Metamorphic Upper Devonian Rooks. 367 



lite, and with quite small garnets, becoming, above, coarser, of 

 rougher surface, and knotted with large staurolites. 



At the south end, nearest the Williams farm, the basal 

 quartzite dips beneath a very iine-grained flat fissile mica slate, 

 which dips 20° in the direction S. 10° E., its surface sparsely 

 pimpled with small garnets, and without other accessories and 

 closely like the underlying schist of the Williams farm section. 



The lowest bed of hornblende rock, here thin and poorly 

 exposed, is followed by a second band of mica slate exactly 

 like the first, which passes with the same dip and strike beneath 

 a massive, dark gray to greenish-black hornblende rock, greatly 

 jointed, and this is exposed in a broad area nearly down to the 

 main road running east from Bernardston, and extending east 

 to the house of S. J. Green, 100 rods west of the locality men- 

 tioned by Professor Dana.* It contains a central band of 

 dark limestone at times 30 cm thick. The hornblende rock is 

 capped by a thin layer, never more than a meter thick, of a 

 shining, white arenaceous mica schist, with scattered scales 

 of biotite, and a similar layer was found to cap a similar horn- 

 blende rock, in so great a number of instances between this 

 point and South Vernon that it attracted particular attention. 

 It was found to pass in every case up into the common dark 

 gray mica schist, and to differ from it only in the entire absence 

 of coaly matter and magnetite ; and it seems possible that the 

 former may have been discharged by the ferruginous matter of 

 the hornblendic band adjacent. It is, however, wanting below 

 the hornblende bands which rest directly on the dark gray and 

 finer mica schist. This makes it probable that none of the 

 hornblende bands are overturned. 



The schists of the area just described are cut off, going east- 

 ward, by a great drumlin, though the quartzite can be followed 

 by the north end of it ; beyond one finds sections which expose 

 the whole thickness of the schists and hornblende bands. 



They are best studied in the area east of the Purple blind 

 road— see map — where, commencing in the chestnut woods 

 northeast of the end of the road, at the basal conglomerate, we 

 pass south over a broad area of the lowest mica schist, broad 

 because of the low dip, and come upon the lowest hornblende 

 rock, a band — about four meters thick — here, as always quite 

 ferruginous and pyritous. Fifteen meters beyond there is a 

 second bed of hornblende rock like the first, and both are 

 capped by the white mica schist layer described above. Going 

 on twenty meters to the top of the ridge, at a large chestnut 

 tree conspicuous in the open field, there is a third rudely 

 laminated layer of hornblende rock, thicker than the others 

 and distinctly laminated. This is capped by a bed one meter 



* This Journal, III, vi, 342, 1873. 



