666 Recombination of Ions in Air. 



Ill this curve the ordinates represent numbers proportional 

 to the relative values of a, while the abscissse represent the 

 temperatures o£ the air in degrees centigrade. 



As will be seen from this curve, it was found that as the 

 temperature of the air increased the value of a also increased, 

 at first fairly rapidly, and at the higher temperatures much 

 more rapidly. The rise in temperature of the gas appears to 

 cause the value of the coefficient of recombination to increase 

 to a considerable extent. 



The density of the air decreases of course with the rise of 

 temperature, but, as in the previous case, this need not be 

 taken into account ; for it was shown in the paper on the 

 relation between pressure and a, that over the range of 

 densities of the air which are involved in this experiment 

 the value of a was practically a constant, so the change of 

 density of the air need not be taken into account in con- 

 sidering the relation between a and the temperature. 



This relation between a and the temperature does not 

 appear to be a very simple one, for the curve showing the 

 connexion between them does not seem to be capable of being 

 represented by any simple mathematical formula. Whatever 

 the changes are which take place among the ions, either in 

 the individual ions or in their relation to one another, when 

 the gas is heated they w^ould appear from this to be somewhat 

 complicated. As seen from the curve, however, the relation 

 between the coefficient of recombination and the temperature 

 of the gas seems to be quite a definite one. 



Summary. 



The results which have been obtained during this investiga- 

 tion may be summed up as follows : — (1) The rate of 

 recombination of ions in air follows the same law, namely, 



— = — an^, at different temperatures, at least over the range 



of temperatures investigated, that is between 15° C. and 

 300° 0. (2) A rise in the temperature of the air causes 

 a considerable increase in the value of the coefficient of 

 recombination, and the relation between the temperature and 

 this coefficient appears to be of a somewhat complicated 

 nature. 



In conclusion I wish to express my appreciation of the 

 kindly interest shown and the advice given by Professor 

 Thomson during this investigation. 



Cavendisli La"boratorj, 

 Cambridge, July 29, 1903. 



