DRIFT DIVISION. 187 
They have but a limited distribution, being found only at the foot and on the northwestern 
declivity of Bellevale mountain, in Warwick.”* 
The region over which Dr. Horton has described the distribution of boulders and erratic 
blocks, is the great valley between the Highland mountains on the southeast, and the Sha- 
wangunk and Kittatinny mountains on the northwest, embracing the Walkill and Long Pond 
valleys, and parts of the Smith’s Clove and Mamakating valleys. The white conglome¬ 
rate or Esopus millstone rock of the northern part of the county, and many of those of 
other parts, are undoubtedly from the Shawangunk mountains, which are mostly covered by 
this rock in place. The lahradorite is the same as the labrador feldspar which forms moun¬ 
tains in Essex county, described by Prof. Emmons ; and no other locality of this rock is 
known in the United States, where it exists in place. Boulders of it, of considerable size, 
are scattered over the Hudson valley from its source to New-York, and down the Great valley 
to Pompton plains in New-Jersey. Boulders of many tons weight are not uncommon in 
many of the high valleys of the Highlands, where I have often seen it. Near Coxsackie, at 
an elevation of more than one hundred feet, perhaps two hundred feet above the Hudson, 
about half a mile below the village, a large block or boulder of this labrador feldspar rock 
was partly exposed above the surface. Masses weighing, it is supposed, fifty to eighty tons, 
were blasted from it down to the level of the ground. 
The boulders of the Mamakating valley are composed of the rocks of the adjacent moun¬ 
tains, mingled with the northern drift. 
The valley of the Schoharie kill contains some boulders of the northern drift, intermingled 
with those of the rocks of the adjacent mountains. At Schoharie, blocks of opalescent feld¬ 
spar like that of Essex county, and boulders of a rock composed of augite, coccolite, sphene, 
plumbago, colophonite and hornblende, occur. Rocks containing the three first of these mine¬ 
rals are common in many places in the Highlands, and the adjoining ranges and hills ; but 
there are peculiarities of appearance in these minerals, and the associated ones in these boul¬ 
ders and blocks, precisely similar to those of rocks in place on Lake Champlain at Wills- 
borough, and on Lake George near Rogers’ rock. The identity of appearance, and mineral 
associations, and no other localities than those mentioned being known, give weight to the 
conclusion that they were derived from those localities. Boulders of granite, gneiss, horn¬ 
blende-rock containing garnet, pyroxene, scapolite and epidote, are common on the mountains. 
Novaculite also is common.t 
In Saratoga county, the boulders between Schenectady and Ballston-Spa are mostly of the 
northern primary rocks, interhiixed with the Potsdam sandstone and other rocks of the region 
to the north. Most of them may be referred to their parent localities at no great distance. 
They are very abundant in many places. About ten miles from Saratoga on the road to 
Sandy-Hill, boulders abound. 
• New-York Geological Report, 1838, p. 159, 161 
t American Journal of Science, Vol. 28, p. 176. J. Gebhard. 
