76 GEOLOGY OF THE SECOND DISTRICT. 
part of Remsen ; then northwesterly, skirting the western border of the valley of the Black 
river, through Lyonsdale to Carthage. From Carthage, the rock runs northeast to the Natu¬ 
ral bridge; and thence it pursues a northwest course to Rossie, and the upper part of Black 
lake. Another range of gneiss, however, runs in a direction a little east of north to Pierpoint 
and Parishville, and then nearly east to Duane, and north through Belmont and the south part 
of Chateau gay. We have now reached the extreme of the great primary tract; and we here 
find the Potsdam sandstone wrapped around the northern prolongation of the primary rocks,- 
resting upon them, and concealing them from view. By this arrangement, the primary does 
not extend into Canada, nor does it approach nearer the provincial line 'than about ten miles ; 
or, in other words, the northern slope which falls off into Canada, is covered by transition 
rocks. 
Turning now to the southeast from Chateaugay, "we find the gneiss to be still the principal 
rock in the direction of Redford, and then passing to Black brook near. Clintonville, where it 
meets the Ausable, whence it turns nearly due east towards Trembleau Point. Trembleau 
Point, however, is composed of hypersthene, forming the most northerly extremity of this 
rock. It therefore breaks the continuity of the gneiss district in this direction; and we may 
now pass over the termination of the Adirondack group, to Willsborough falls. If a line is 
drawn from near the falls, through Westport and East Moriah, to Minerva, then northwest to 
the foot of Long lake, and then again north to Black brook through the waters of the Saranac, 
it would separate the hypersthene rock from the gneiss. To complete the entire view of the 
gneiss district, I may add, that Minerva, Schroon, Johnsburgh, Chester, Athol, the country 
about Lake George, and Luzerne,' may be set down as being composed principally of this rock. 
By an inspection of the map, and by following the lines traced upon it, it will be seen that a 
very great proportion of the primary district of the northern counties is gneiss. The rock which 
ranks next as it regards the area it covers, is hypersthene rock, which has been already noticed, 
and its boundaries and limits pointed out. It is proper to observe, in this place, that the line 
which limits the gneiss was represented as passing over to Parishville. There is left out a 
primary district in this description, lying to the west, viz. that of Alexandria, Antwerp in the 
county of Jefferson, and Wells’s island in the St. Lawrence river. But many of the rocks 
in these and the adjacent places are of an intermediate character ; sometimes a gneiss, pass¬ 
ing into granite or hornblende ; or granite passing into, or irregularly mixed with, limestone. 
In the district which has just been described as gneiss, there are numerous patches of 
granite, limestone, hornblende, and trap rocks of minor extent. I have, however, considered 
these as of little consequence, as they do not change the essential features of this area. Thus, 
at Lake Genet in Hamilton county, there are beds of limestone; also, at Brant lake in War¬ 
ren county, and in Chester, Johnsburgh, Chesterfield, and in township No. 14. Beds of 
granite, in addition to those already noticed, are found in Johnsburgh, at the base of Crane’s 
mountain, at Parishville, Malone,' Saranac, Clintonville at the Arnold ore bed, and in Dekalb. 
Hornblende rock occurs one mile west of Westport village, and in Dekalb at or near the 
Osborn marsh, where it is found in fine crystals as well as in extensive masses. 
It will be seen, from the foregoing remarks, that gneiss forms the principal and most im- 
