Plagioclase Feldspars. 595 



The Question of Iviister's Rule. 



Both the equations derived and the decisive nature of the 

 experimental results demonstrate that Krister's rule* can apply 

 only in a certain limiting case. This rule states that the melt- 

 ing points of isomorphous mixtures should lie on the straight 

 line joining the melting points of the pure components. It 

 can be seen from inspection of the equations for the liquidus 

 and solidus that the two curves may lie close together if either 

 Z x or Z 2 is very small, and that both may approach straight 

 lines if both Z, and Z 2 are small, but that they can not 

 become a single straight line unless T x — T n , i. e., the difference 

 between the melting points of the components, is zero.f That 

 vapor pressures, densities and like properties, when measured 

 at a definite temperature, should be linear with respect to com- 

 position for solid solutions which are perfect physical solu- 

 tions, is a necessity. That such a rule should apply to melting- 

 points, themselves variable temperatures, does not follow. 



Isomorphism. 



The establishment of complete solid solution between the 

 feldspars raises the whole question of the use of the terms 

 "solid solution'' and "isomorphism." The latter has been 

 avoided hitherto m this paper on account of the variety of 

 meanings which have at different times been attached to it. 

 Some authors use " isomorphism " to designate complete solid 

 solution, others speak freely of " limited isomorphism," and 

 still others use the term in its original significance of simple 

 crystallographic similarity, without regard to the question of 

 miscibility in the solid state. In view of the recent very com- 

 plete review of the subject by Hlawatsch,^: nothing more need 

 be said of this matter except that, for the conception of misci- 

 bility in the solid state, the term, solid solution, seems to be 

 much more useful, both in avoiding ambiguity, and because of 

 a certain connotation of similarity with liquid solutions which 

 has been amply justified in the present instance. 



It has hitherto proved impossible to agree upon a deciding 

 factor which shall determine w T hether or not two substances 

 shall form a complete series of solid solutions. The facts indi- 

 cate rather that no single criterion may be safely used alone. 

 Similarity of chemical composition, for example, is known to 

 be favorable. Except that both are alumino-silicates, no great 

 similarity of composition between albite and anorthite, as their 

 formulge are ordinarily written, can be claimed. For this 



*F. W. Kiister, Zs. phys. Chem., viii, p. 577, 1891. 



f Cf. Van Laar, loc. cit., p. 439, and Day and Allen, loc. cit., p. 68. 



JZs. Kryst., li, p. 417, 1912. 



