ESSAY-REVIEWS 71 



of pressure in areas undergoing tensional faulting. One of these 

 alternatives is the series gabbro-diorite-syenite, etc. ; another 

 is gabbro-essexite-augite-syenite, etc. In them orthoclase is 

 able to crystallise at a moderately early period, owing to the 

 normal delay controlled by water and other volatile fluxes 

 being somewhat weakened. 1 In this way practically the whole 

 field of igneous rocks is brought under survey, and their evolu- 

 tion is traced directly back to a primary basalt magma. While 

 Bowen does not deny the fact of assimilation, he assigns to it 

 only a minor role. Magmatic stoping and abyssal assimila- 

 tion on the scale advocated by Daly are considered now to be 

 unnecessary, since granite — the chief rock type to be ex- 

 plained — may be derived from basalt magma, given a sufficiently 

 large intrusion and a sufficiently long period of cooling. The 

 basic borders of many batholiths represent on this hypothesis a 

 less advancedstage in the differentiation of the magma, that stage 

 having been arrested by the completion of crystallisation around 

 the cooler margins, while later stages — granodiorite or granite — 

 were reached in the hotter and more slowly cooling interior. 



It should be carefully noted that experimental investigation 

 has not yet passed beyond a demonstration of the possible 

 evolution of dioritic and syenitic magmas from basaltic magma. 

 Two difficulties occur to the present writer that deserve dis- 

 cussion before complete acceptance of the further step — the 

 evolution of all granitic rocks — becomes possible. Some basalt 

 magmas could undoubtedly give rise to about 10 per cent, of 

 granite, provided that they had previously differentiated in 

 the opposite direction so as to produce a similar amount of 

 peridotite. The gabbro, however, which would consolidate at 

 intermediate depths, would afterwards be " dead " from the 

 point of view of granite generation. The composition of 

 gabbro as compared with basalt clearly shows this to be the 

 case. The refusion of gabbro might lead to stoping and 

 assimilation, but no appreciable supply of fresh granite could 

 be extracted. The amount of gabbro, or potential basalt magma, 

 in the earth is limited, for the distribution of the radioactive 



1 It is worthy of notice that Dr. J. W. Evans, in a communication to the 

 Twelfth International Geological Congress, Toronto, expressed his opinion that 

 in the presence of sufficient water basalt magma was enabled to produce a 

 differentiate of granitic composition ; whereas, if water were deficient, the 

 basic portion would retain more alkalies, and on further differentiation would yield 

 the different rocks of the alkali-series, 



