286 HAROLD L. ALLING 



■ The possibility of inversion^ of orthoclase to microcline is not pro- 

 vided for in Part I, although Warren's figure introduced it. This 

 change in modification he emphasized as a partial explanation of the 

 formation of perthitic intergrowths.^ The other suggested modes 

 of origin are the " unmixing "^ or "ex-solution"'* of the two phases 

 upon falling temperature and the crystallization of the eutectic. 

 It is well to note that the diagram relates to pegmatitic feldspars 

 only and does not apply to feldspars of other origins. 



One of the difficulties with the Vogt-Warren diagram is that it 

 was based upon chemical and quantitative microscopic studies^ 

 of pegmatitic feldspars at normal pressures and temperatures, and 

 was not checked by thermal analysis. Had these data been avail- 

 able, it is believed that a different position of the eutectic point 

 would have been proposed, as well as certain other conclusions 

 drawn. The reason for this is that the proportion of the two feld- 

 spar phases in a given intergrowth depends upon the amount of 

 exsolution that has taken place. Rapid cooHng and the absence 

 of mineralizers promote excessive undercooling and thereby retard 

 exsolution. This condition prevails in many surface and porphy- 

 ritic rocks, and hence we should not draw too general conclusions 

 from special or local phenomena. 



THE DIAGRAM OF DITTLER 



Dittler^ investigated the potash-soda series, and found that 

 natural specimens of various composition melt between 1125° and 

 1200° C. either incongruently'' or congruently, and came to the con- 

 clusion that the form of the diagram should be a series of soHd solu- 

 tions with a minimum.^ (Fig. i. Part II.) 



^ See J. B. Ferguson, Science, N.S., L (1919), 544-46. 



= J. H. L. Vogt, Tsch. Min. Petro. Mitt. (2), XXIV, 537-41 ; Alfred Harker, Natural 

 History of Igneous Rocks, 1909, pp. 259-260. 



3 C. H. Warren, Proc. Amer. Acad. Arts and Sci., LI (1915), No. 3. 



4 The writer. See Part I, p. 222, footnote. 



5 The Delesse-Rosiwal method. See Arthur Hohnes, Petrographic Methods, p. 313. 

 *E. Dittler, "Die Schmelzpunktskurve von Kalinatronfeldspaten," Tsch. Min. 



Petro. Mitt., XXXI (1912), 513. 



7 See Morey and Bowen, Amer. Jour. Sci. (5), IV (1922), 1-22. 



8 Type III of Bachius Roozeboom. 



