ORIGIN OF RAND UANKETS. * 359 



magma, there ought to be a heterogeneous collection of products 

 according to the varying nature of the sediments absorbed, 

 whereas there is a fairly constant tyj)e of rock exhibited in the 

 selvage. The answer to this is to be found in the selective action 

 of the crystallising forces, these having the power to control the 

 material out of which the crystals are being formed, retaining 

 those substances which heh) to build up the type crystals and 

 rejecting unsuitable substances. The action can be studied in 

 the andalusite and cordierite rocks that have been produced as a 

 result of contact metamorphism round the laccolite, or, in fact, in 

 any contact zone. I first observed the facts in the andalusite 

 schist of George* The first indication of metamorphic action 

 is the production of a f els or hornstone ; the andalusite substance 

 is deposited as a cement in patches throughout the sedimentary 

 rock, and the minerals and fragments in the sediment are as yet 

 unaltered, (iradually. as the action of metamorphism is ])ro- 

 longed, the andalusite dissolves the edges of the included grains, 

 assimilates substances suitable to its growth, and passes out the 

 unsuitable materia!, at the same time gradually widening its 

 cavity and lining up its crystalline boundaries. The included 

 fragments and grains of chlorite, mica,, quartz, etc., have now 

 rounded edges, and are much more sparsely scattered throughout 

 the crystal. Some substances in the sediment cannot be assimi- 

 lated, and in many cases these are arranged by the crystallising 

 forces in certain axial lines which form the dark crosses in the 

 variety known as chiastolite ; lumps of material that in the be- 

 ginning were not penetrated by the andalusite substance become 

 whittled away to little specks, and remain like cysts in the crystal. 

 The same laws that apply to the individual crystal apply to 

 the collection of crystals forming the rocks, and hence the rock 

 type grows by the absorption of sedimentary rocks, but does not 

 lose its individuality in the process ; the little cysts of undigested 

 material in the indalusite crystals of George find their parallel 

 in the xenoliths, t)r masses of sedimentary rocks included in the 

 igneous rocks. Marker [ has adduced as an argument against 

 the theory that igneous rocks " stope " out the chambers that 

 they occupy and assimilate the blocks that fall into the liquid 

 mass, the fact that in Skye masses of Cambrian limestone rest as 

 included bodies :n the granite and gabbro indiscriminately. The 

 cysts in the andalusite may ver^' well account for these, as the 

 xenoliths of limestone would represent kernels which escaped 

 absorption ; in the case of the granite the viscosity of the liquid 

 enabled the limestone to withstand absorption, and in the case 

 of the gabbro, the rock was sufficiently rich in lime to reject 

 further additions of this material. 



*_"T]ic Anclalusite schist of George," Records Albanv Museum, 

 vol. ii., pt. ii., 1907. 



t " TcM'tiary Toneous Rocks of Sk\c "' (1004). ]). 97. 



