248 G. C. Martin — Dunite in Western Massachusetts. 



extremely contorted. There is a general tendency toward a 

 north and south strike with a steep easterly dip, hut this varies 

 extremely, as may be seen by the map. The dunite mass in its 

 southern part is parallel to the strike of the gneiss, but at the 

 northern end the bedding of the gneiss changes position, possi- 

 bly forming a synclinal fold whose axis is cut by the dunite. 



The contact of serpentine or dunite with gneiss is nowhere 

 revealed. Outcrops of the two rocks are everywhere separated 

 by eighty feet or more of swamp or by a drift-filled stream bed. 

 In these places abound fragments of rotten gneiss and vesicular 

 bowlders of gneiss, which latter are seldom elsewhere seen. 



Structural Relations. — Concerning the structural relations 

 of the dunite to the surrounding rocks, there are five hypoth- 

 eses which may be considered. 



1. That the dunite is a chemical deposit. 



2. That it is a clastic. 



3. That it is a bedded flow. 



4. That it is a knob older than the gneisses and about which 

 they were deposited. 



5. That it is an intruded mass. 



Of these the last seems best to accord with the following facts. 



The unaltered dunite is uniform in character, has no trace 

 of fragmental material, of recrystallization, or of bedding or 

 banding of any kind. 



There are an abundance of angular chloritic masses near the 

 border, which on account of their aluminous character are 

 probably the remains of feldspathic inclusions. 



There are a number of outlying masses of dunite which can 

 be connected with the main mass only as apophyses. 



At several places the strike of the gneiss is directly toward 

 the mass and in close proximity to it. (See map ; east side, 

 Loc. 346, 114a; west side, Loc. 353). These show conclusively 

 that the dunite is discordant with the bedding of the gneiss. 



The contact is deeply disintegrated and is marked by the 

 presence of vesicular gneissoid bowlders, suggesting an altera- 

 tion by an intruded mass. 



Note. — In this connection it may be noted that there are in 

 the Hoosac Range and to the east of it a considerable number 

 of serpentine beds, many of which are not known to have 

 been studied since the Hitchcock Survey and whose origin has 

 not been made out. Some of them, however, are known to be 

 derived from peridotites,* but these peridotites resemble all 

 others in America, except those from Quebec, Xorth Carolina, 

 and the one here described, in containing, as original constitu- 

 ents, other ferro-magnesium silicates in addition to olivine. 



Cornell University, Ithaca, X. Y. 



* Bulletin 126, U. S. Geol. Survey, pp. 10, 56 : 152. 



