198 DE. C. E. TILLET ON THE PETROLOGY OF [vol. lxxix, 



In the third type of rock, the rocks of composite origin, the 

 minerals developed include those characteristic of the Green Schists, 

 and the essential minerals of the mica-schists. The lowest grade 

 of metamorphism is represented by the quartz-muscovite-chlorite- 

 albite types. The succeeding stage is the development of red 

 garnet which (as previously noted) may reach 1/16 inch in 

 diameter, forming idiohlasts in a museovite-chlorite ground-mass. 

 This garnet is an almandine containing 5 "5 per cent, of manganous 

 oxide, corresponding to a spessartine content of 12"S per cent. 



There is no indication that biotite has ever been a constituent 

 of these rocks, nor that chlorite has been secondarily derived from 

 it. Almandine is the common garnet of metamorphosed argilla- 

 ceous sediments, and as such appears in rocks showing progressive 

 metamorphism after biotite has developed. This is abundantly 

 clear from the work of Mr. G. Barrow in the South-Eastern 

 Highlands of Scotland, 1 and receives confirmation from the zones 

 of metamorphism in the Trondhjem district of Southern Norway. 3 



The presence of a notable content of manganese in the metamor- 

 phosed sediment appears, however, to lead to an earlier development 

 of garnet, with the result that the order of development of these 

 two minerals is reversed. This receives support from other regions, 

 for example in the spessartine-bearing phyllites of the Ardennes 

 (as described bjr Renard), and still more recently in the Stavanger 

 district of Southern Norway. In this area Prof. V. M. Gold- 

 schmidt has shown that progressive metamorphism in the Cambro- 

 Silurian sediments has led to the development of zones characterized 

 in ascending sequence by chlorite, garnet, and biotite. Analysis 

 of the garnets of these rocks shows that they contain notable 

 amounts of the spessartine molecule. 



The highest stage of metamorphism in the composite rocks is 

 that of the garnet-bearing types. The source of the manganous 

 oxide of these garnets is doubtless the sedimentary material, and 

 not the part contributed by the igneous rocks. The garnet owes 

 its development to chlorite combined with manganiferous material 

 in the sediment. This is the more probable when it is considei*ed 

 that garnet never appears in the true Green Schists rich in chlorite, 

 but devoid of sedimentary detrital material. 



Hornblende is a rare constituent of this type of rock, and when 

 it does appear, is associated with chlorite usually free from 

 muscovite. Its rarity here, when compared with its almost 

 universal presence in the true Green Schists, is noteworthy. It is 

 highly probable that in rocks of this composition, hornblende 

 appears characteristically in a higher grade of metamorphism, and 

 later than in the associated Green Schists. 



Among the minerals of igneous rocks, hornblende and muscovite 

 appear to be mutually exclusive, and, while we are unable to apply 

 completely the same restriction in metamorphosed rocks, there are 



1 Proc. Geol. Assoc, vol. xxiii (1912) pp. 274-79. 



2 V. M. Goldschmidt, Vidensk. Selsk. Skrifter, No. 10 (1915) pp. 36-37. 



