THE HYDRATE THEORY OF SOLUTIONS 205 



mobilities of kathion and anion was equal to the molecular 

 conductivity 



for the given dilution. This method of representation un- 

 doubtedly possesses certain advantages, in view of the fact that 

 the coefficient of ionisation cannot be accurately determined 

 except in the most dilute solutions. In moderately dilute 

 solutions a fair approximation can indeed be reached, but the 

 coefficients for concentrated solutions are absolutely unknown 

 quantities, and in the case of fused salts no method has yet been 

 devised for determining even approximately the value of the 

 coefficient, so that it is actually impossible at the present time 

 to say whether the whole or only a part of the salt is directly 

 concerned in electrolysis. Arrhenius's method of dealing with 

 ionic mobilities differs from that of Kohlrausch in assuming that 

 the ionic mobilities are substantially constant in dilute solutions, 

 and that the decrease of molecular conductivity when the con- 

 centration is increased is due to decreased ionisation, and not to 

 decreased mobility. Thus he wrote 



Aoo = z^ 00 + 2/00 



but (in place of K = u„ + v n ) 



\ n = a ill 00 + v 00) 



— a novel method of representation that has proved to be of very 

 great advantage in the study of the dilute solutions which have 

 hitherto formed the principal subjects of investigation. 



In a few exceptional cases, such as that of heated glass 

 (Bousfield and Lowry, Proc. Roy. Soc. 1902, 71, 45) or concentrated 

 caustic soda (ibid. Phil. Trans. 1905, 204, 253-322), the mobility 

 of the ions may be the most important factor in determining the 

 conductivity of an electrolyte, but in all the more ordinary 

 solutions the variations of mobility are relatively small. The 

 factors governing them (ionic size, viscosity, etc.), which are 

 discussed below, are therefore of secondary importance com- 

 pared with the fundamental problems associated with the 

 coefficient of ionisation — as, for example, why a solution of 

 hydrogen chloride in water should be a good conductor, a 

 solution in alcohol a moderate conductor, a solution in ether 

 a very poor conductor, and a solution in benzene an excellent 

 insulator. 



In considering this problem, the theory put forward almost 

 simultaneously by J. J. Thomson {Phil. Mag. 1893, v. 36, 320) 



