C— GEOLOGY. 55 



individuals had practically ceased to grow. It represents the original 

 basalt magma almost depleted as regards lime and magnesia, but retaining 

 abundant alkalies and dissolved water. It evidently retained its fluidity 

 over a fair range of temperature and would thus be capable of separation 

 under gravity or any externally applied stress. This process of the 

 migration of the acid differentiate can be studied in the Carboniferous 

 quartz-dolerites of the Lothians, and we assume that it has taken place 

 under the action of gravity in the quartz-gabbro ring-dyke of Glen More 

 in Mull. 



It would appear, therefore, that stresses produced by subsidence acting 

 upon a partially differentiated magma of this character could bring about 

 the separation of the fluid acid differentiate and cause intrusions of 

 granophyre or allied rocks. 



The separation of an acid differentiate that exists in the interstices of 

 an almost solid magma, by external stress, has repeatedly been used by 

 Dr. Bowen to account for the close association of widely divergent rock- 

 types. He considers that an external pressure applied to a magma that 

 had become about 80 per cent, solid by the formation of a mesh-work of 

 crystals could break down this mesh-structure. The interstitial liquid 

 would then feel the pressure which would be transmitted hydrostatically 

 to all parts, and if continued would produce a separation and intrusion of 

 the acid differentiate in a comparatively pure state. Such action is 

 supported by the frequently bent and broken state of felspars and early 

 formed augites of the quartz-dolerites. 



Although the magmatic sequence, in Ardnamurchan, is from basic to 

 acid, like that of other centres, the ring-dykes almost invariably have 

 margins of varying width which are more acid than the bulk of the rock 

 forming the particular intrusions. In these less basic margins we 

 certainly see, not the result of differentiation in place, but the effects of 

 the injection and partial chilling of a magma that was followed immediately 

 by one of somewhat more basic composition. Such a condition of things 

 might conceivably be brought about by the subsidence of a crustal mass 

 into a basin filled with partially differentiated magma. The first portion 

 to be intruded would be that richer in the acid differentiates, while as 

 subsidence continued there would be an uprise of the more basic and less 

 differentiated magma that represented the bulk of that contained in the 

 upper part of the reservoir. Such a process appears to me to be the 

 explanation of the quartz-dolerite margins to the Ardnamurchan gabbros 

 and eucrites, and the felsitic margin to the tonalitic ring-dyke of the 

 central complex. 



The alkaline rocks call for passing notice on account of the varied and 

 sometimes fanciful suggestions put forward to explain their origin. In the 

 Tertiary province they form a very insignificant group so far as bulk is 

 concerned, but are important on account of their unusual composition. 

 In the plateau-lavas certain segregation veins that represent a late phase 

 in the consolidation are of definitely alkaline character. They prove that 

 the differentiation of the plateau basalt magma, even under superficial 

 pressure and rapidly falling temperature, is capable of producing rocks of 

 alkaline character. Dr. Bowen on theoretical grounds concludes that th< j 

 alkali-rocks can, and do, originate from the same primitive magma as the 



