740 JOHN JOHNSTON 



component, the mutual solubilities^ enter as factors in the result. 

 These are affected by pressure because it changes the relative con- 

 centrations or activities of the several components. This influence 

 is slight with solids and liquids, because their compressibilities are 

 very small; but in systems with a gaseous phase it is very great, 

 because the concentration of a gas is directly proportional to the 

 pressure, so that in such systems change of pressure will have a 

 marked influence on the relative concentrations of the several 

 reacting components. 



In accordance with this, the influence of uniform pressure upon 

 solubility and upon chemical equilibrium in condensed systems — 

 that is, systems in which there is no vapor phase — is slight, and 

 for practical purposes neghgible in comparison with the influence 

 of temperature or change of composition of the solution. Thus the 

 few trustworthy experiments made hitherto show that a pressure 

 of i,ooo atm. changes the solubility of a salt in water by only about 

 I per cent of its value — a change which may readily follow a tem- 

 perature change of a few degrees or a slight change of composition 

 of the solution.^ Again, at ordinary temperatures the liquids water 

 and methylethylketone are only partially miscible at i atm., but 

 they become completely miscible at about i,ioo atm.,^ whereas at 

 I atm. this does not take place until a temperature of 152° is reached. 

 The opposite effect — i.e., unmixing by increase of pressure — ^could 

 equally well be produced, just as it may be by increase of tempera- 

 ture. But it seems probable that neither of these effects is of 

 importance in liquid mixtures of silicates, because liquids which 



^ It is to be remembered that mutual solubilities can be altered by the addition 

 of a third component, even though it forms no part of the solid phase which separates. 

 Thus, as is evident from fig. 6 of Rankin's paper {Am. Jour. Sci., XXXIX [1915], 

 1-78), a mixture composed of 60 per cent SiOa, 40 per cent AI2O3, on cooling deposits 

 AlaSiOs (sillimanite), whereas in presence of CaO, in amount ranging from 3 to 30 

 per cent of the whole, the soHd phase which separates is piire AI2O3. This illustrates 

 how even a small change in composition of a solution may alter the course of crystal- 

 lization. 



2 E. Cohen and L. R. Sinnige, Z. physik. Chem., LXVII (1909), 432; LXIX (1909), 

 102; E. Cohen, K. Inouye, and C. Euwen, ibid., LXXV (1911), 257. These authors 

 give a critical resume of earlier work along this line. 



3 P. Kohnstamm and J. Timmermanns, Proc. K. Akad. Wetenschappen, XV 

 (1913), 1021-37. 



