250 DE. H. H. THOMAS OjS" xeis^olithic [vol. Ixxviii^ 



(v) The Conditions of Metamorphism. 



It is clear that the metamorphism of the aluminous sediment that 

 gave rise to the xenoliths was effected before the intrusion of the 

 sill. Further, it is evident that the xenohths, as we nov/find them,, 

 owe their main characters to both physical and chemical action on 

 the part of the molten igneous rock, the one producing simple 

 melting without transfusion of material from the magma, the other 

 bringing about a local enrichment of certain bases (lime, ferrous 

 oxide, and magnesia), which resulted in the formation of anorthite, 

 sapphire, and spinel. There is little doubt that the greater part of 

 the metamorphism of the aluminous sediments was actually effected 

 in the walls of the magma -basin, and that the xenoliths are mainl}'^ 

 due to the disruption of the metamorphosed lining of the basin 

 preliminary to the intrusion of the magma into its present higher 

 position. Only in this way is it possible to account for the lack of 

 symmetry of the metamorphic zones in the xenoliths and the trans- 

 gression of the enclosing igneous rock across zones that clearly 

 indicate different phases of metamorphism. A general idea of the 

 process was many years ago clearly outlined b}" Sir Jethro Teall^ 

 who said : — 



' The subterranean magmas act powerfully on tlieir containing- walls, and 

 transform highly argillaceous sediments into crystalline rocks composed of 

 cordierite, sillimanite, biotite, quartz, and sometimes spinelle and corundum. 

 The rocks of the inner contact-zone become shattered, and the igneous magma 

 insinuates itself between the cracks, or may even permeate the mass. Portions 

 of the metamorphic rock float off into the molten material, and travel ^^dth it 

 through dykes and other channels to the surface.' ^ 



In the cases under description the first change effected in the 

 aluminous sediment was simple fusion forming a viscous melt, from 

 which the excess of alumina was early to separate as corundum and 

 sillimanite. The foriner exists as small hexagonal plates, the latter 

 as minute elongated prisms. The rajDid separation of sillimanite 

 rendered the fused layer still more viscous, and presumabl}'' enabled 

 it, without application of external stress, to retain its position as a 

 lining to the magma-basin. Simultaneousl}^, however, the trans- 

 fusion of bases, more particularly of lime, was proceeding from the 

 m.agma into the aluminous melt that formed the matrix of the 

 already separated sillimanite. This transfusion produced a melt of 

 such composition that anorthite was capable of crystallization as a 

 primary phase, and, owing to the partial resorption of sillimanite 

 to furnish the necessarj^- silica, there was a simultaneous precipitation 

 of corundum. 



The anorthite thus formed enclosed the undissolved sillimanite- 

 needles of the original melt (PI. VII). Similarly the transfusion 

 of the bases ferrous iron and magnesia gave rise to spinels of 

 various compositions, which occur occasionally intergrown with 

 corundum, and either completely enclosed in anorthite or in 

 approximate eutectic relationship to this mineral. 



1 ' The Natural History of Cordierite & its Associates ' Proc. Geol. Assoc. 

 ToL xvi (1899) p. 65. 



