Ions formed in Gases at High Pressures. 



179 



If we put n = 2, A=166, [pv^-504:, and £ = '00316 

 (p being expressed in cms. of mercury) } we obtain numbers 

 for pv which agree very well with those found experimentally 

 by Langevin. The greatest difference amounts to a half per 

 cent., and is therefore well within the experimental error. 

 Langevin's numbers for pv are tabulated alongside those 

 calculated from the above formula in the following table. 



p (in cms.). 



pv/76 (found). 



pv/7Q (calculated). 



7-5 



647 



645 



20 



580 



580 



41-5 



530 



5305 



76 



510 



513 



142 



505 



506-5 



Since there are two constants beside n in the above formula, 

 which has only to fit five observations, it might be thought 

 that some other number would do equally well. It was found, 

 however, that both n = l and n = 3 gave a far worse agreement 

 with the numbers, whilst higher values of n are readily seen 

 to be impossible. 



These results lead us to interpret Langevin's numbers in 

 the following manner: — The negative ions, which occur at 

 high pressures and which have a velocity of 504 cms. per sec. 

 under a field of one electrostatic unit per cm. at atmospheric 

 pressure, dissociate as the pressure is reduced. They then 

 give rise to two molecules of gas and a smaller negative ion 

 which would have a velocity of 670 cms. per sec. under a 

 field of one electrostatic unit per cm., if it could exist in air 

 at atmospheric pressure. The actual velocity measured at 

 any pressure is the average for the actual numbers of the two 

 kinds which occur. 



Since the velocity of the positive ions at high pressures is 

 smaller than that of the negative, the theory requires that 

 they should dissociate similarly at low pressures. The expe- 

 riments show that there is no evidence of dissociation until 

 pressures of about 7 cms. are reached, when the velocity of 

 the positive ions experiences an increase. 



The negative ions for which pv/76 = 670 do not necessarily 

 represent the ultimate stage in the dissociation. It is probable 

 that these dissociate further at lower pressures into corpuscles, 

 which would have a much higher value for pv. 



N2 



