316 



PHOF. T. Gr. BOSTNEY OK THE 



three grains have once formed portions of a larger one. The mica- 

 flakes also appear often as if they had been twisted out of their 

 original position, and the bands themselves are sometimes rather 

 rumpled. In short, the microscopic structure of the rock recalls 

 that which I have observed in other gneisses and schists which have 

 been exposed to a pressure not sufficient to develop a marked cleavage- 

 foliation. 



(2) The largest mass of the porphyritic granite is near the eastern 

 end of the section examined. The same rock, I believe, forms the 

 numerous skerries and tors east and south-east of the town. This 

 granite cuts both the amphibolite and the gneiss*, being completely 

 welded to, and sometimes including strips of, the latter. As said above, 

 it is coarsely crystalline up to the junction-surface ; one would there- 

 fore suppose that it cooled slowly. Microscopic examination shows 

 it to consist of quartz, felspar, brown mica (partly altered here and 

 there into a green mineral), and a little apatite and iron oxide. 

 Some of the felspar is orthoclase, which is, in places, much altered, 

 being replaced by minute kaolinitic and micaceous minerals ; but 

 plagioclase is more common, showing lamellar and " pericline " 

 twinning. The rock has sometimes a slightly foliated aspect, and 

 under the microscope shows indications of disturbance, though this 

 is less marked than in the gneiss and the amphibolite f. 



(3) The fine-grained granite occurs in irregular dykes, cutting 

 right across the banding of the gneiss, and including fragments of 

 the porphyritic granite (fig. 6). In one case, at its junction with 



Fig. 6,— At Roscoff. 



D 



A. Gneiss, banding very distinct ; lighter 



layers, sometimes full \ inch thick. 



B. Fine-grained granite. 



C. Included fragment 



phyritic granite. 



D. Quartz-felspar vein. 



of dark por- 



* In one case it has lifted up and bent into a curve the top bands of the 

 gneiss. 



t One or two grains show "corrosion quartz," and in others the lamellar 

 twinning occurs in a part of a grain only, near to a slight bend, which seems 

 favourable to Professor Judd's view of this twinning being sometimes due to 

 subsequent mechanical action. 



