DOUBLE SALTS 153 



of as a waste product, whilst the carnallite that had separated 

 from it is again worked up in the same manner as the original 

 mineral. 



Van't Hoff, in his Chicago lectures on " Physical Chemistry 

 in the Service of the Sciences," has pointed out that application 

 of the physico-chemical method to the carnallite problem reveals 

 all the possible methods of splitting it up, methods of which 

 the above process is only one amongst several. He proceeded 

 to suggest another in which, instead of effecting the separation 

 of the KC1 by use of the solvent water and concentration of 

 the solution obtained, temperature changes are resorted to in 

 order to produce transformation, melting, and solidification. 

 The process advocated (which has been patented by Van't HoflPs 

 late colleague, Meyerhoffer) is as follows : 



The carnallite is heated in closed vessels to 168 , the 

 transition point, when three-fourths of the KC1 separates 

 out and is removed by filtration, the water split off retaining 

 in solution all the MgCl 2 and one-fourth of the KC1. The 

 liquid is then cooled to 115 , when carnallite crystallises out, 

 leaving a liquid which is an almost pure melted hydrate of MgCl 2 . 

 The solid carnallite is separated by filtration and again worked 

 up, whilst the liquid, which at ordinary temperatures becomes 

 solid, constitutes the waste product. 



The essential difference between the two methods — the 

 empirical and that based on the application of physico-chemical 

 methods — is that in the latter the MgCL separated from the 

 KC1 is obtained, not as a mother liquor, but in the more 

 convenient form of a solid. 



The number of illustrative cases of the application of physico- 

 chemical methods to the special problems encountered in the 

 formation and decomposition of double salts could, of course, 

 be greatly increased ; but even the slightest respect for pro- 

 portion makes it necessary to pass on to another and altogether 

 different aspect of this subject. 



The physico-chemical method, as developed on the basis of 

 the phase rule, considers the existence of double salts in its 

 dependence on temperature, pressure, and concentration, that 

 is, on factors which it is possible to vary arbitrarily ; and in 

 this it differs essentially from the chemical method, whose 

 object it must be to find an answer to the question : What 

 relations must exist between two salts in order that — apart 



