220 N. L. Bowen—The Ternary System: 



General Character of the Boundary Curves. 



The composition diagram, with isotherms, is -a scheme for 

 representing both composition and temperature on a plane sur- 

 face (in plan). It is a projection, on the plane of the base, of 

 the solid figure showing the fusion surfaces. Now a vertical 

 section of the solid figure shows the temperature relations 

 somewdiat more clearly than do the isotherms, just as a vertical 

 section along a given direction in a mapped area shows the 

 relief more strikingly than does a contoured map. Fig. 7 has 

 therefore been drawn to show the temperatures along the 

 boundary curves. It represents vertical sections of the solid 

 figure along the boundary curves. Composition is, of course, 

 not indicated and is known only by referring the curves to 

 their position on the composition diagram. 



The figure shows that the boundary curve, forsterite- 

 pyroxene, falls continuously, becomes gradually flatter as it 

 approaches the point where it cuts the pyroxene conjugation 

 line and thenceforth is nearly flat. 



The boundary curve pyroxene-silica is, strictly speaking, 

 composed of two boundary curves, pyroxene-cristobalite and 

 pyroxene-tridymite. The difficulty of getting silica to assume 

 the equilibrium form in the neighborhood of the inversion 

 point tridymite-cristobalite (1470°) is very great and in quench- 

 ings near this temperature both forms were usually found. 

 Any sudden change of direction which this boundary curve 

 may have at the inversion temperature is, therefore, not dis- 

 coverable experimentally. The two curves may, for con- 

 venience, be considered as the one boundary curve pyroxene- 

 silica. 



The pyroxene-silica boundary falls continuously from the 

 eutectic, clino-enstatite-silica to the eutectic diopside-silica. It 

 becomes considerably flattened at the lower temperatures, but 

 there is clearly a continuous fall, without a minimum. 



Such boundary curves can be interpreted only in terms of 

 complete solid solution between diopside and clino-enstatite, a 

 fact which is confirmed by other evidence presented later. 



Brief Theoretical Discussion of Solid Solution in Ternary Systems. 



The relations in the present system are somewhat complex 

 and, so far as the writer is aware, a system showing the same 

 general relations has not been investigated nor has such a 

 system received theoretical treatment. Before proceeding 

 further with the discussion of the equilibrium diagram it 

 would, therefore, be well to discuss some theoretical cases of 

 solid solution in ternary systems which have been treated by 



