14 Mr. W. Sutherland on Ionization in Solutions 



Here water is again exceptional in a marked way. The 

 values of 1Q0A, ?7 for water and the alcohols fall according to 

 the series 136, 64, 56, and 52, the values of dielectric 

 capacity being 82, 34, 24, and 21. There is some suggestion 

 of X ?7 being a function of K. 



In my previous paper the assumption that polarization of 

 the dielectric could be ignored led to the conclusion that 

 ?v ?7o must be proportional to K; imperfect data for water 

 and the alcohols seemed to confirm this. But Table J. affords 

 proof that A ?7 is only slightly dependent upon K. The ex- 

 ceptions which we have encountered will probably be under- 

 stood only when the nature of the ions carrying current in 

 each solvent is investigated. In our comparisons the ions 

 are supposed to be always the same when derived from a 

 given solute. 



5. A Special Consideration of Typical Aqueous 

 , Solutions. 



In my previous paper I took solutions of NaCl, BaCl 2 , 

 (NH 4 ) 2 S0 4 , and Zn!SQ 4 as types and tabulated for them the 

 values of X/A , y l?h anc ^ ^vl^oVo for concentrations ranging 

 from O'OOOl gram-equivalent per litre up to saturation or 

 near it. These data were represented in a graph (loc. cit. 

 p. 171) which brought out the interesting fact that \r]/^oVo 

 passes through a minimum at a certain concentration of the 

 solution in each case. Now for a given solute in water 

 (A, + A 2 )/Ao=- Vo/V an d with i = l equation (13) can be written 

 in the form 



^^] + b(v in )i n Jv (15) 



This simple formula is to express the molecular specific 

 conductivity of a solution at all concentrations up to satu- 

 ration. The values of b derived from Kohlrausch's data 

 are for NaCl 4'035, BaCl 2 6*68, (NH 4 ) 2 S0 4 7*856, and Zn$0 4 

 31*55. These yield the following comparison between 

 the experimental values of \ r)J\w and those calculated 

 by (15). 



