C— GEOLOGY. 57 



high percentages of alumina, lime and magnesia when compared with the 

 plateau basalt and rocks of the other series with similar percentages of 

 silica. Thus, for instance, the normal quartz-dolerite magma, represented 

 in Mull and Ardnamurchan by innumerable sheet-like intrusions, and by 

 the great mass of Ben Hiant, has a similar silica percentage to the magma 

 represented by the quartz-gabbros of these regions. The quartz-gabbro 

 magma, however, is richer in lime and alumina, a fact that indicates a 

 concentration of lime-felspar greater than that encountered in the plateau- 

 basalt and its normal derivatives. 



One other point. There are still petrologists who see in the variation 

 of igneous rocks the results of wholesale contamination of a magma by 

 the assimilation of country rock. But, whatever the evidence elsewhere, 

 in the Tertiary province the idea of serious modification of a magma by 

 such means receives no support. 



The chemical and mineralogical characters of all the major intrusive 

 bodies are quite normal and are just such as would be presented by 

 straightforward differentiates. A study of the margins of intrusive masses 

 shows that the effects of interaction and assimilation are quite local, of 

 relatively insignificant extent, and where encountered are distinguished by 

 characters that are unrepresented in normal igneous rocks. Even where 

 interaction between a magma and its retaining walls can be presumed, as 

 in the case of the magma that supplied the xenolithic sills of south-west 

 Mull, the extent of interaction was very limited. It would appear that a 

 narrow reaction zone was established and that the early precipitation of 

 insoluble phases in this zone protected the country rock from further 

 attack. The magma as a whole remained unmodified, and the reaction 

 zone presented quite abnormal chemical and mineralogical characteristics. 



If assimilation was going to be operative on a large scale the place to 

 look for it would be in the deep-seated basin of the primary magma, where 

 elevated temperatures and the unsaturated state of the magma as regards 

 silica would be in its favour. The greatest argument, therefore, against 

 serious magmatic modification by assimilation is furnished by the repeated 

 appearance of the plateau basalt magma throughout the Tertiary province 

 in an unmodified form at widely separated periods — first as the plateau- 

 lavas, then as cone-sheets and sills, and lastly as basic-dykes. 



