46 ELISHA MITCHELL SCIENTIFIC JOURNAL 



which they occur in the mineral. This can be successful 

 only in case the chilled melt is annealed thoroughly. Merely 

 allowing- the melt to cool will not give the desired result. 



Another matter in which the phase rule has been of great 

 value is in analysis in cases where a pure compound cannot be 

 isolated. Under these circumstances, the ordinary methods 

 of gravimetric and volumetric analysis are inapplicable. Bv 

 methods based on the phase rule, it has been possible to deter- 

 mine with accuracy the composition of compounds crystalliz- 

 ing from a molten mass of metal. 



Analyses of efflorescent salts containing water or acid of 

 crystallization are necessarily inaccurate because there is no 

 way of drying the compounds enough to remove all the mother 

 liquor without running the danger of removing some of the 

 volatile component. By the new methods this difficulty no 

 longer exists because the compound can now be analyzed 

 while in the solution. 



As yet no one has applied these methods to the analysis of 

 colloidal precipitates, but there seems to be no reason to sup- 

 pose that this cannot be done. The methods of the phase 

 rule have already been used with success in determining the 

 composition of hydrated beryllium sulphate. The whole 

 question of colloidal precipitates has taken on a new aspect 

 since van Bemmelen began the study of the equilibrium rela- 

 tions with reference to the phase rule. 



No matter where one turns, in the whole field of inorganic 

 chemistry, one finds the phase rule useful as an instrument of 

 research. This has come about in the last few years and is 

 but an earnest of what is to be done in the future. When the 

 domain of the phase rule has been extended to cover the 

 whole of organic chemistry and of electrochemistry, every 

 one will then admit the truth of 1113^ theses "that the phase 

 rule offers the rational basis for the classification of all chem- 

 ical phenomena and that it is perhaps the most valuable in- 

 strument of research that we possess." 



