178 [ASSEMBLT 



and it is traced into New Fairfield and Danbury, Connecticut. Dr. 

 Percival did not recognize in it the same general characters as in his 

 " Granitic Formations," but put it under the head of " Calciferous 

 Micaceous Formation (gneiss)" and represented it by H, 2. B. on his 

 geological map. He calls it a table-land bounded north and south 

 as indicated above. In his description of the rocks of this division 

 or formation a granitic gneiss is mentioned as predominating witli 

 beds of dark liornhlendic and suh-hornUendic rocks and dark micaceous 

 rocks. And he adds that a range of the latter is traceable west 

 beyond the Oroton, between Carmel and Somers.* An examination 

 of the outcrops on two sections across this division of Dr. Percival, 

 one from Brewster's, north-east to Valleyville and Quaker brook hol- 

 low, and the other from Brewsters through Southeast Centre and 

 Milltown and across Joe's hill, failed to show sufficiently marked and 

 distinctive rock characters to separate it from the gneisses, granulytes, 

 syenite-gneisses and micaceous schists common to the Highlands to 

 the west. 



The southern limits of the Ai-cha3an rocks of the Highlands from 

 the State line east of Brewsters to the Hudson river may be described 

 as follows : Beginning near Mill Plain, Conn., and at the southern 

 foot of the Joe's hill ridge, the line follows the east and west valley 

 to near Southeast Centre and then the Croton river, passing south-east 

 and south of Brewsters. This valley has in it a white limestone and 

 a micaceous quartzite, as also mica schist, and these sedimentary 

 rocks have a general east and west strike and a northward dip. Joe's 

 hill is the southernmost ridge of gneiss of the belt, whose boundaries 

 are here given, and the limestone and schistose rocks are the adjacent 

 bounding formations. South-west of Brewsters the tracing of the 

 geological line is rendered difficult by the absence of distinguishing 

 characteristics in the rocks which occur in the undoubted Archaean 

 belt north-west of the Croton river, and some of the outcrops of 

 gneiss which are traceable in a narrow belt eastward from north of 

 Croton falls to Peach lake and into Eidgefield and Danbury in Con- 

 necticut. The schists in the southern part of the town of Southeast 

 and north of Peach lake border this belt of gneiss on the north and 

 the Salem limestone is contiguous on the south. The prevailing 

 north-east dip makes an angle with the mean direction of that of the 

 beds in the outcrops of the Theall mine and along the Croton valley, 

 but no contact phenomena or localities of unconformable strata were 

 observed. So far as rock composition is concerned there is more mica 

 and a more schistose structure in the rocks of this Croton falls and 

 Peach lake gneissic tongue, if it be viewed as an extension eastward 

 of the Archaean. Garnet also is common in these rocks; and they 

 have a striped or banded appeai'ance. But there is the absence of 

 well-marked distinctions which are at once recognized in the outcrops 

 on the two sides of the boundary line as traced across Putnam and in 

 Dutchess col^nties. 



From the Croton river west the general course of the southern line 

 of the Highlands belt is westerly, but it is marked by several loops, 

 which stretch southward around the Archfean projecting ridges, as it 



* Fercival'e Ji(^)ort on the Geoloijij of Connecticut, 1842, pp. 83 and 92, 93. 



