B. K. N. Wyllie & A. Scott — Plutonics of Garahal Hill. 543 



variations, but also the products of final consolidations in the form 

 of schlieren ; the porphyritic pyroxenite veins of Loch Garabal in the 

 ultrabasic series, the hornblendites and the felspathic diorite- 

 pegmatites of loc. v (Map, Fig. 1) in the diorites, and finally the 

 aplites and acid pegmatites associated with the last intrusion. 



For these reasons it seems to us exceedingly difficult to suppose 

 that the whole of these rocks are " the result of the differentiation of 

 an originally homogeneous magma ".^ Firstly, it is most improbable 

 that a magma is ever in a state approaching homogeneity. Local 

 variations of physical conditions as well as assimilation phenomena 

 on the margins must be sufficient to keep the equilibrium hetero- 

 geneous. Secondly, while it is undoubtedly true that the basic 

 elements would crystallize out first, the crystals thus formed would 

 tend to sink, so that the upper parts of the magma would be relatively 

 enriched in silica. Thirdly, the absence of intermediate types 

 militates against the differentiation theory. This is also an argument 

 against any theory of assimilation, as we should then expect 

 a perfectly continuous sequence. Again, it is obvious that an 

 original ultrabasic magma can never assimilate enough schistose 

 material to form a toualite, as the schist is more basic than the 

 tonalite, while on the other hand an original acid magma could 

 never form peridotite. In addition, it seems to us that difficulties 

 arise when we come to consider the mechanism of differentiation. 

 The application of Soret's principle ^ involves a large degree of 

 diffusion and ignores, to some extent, the important role played by 

 viscosity, so that its applicability is probably confined to marginal 

 phenomena on a fairly small scale. Similar objections exist with 

 regard to the principle of Gouy and Chaperon.^ The difference of 

 concentration between the top and bottom of a magma due to 

 a gravity concentration of the heavy molecules in the dissolved state 

 would be so small as to be negligible, unless the magma were of 

 enormous depth.* It is most probable that gravity only operates on 

 the heavy molecules where the latter form an immiscible liquid or 

 have crystallized out. Hence it seems to us that a separate origin of 

 the ultrabasic and the acid magmas must be postulated. While it is 

 possible that the diorite has formed from the tonalite by assimilation, 

 no such possibility exists with respect to the peridotites. It may 

 be that ttie igneous material of the earth's crust is stratified to 

 some extent and thtit the pyroxenites, etc., originated in some 

 ultrabasic infraplutonic zone similar to that postulated by Fermor.^ 

 The almost invariable association of acid rocks with the earlier 

 peridotites suggests that the former were forced up the margins of the 

 earlier intrusions, which would constitute areas of crustal weakness 

 and hence be the most ready outlet for the acid material. Any 

 attempts to verify these ideas, however, are so much hampered by our 



^ Teall & Dakyns, loc. cit., p. 117. 



^ Ann. de Cliim. et de Phys., ser. v, xxii, pp. 293 et seq., 1881. 

 ^ Ibid., ser. VI, xii, pp. 384 et seq., 1887. 

 * Cf. Lelimann, Molekularphysik, i, p. 488, etc., 1884. 

 ^ " Garnet as a Geological Barometer " : Eec. Geol. Surv. India, xliii, pt. i, 

 pp. 44-5, 1913. 



