THE NATURAL HISTORY OF IGNEOUS ROCKS. 215 



frequently this can be verified, when the edges of the com- 

 ponent dykes are indicated by compact selvages, bands of 

 amygdules, etc. ; and we may suppose that other cases are 

 overlooked in the absence of such decisive evidence. Ap- 

 parent bifurcation is often due to the separation of two 

 dykes or sills which for some distance have run in contact 

 with one another. 



We have assumed the several injections which go to 

 build up a multiple intrusive body to belong to a single 

 rock-type, but mention must also be made of the modifica- 

 tions in form and habit produced by the association of 

 different rock-types in composite multiple dykes and sills. 

 Good illustrations are again afforded by the Tertiary rocks 

 of the British province, which include both acid and basic 

 types. In the larger intrusive bodies these are represented 

 by granite, often with a micographic tendency, and gabbro, 

 graduating into ophitic diabase, etc. It is well known that the 

 two rocks occur together in the Carlingford district, in Ar- 

 ran, Mull and Rum, in the central part of Skye, and as far 

 away as St. Kilda, and always in close association. In the 

 dykes and sills, however, the relations of the two correspond- 

 ing rocks — granophyre and basalt — are sometimes of a more 

 intimate nature. In particular, Sir A. Geikie has drawn 

 attention to a remarkable group of composite sills intruded 

 at various horizons in the Lias of Skye, in which the acid 

 and basic rocks exhibit a curiously symmetrical arrange- 

 ment. Usually there is a central and thicker sheet of 

 granophyre, having in contact with it above and below two 

 thinner sheets of basalt. These composite triple sills are 

 not due to "differentiation in place," but the acid magma 

 has found its way between two basalt sheets which pre- 

 viously constituted a " double sill ". More complex cases 

 are also met with. 



Triple dykes showing a similar association are not in- 

 frequent in Skye, and Geikie noted one consisting of a 

 central dyke of spherulitic granophyre, eight or ten feet 

 wide, flanked by two narrower dykes of basalt. Judd had 

 already remarked similar cases in Arran. One at Cir 

 Mhor, which was specially studied, has a central part of 



