THE PROBLEM OF THE ANORTHOSITES 95 



but let us, at the same time, study the crystaUization of melted 

 mixtures of natural minerals and of fused igneous rocks. 



Bowen gives two points as special evidence for gravitational 

 differentiation: (i) those intrustive sheets whose lower part consists 

 of gabbro (diabase) and the upper of granite (granophyre) ; (2) the 

 connection of nephelite syenites with granites (he admits that the 

 former overlie the latter). 



1. Examples of the first case are well known. The fact itself 

 is beyond doubt, but the question lies in its interpretation. Every 

 author of a general theory of differentiation makes use of these 

 facts to support his own views, and it is therefore easy to see how sub- 

 jective the interpretations become. A very instructive example is 

 afforded by the sills of the Purcell Mountains. While for Bowen 

 they are a confirmation of his view that the granite did not come 

 from a distinct magma but was derived from the basaltic magma 

 which is the one and only mother-magma of all igneous rocks, for 

 Daly the same intrusive bodies illustrate his eclectic hypothesis, of 

 a syntexis of the basaltic magma with quartzite, followed by differ- 

 entiation. Bowen asserts that the granites are everywhere under- 

 lain by granodiorites, diorites, and gabbros, and that these rocks 

 can be seen wherever the granites are cut through by denudation. 

 He says further that the thickness of the granite layer does not 

 exceed from 10 per cent to 15 per cent of the whole intrusive body. 

 Where are the proofs for such assertions, and how can they be 

 brought into agreement with the fact that the dimensions of 

 granite bodies exceed by far those of other intrusive rocks ? 



2. The frequent, or perhaps even constant (?), association of 

 nephehte syenites with granites has been mentioned by Daly, and 

 this is really a very frequent combination, but there are no proofs 

 that nephelite syenites always overlie granites. 



As already mentioned, the theory of the origin of the anortho- 

 sites (and other monomineral igneous rocks) by the accumulation of 

 soKd crystals, and the denial of their existence in the Hquid state, is 

 founded on the following geological and petrographical data: 

 (i) protoclastic structure, (2) lack of dikes, (3) lack of effusive 

 equivalents, (4) absence of mineraHzers, (5) the forms of the rock 

 bodies and their stratigraphical relations. 



