744 . Dr. S. R. Milner on the Effect of Interionic 



determinations as to the fraction o£ the electrolyte which 

 is dissociated. Finally, the calculated effect is of just the 

 right magnitude to account for the observed depressions in a 

 completely dissociated electrolyte. 



The view that an electrolyte is completely dissociated at 

 all concentrations gets over the long outstanding difficulty 

 of the want of agreement between the observed freezing- 

 point variation and the law of mass-action, for the latter lavv r 

 clearly does not now come into the matter at all. On the other 

 hand, at first sight an equally great new difficulty is raised 

 by the existence of a variation in the molecular conductivity 

 with the concentration, for this can now no longer be ascribed 

 to incomplete dissociation. However, the interionic forces 

 in a completely dissociated electrolyte may be expected to 

 produce an effect on the conductivity as well as on the 

 osmotic pressure, and I hope to show in a later paper not 

 only that such is the case, but also that the theoretical effect 

 is in good accord with experiment. 



It was proved in the paper on the virial referred to above, 

 that in a gaseous mixture of N+ and N— ions confined in a 

 volume V the ions will not be distributed at random in the 

 volume, but that the ions of opposite sign will be on the 

 average slightly nearer together, and those of like sign on 

 the average slightly farther apart, than they would be on a 

 purely random distribution. It is because of this deviation 

 from the random distribution, which is brought about by 

 the action of the interionic forces, that there exists a finite 

 virial for the mixture, for the virial of a purely random 

 distribution of the ions would be zero. The virial "* we are 

 concerned with is defined as 



E = 2 ±4*jr 9 



where q is the ionic charge, and r is the distance apart of 

 any pair of ions. The summation is to be extended with 

 appropriate sign over every pair which can be formed out 

 of the 2N ions in the volume, and then averaged over every 

 possible configuration of the system, due regard being paid 

 to the probability of occurrence of each configuration. It 

 was found that E could be expressed in the form 



E = Nw7i</)(/0 (1) 



In this equation 



RT 



IV = , 



V 



where v is the number of molecules in a gram molecule, 



* Boltzmann's definition of the virial as 5 it*" hHS been employed in 

 this and the preceding 1 paper. Some authors define the virial as A^Kn 



