. - - SINGLE MINER^VLS AND MINERAL MIXTURES 163 



natural rock formation, and are, therefore, an essential part of the 

 problem. The disappearance of the volatile ingredients from extruded 

 magmas has left them viscous and caused obsidian formations. Their 

 presence lowers the formation temperature, gives mobility to the 

 fluid magma and larger individual crystals. It is through the effects 

 produced by the volatile components also that one of the chief effects of 

 pressure as a controlling force in rock formation has been brought to 

 light; the pressure serves to retain the fugitive ingredients in the rocks, 

 and these are probably chiefly responsible for the low formation tempera- 

 tures. No doubt the field experience of the petrologist will suggest to 

 him many other relations which result from the same causes. 



Mixtures of Minerals 



It now remains for us to consider mixtures of minerals from the labo- 

 ratory viewpoint. Given a homogeneous mixture of two minerals in the 

 molten condition, contained in a suitable crucible, and suppose them to 

 be in complete equilibrium for the prevailing temperature. We will also 

 suppose that neither of the components will boil within the temperature 

 range of investigation, and therefore that the effect of pressure is insig- 

 nificant and negligible. If this mass is gradually cooled, a temperature 

 will soon be reached when the solution will contain more of one of the 

 ingredients than it can hold in equilibrium. Now, whether it will crys- 

 tallize or not will depend primarily upon two opposing forces, the power 

 of the molecules to arrange themselves in the order characteristic of the 

 particular mineral now in excess, and the power of the viscosity or some 

 other opposing property to prevent this or any other rearrangement, ex- 

 actly as in the case of a single mineral. If the viscosity prevails, the 

 tendency of the excess component to crystallize is not strong enough to 

 overcome the opposing force, crystallization is delayed, or maybe indefi- 

 nitely deferred, and the excess component cools without displaying any 

 evidence of its relation to the rest of the solution, for viscosity increases 

 as the cooling goes on, so that if the excess mineral does not crystallize 

 at an early stage it may find no better opportunity farther down. Thus 

 our whole solution may cool gradually from its initial temperature to the 

 temperature of the laboratory, without anything whatever happening 

 within it to distinguish one component from another or the whole mass 

 from any other mass of random composition. The solution merely per- 

 sists, as a liquid (glass) beyond the temperature where the ingredients 

 should separate, and reveals none of its characteristics. 



