440 SEE VOGE 
against the hypersthene (see, for example, Figs. 15 and 16) or the 
diallage, which ordinarily entirely lack idiomorphism. We may 
consequently draw the conclusion that the crystallization of labra- 
dorite in these plagioclase-rich rocks must have commenced before 
the beginning of the solidification of the pyroxene. The idio- 
morphism of the labradorite, however, is often only quite slightly 
developed, as illustrated in Figures 15 and 16. This tells us that 
Fic. 15.—Photomicrographic (19:1) Fic. 16.—Drawing (35:1) 
Hyperitic- (or ophitic-) structured norite from Erteli, Norway. Consists of 
about 56 per cent labradorite (ca. Ab,;An,), 41 per cent hypersthene (0.31 FeSi0,: 
0.69 Mg SiO,), a little magnetite and pyrrhotite (0.07 per cent S, see Fig. 46), 0.09 
per cent apatite; see analysis in Part II. The drawing (35:1) represents the central 
and lower parts of the photograph (19:1). 
only a certain small part of the labradorite, in this case, had solidi- 
fied before the pyroxene began forming. 
Rosenbusch‘ emphasized, and rightly, with regard to the gabbros, 
that in rocks rich in plagioclase, the plagioclase, and in the varieties 
rich in diallage, the diallage, develops in idiomorphic individuals, 
and further, that in rocks rich in diallage, the idiomorphism of 
the diallage is the more prominent the greater its quantity. “Man 
wird also scheinbar gen6tigt ein gewisses Schwanken in der Reihen- 
folge der Ausscheidungen anzunehmen.”’ But Rosenbusch did not 
engage in the physicochemical interpretation of the phenomenon. 
* Mikroskop. Phys. d. Mass. Gest. (4th ed., 1907), II, 1, p. 364. 
ey ce 
