226 REPORTS ON INVESTIGATIONS AND PROJECTS. 



of stable existence in any proportion of the two component minerals. 

 The proof of isomorphism depends upon two facts which we were 

 able to establish experimentally : First, a continuous change in the 

 temperature of melting corresponding point for point with the change 

 in percentage of the two components ; second, upon the continuous 

 change in the specific gravity with the chemical composition in the 

 same way. We were able to prepare artificially all the feldspars 

 which were investigated, in high chemical purity, according to the 

 strictest standards of quantitative chemistry. They were also studied 

 optically and identified with established natural types. The chem- 

 ical purity of the specimens is responsible for a large measure of the 

 success of the investigation, for thermal measurements, even when 

 carefully made, have but little significance if disturbing subordinate 

 reactions are present. For the same reason good specific- gravity 

 determinations are possible only in chemically pure and homogeneous 

 material. 



As the percentage of the more alkaline feldspar in these mixtures 

 increases, the viscosity of the molten mineral comes to be of the same 

 order of magnitude as the rigidity of the solid crystals. This gives 

 rise to an entirely new phenomenon, both to physicists and mineral- 

 ogists. It becomes possible to superheat the crystals 150 or more 

 above the point where melting begins, on account of the mechanical 

 resistance to molecular deorientation which is offered by the hyper- 

 viscous liquid. 



The investigation contributed not only a number of positive con- 

 clusions on the subject of rock formation, but also a valuable verifi- 

 cation of the theory of isomorphism which had been developed by 

 Professor Roozeboom, but which had never been put to a fair test 

 over a long range of temperature. 



Incidental to the investigation, it was found that considerable 

 viscosity in liquid minerals means considerable undercooling before 

 solidification occurs, unless some powerful outside agent can be 

 brought in to restore the equilibrium — a fact which makes it danger- 

 ous to draw far-reaching conclusions from observations of solidifying 

 temperatures, but which clearly shows the possibility of feldspar 

 formation far below the melting temperatures of these minerals. A 

 liquid feldspar magma of any composition, if cooled rapidly, usually 

 does not solidify at any particular point, and may become mechan- 

 ically solid without crystallization at all — that is, it may form a glass. 

 Some study of glass formation is almost inseparable from any thermal 

 investigation of rock-forming minerals. 



