MOLECULAR COMPOSITION 169 



(CaSiOs and Ca2Si04), except near the ends of the series, where the pure 

 oxides remain over in excess. If combination occurred only at the mo- 

 ment of crystallization, the accompanying energy change would be of a 

 different order of magnitude, according as the component appearing in 

 excess is one of the original oxides or one of the silicate combinations. 



Characteristics of the coOled Products — some Definitions 



The next important step, after differentiating the various situations 

 through which a cooling magma passes, is to learn to recognize the dis- 

 tinctive characteristics of the products which result — the glasses, isomor- 

 phous mixtures, eutectics, and new compounds. 



Up to the time of Lagorio but little attention was paid to the glasses 

 in the study of rock synthesis. Lagorio endeavored to make use of them 

 as hypothetical standard solvents in which various oxides could be dis- 

 solved to form rocks of different character, making it the counterpart in 

 the mineral world of water in aqueous solutions. Each rock was to be 

 considered as a mineral or group of minerals in solution in one of two 

 common solvents. This viewpoint laid much emphasis upon the impor- 

 tance of rock glasses, but it has not yet contributed materially to the elu- 

 cidation of rock formation. Doelter has endeavored to ascribe individual- 

 ity to glasses, and has experimented with a great many glasses of different 

 compositions without arriving at any conclusion of general interest. 

 Barus, in his determinations of physical constants, esteemed it a matter 

 of secondary importance whether a mineral cooled to a glass or crys- 

 tallized. Modern physical chemistry ascribes little individuality to 

 glasses as such, but considers them merely as under-cooled liquids which 

 can be crystallized only with the greatest difficulty, or not at all, and which 

 in consequence possess only tentative and general properties. They ac- 

 cordingly possess about the same significance as any other incomplete re- 

 action which is caught in transition between states. It is much more 

 important to be able to define these end states and their relations than to 

 be able to fix the progress of the interrupted operation in individual cases. 

 We find obsidians in which some crystallization has occurred, and rocks 

 in which but a small portion of the total content is glass, all of which 

 merely serves to show that the progress of individualization of a magma 

 may be interrupted anwhere whenever the cumulative action of the op- 

 posing forces has reached a sufficient magnitude. It may also be resumed 

 on the advent of new, favorable conditions, such as would be produced if 

 a supersaturated magma were to come in contact with crystals of the 

 excess component. 



An isomorphous mixture, as its name indicates, is an intimate (crystal- 

 line, not vitreous) mixture of two or more component minerals in which 



