248 DE. H. H. THOMAS OX XENOLITHIC [vol. Ixxviii^ 



be studied in specimens from Locality 16, figured in PL A'lII, 

 figs. 3 & 4. Here spinel-crystals, separated from a glass}" base, haye 

 acted as nuclei upon Avhich haye grown crystals of anorthite, in 

 such a manner as completely to surround the spinel Avith an 

 irregular but thin zone of this mineral. This relationship existing 

 betAyeen spinel and anorthite is one of the most striking features of 

 the xenoliths, and is easily explicable in the light of work recently 

 carried out on the ternary system anorthite-forsterite-silica.i^ 



As a much darker yariety (black, brown, or dark plum-colour),, 

 spinel builds regular well-shaped octahedra in the outer and 

 most highly-modified portions of the xenoliths. There is good 

 reason to belieye, from the microscopic eyidence, that these spinels 

 result from a later interaction of magmatic and xenolithic material, 

 whereb3"the ah-eady-formed anorthite that contained sillimanite has 

 been resorbed by a mao-ma relatiyely rich in mao'nesia ; and that 

 the excess of alumina so proyided has been precipitated as spinel, 

 accompanied by the separation of a sillimanite-free less basic 

 plagioclase (oligoclase) and glass of magmatic nature. Sillimanite- 

 bearing anorthite in all stages of dissolution, and the separation of 

 spinel together with the formation of less basic felspars, are features 

 illustrated by seyeral examples in Plates IX & X. The sharply 

 idiomorphic form of the spinel-crystals is well shown in PL X, 

 fig. 4 and PL VIII, fig. 2. A late conyersion of spinel into yellow 

 or green serpentine is not an uncommon feature of the xenoliths. 

 The serpentine either forms a peripheral skin (PL IX, fig. 2), or 

 completely replaces the crj^stal. 



(iii) Probable Nature of the Unaltered 

 Xenolithic Material. 



The aluminous xenoliths, although complex in mineral contents,, 

 and yarying considerably in composition in different parts of their 

 mass, are clearly fragments of higlily-aluminous shale that haye 

 suffered progressiye metamorphism by a basic igneous magma. 



There is no doubt, from the character and composition of the 

 sillimanite-buchite, that this material has been produced by the 

 simple fusion of aluminous sediment, and that it is unmodified in 

 any way by the transfusion of igneous matter. Its composition is 

 represented by Analysis IV (p. 236), from which it will be seen 

 that it is comparable to a highly aluminous fireclay, and that it is 

 free from all bases such as would be present if contamination by an 

 igneous magma had taken place. Microscopically, it consists of 

 glass in which are embedded, often with clear fluidal arrangement, 

 minute but abundant prisms of sillimanite, an occasional crystal 

 of corundum, and a yery small quantity of magnetite or some other 

 spinellid. The original rutile-needles of the sediment appear, in 

 some instances, to be preseryed. 



The analysis indicates that corundum, separately determined^ 

 constitutes about 0"o per cent, of the whole, and sillimanite 



^ 0. Andersen, 'The System Anorthite-Forsterite-Silica ' Amer. Jour. Sci.. 

 ser. 4, Tol, xxxix (1915) p. 407. 



