i-etains its constitution, we 

 should expect that owing to the second cause its mobility should increase 

 proportionately to the square root of the absolute temperature. 



Before dealing with the measurements of mobility at different tempera- 

 tures we give the results of an experiment made with a steady supply of 

 ions passing into the apparatus, the temperature being gradually raised. 

 The following were the values of the current to the central terminal, a 

 saturation voltage being applied. They are thus a measure of the number 

 of ions coming into the measuring apparatus. 



Temperature. Current. 



8= C. 120 



50^ C. 120 



81° C. 120 



106= C. 119 



128= C. 112 



147= C. 100 



168° C. 86 



194° C. 64 



The reduction in the number of ions at high temperatures is due to the 

 increased recombination of ions which have had their mobility increased. 

 As we shall see, this increase in mobility is so marked that it must be 

 attributed to a radical change in the ion itself. It is noteworthy that the 

 increased recombination does not set in until the temperature has gone 

 above 100= C. 



In fig. 3 are shown the cuiTeut- voltage graphs plotted for various tem- 

 peratures between 17° C. and 167° C. There is not much difference between 

 the curves for 17° C, 55° C, and 100= C. The number of ions is unaltered 

 in agreement with the results given in the table above. Each curve shows 

 saturation at very nearly the same point coiTcsponding to the ordinaiy ion 

 of mobility -OOOSS. The increasing conve3dty of the cunes as the tempera- 

 ture increases shows that a certain number of the more mobile ions are 

 appearing. The smaU shift in the satur-ation voltage between 17= C. and 



