504 B. K. N. Wyllie & A. Scott— Flutonics of Garabal Hill. 



We note (1) that the lime content of davaiiiite is on. a par with 

 that of the wehrlite analyses, while the comparatively low magnesia 

 percentage is consistent with the absence of olivine. (2) The amounts 

 of Fe 0, Ti Oo, Fco O3 in davainite would correspond with 20 per cent 

 of magnetite and ilmenite in the rock, if all were so present. But 

 3 or 4 per cent is all that can be allotted to these minerals. Hence 

 it appears that the hornblende of the rock must contain a large 

 proportion of Fe and Fco O3 and probably Ti Oj. (3) Fe^ O3 is 

 decidedly in excess of Fe in davainite, which is the reverse ol' their 

 normal relation in diallage-bearing rocks. We conclude that our 

 assumption of oxidation is supported by this analysis. 



We have no evidence of any considerable action on the ultrabasic 

 rocks by the diorite magma other than this metamorphism of 

 pyroxene : there has been no wholesale absorption, since quartz- 

 bearing diorites are the rule close up to the contact ; though the 

 absence of olivine from the border-zone of the ultrabasic mass — at 

 least, in all the sections of davainites we have cut — makes it possible 

 that a slight amount of siliceous material has penetrated there. This 

 is supported by the presence in davainite of a considerable amount of 

 rhombic pyroxene ; tlie clustered grains mentioned above will in that 

 case represent original olivine. It is all the more significant that 

 some of the rhombic pyroxene seems to be new-formed and of 

 a different composition from that of the unchanged peridotites. 



It was mentioned that among the rock-types a norite was found. 

 This occurrence will now be described, as it is pertinent to the 

 foregoing discussion. Tlie rock in question forms thin irregular veins 

 in the peridotite at locality iii (see Map, Fig. 1). In thin sections 

 these are seen to be composed of felspar and pyroxene, with granular 

 aplitic texture. Felspar forms about 80 per cent of the bulk except 

 where the femic minerals are clotted together — and that always happens 

 close to the surrounding peridotite. It consists of orthoclase, with 

 a basic plagioclase (andesine or labradorite), the latter being far in 

 excess of the former. In the whiter parts a few grains of quartz can 

 be detected. 



Of the femic mirlerals, the most striking is a richly pleochroic 

 hypersthene, which builds large idiomorphic crystals, and is much in 

 excess of the others, which are colourless or faintly greenish enstatite 

 and diallage. 



Now the surrounding rock is a peridotite, containing diallage and 

 colourless non-pleochroic enstatite, with scattered flakes of mica. 

 Diallage and enstatite in the two rocks are identical, and the vein 

 rock presents in the same way a few stray pieces of biotite. The 

 co-existence in the same melt of two different members of the iron- 

 magnesium metasilicate minerals, which form a continuous series of 

 solid solutions, is obviously impossible. Hence one of them must be 

 xenocrystic, and that one can only be enstatite. The explanation is 

 therefore plain : the vein rock is a hybrid, formed by partial solntion 

 of peridotite matter in a quartz-aplite, sucli as are abundantly 

 associated with the later tonalite magma. Enstatite has remained 

 unattacked. The solution of olivine has produced hypersthene, with 

 elimination of nearly all the free silica of the solvent. Diallage has 



