THE POSTERIOR SHEET. 473 



steam and water might occm-, and an indraft of muddy water, wliicli W(.uld 

 .siuldeuly coat the surface Avith fine, tliin-laminated uTud, and this would 

 then be mingled with the plastic la\a so as to produce veins like those seen 

 in Castile soap. 



The current was liere passing directly into the apex of the area of the 

 Chicopee shale, and the fragments in the trap are of exactly the same char- 

 acter as this rock. This mixed layer appears in its normal j)Ositiftn on the 

 surface of the trap at the quarry, and the heavy-hedded sandstones aljove 

 show that tlie)^ were very rapidly accumulated over the still heated trap 

 by the abundance of specular iron that coats all their fissures. Here in 

 the bluff the whole former surface of the trap is under- rolled and appears 

 imerted upon the sandstone of the old sea bottom. 



THE ROARING BROOK FAULT AND THE DISAPPEARANCE OF THE POSTERIOR 



SHEET. 



The amygdah^idal surface of the trap, so well exposed at Delaney's 

 quarry, can be followed continuously a few rods north in the bed of Roar- 

 ing Brook. It is here more coarsely araygdaloidal and without inclusions 

 and is covered by a thin bed of tuffaceous sandstone full of flat fragments 

 of a white volcanic rock (see p. 474). Following up the brook across the 

 road to a small waterfall on this amygdaloidal surface, with the sandstone 

 forming the bank, one comes upon the first fault. At the foot of the fall 

 is the scoriaceous surface. The trap over which the Avater plnng'es is com- 

 pact, and here there is a fault with upthrow on the west which amounts 

 nearly to the thickness of the trap sheet. A little west there is another 

 fault in the same direction and throAv, running about N. 20° E., jiarallel 

 w'itli and a little west of the brook, which seems to have given the lirook 

 its direction, and which continues north along the east foot of the eastern 

 bluff of the hill 30 rods north of the brook that contains the inverted 

 section described above (p. 472) and crosses the next brook on the noi-th 

 10 rods west of the road and of the broad surface covered with bird-tracks 

 beside the road. It has the sandstone on the east and the tuff and trap on 

 the west. 



As the surface of the trap sheet is exposed in the bed of Roaring Brook, 

 and as the base of the same sheet appears 30 feet higher in the hill 30 rods 

 north, the aggregate throw of the faults is al)Out 60 feet. 



