CRYSTALLIZATION IN IGNEOUS ROCKS 459 
that augite continued to crystallize after plagioclase had ceased. 
The order of cessation of crystallization is the only certain informa- 
tion given. 
In Fig. 3 but one crystal is shown idiomorphic against the 
augite crystal. If a great many other crystals showed the same 
relation to the augite, in short, if the augite crystal were inter- 
stitial to a number of other minerals, no more definite conclusion 
S 
Ss 
(prismatic plagioclase) (tabular plagioclase) 
Fic: 3 
than the above could be drawn. The augite crystal continued 
to grow after the others ceased, but it might have begun to form 
before the crystals to which it is interstitial. 
Obviously, then, the observation of allt the relations of outline 
of a number of minerals in a section of a holocrystalline granular 
rock leads to a safe conclusion.only as to the order in which the 
minerals have ceased to crystallize. Any statement of order of 
crystallization based on such observations should be modified to a 
statement merely of order of cessation of crystallization. 
1Cf. A. Harker, The Natural History of Igneous Rocks, 1909, p. 179. 
