RECENT ADVANCES IN SCIENCE 469 



decinormal solution of KC1, and 9 per cent, lower than in 

 the case of HC1 at the same concentration. It will also be 

 seen on comparing these data with those of HC1 that the 

 numerical values of both a and 7 are less in the case of KC1 

 than they are in the case of HC1. There is therefore no par- 

 ticular reason for employing the assumption that the true 

 degree of ionisation for HC1 is identical with that for KC1. 

 Thus far we have not succeeded in clearing up the difficulties 

 of the difference between a and 7, but it is just here that Ellis 

 has brought forward some additional evidence which has a 

 notable bearing on the problem. A comparison may be made 

 between the values of the activity coefficients of KC1 derived 

 from electromotive force measurements and those derived from 

 freezing-point determinations by the method recently described 

 by Bates. With the aid of the laws of thermodynamics and 

 of the two assumptions (1) that the two ions in the same 

 solution have equal osmotic pressures and (2) that the true 

 value of the ionisation is given by the conductivity-viscosity 

 ratio, Bates found that the change at o C. of the osmotic 

 pressure it of either ion with its concentration cy is expressed in 

 the case of KC1 by the equation d7r/d{cy)=RT(i -0-0552 (cy) om ). 

 From this equation and the two thermodynamic relations : 



— AF = \ ^d-rr/cy, and — AF — RT log c x a<\l c 2 a 2 



JTT-2 



(where — AF denotes the decrease in free energy involved 

 in the transfer of one mole, of KC1 from osmotic pressure ir x 

 to osmotic pressure 7r 2 ), supplemented by the assumption that 

 the activity coefficient a approaches unity as the salt concen- 

 tration c approaches zero, the activity coefficients corresponding 

 to Bates's osmotic pressure calculations are found to have 

 the values quoted in the second column of the following table. 

 At the same time the activity coefficients calculated directly 

 from Maclnnes and Parker's E.M.F. measurements are recorded 

 in the third column. 



