I ART. 10. — K. TKî:T)A : RTITDTES OX THE 



Introduction. 



The theory of clikite sohitious, with which the name of 

 VAN t'Hoff is most prominently associated, has been one of the 

 chief agencies in the rapid development of physical chemistry in 

 the last two decades. Bnt its applicability is by its very nature 

 limited, and for further progress a more general theory, such as 

 Avill enable us to treat quantitatively problems of solutions of 

 any composition, must be worked out. This want has been 

 generally felt, and attempts to meet it have been made by 

 various physicists and chemists. 



Among these the elaborate molecular theory of binary mixt- 

 ures ])y VAN DER Waals stauds preeminent, and has led to 

 many investigations theoretical as Avell as experimental. But 

 his theory is bound uj:) with the equation that bears his name, 

 and requires complicated mathematical apparatus in solving even 

 apparently simple problems. For systems in Avliich chemical 

 reactions take place, or in which the number of components 

 exceeds two, the difficulties become so great that no noteworthy 

 progress has as yet been achieved in these directions. 



A simpler method of procedure might perhaps lead to the 

 goal more quickly. It has been demonstrated by the investigations 

 of Dahms, Hartmann, Lechatelier, Ltnebarger, Schroeder, 

 VAN Laar, Young, Zawidzki, and others, that there are some 

 solutions in which the quantitative relations of the heterogeneous 

 equilibria are remarkably simple. These approximate pretty 

 closely to what has been called the " ideal solusion." But the 

 majority of the solutions hitherto studied show more or less 

 marked deviations from these simple relations. Are we to con- 



