36 TACONIC PHYSIOGRAPHY. 
upper terraces of the Hudson, has a meandering course owing to th< 
horizontality of the surface (PL VII, A). At Schaghticoke it flows 
first southwest, then northwest and south, then north-northeast, anc 
finally west-northwest, all in a distance of 2 miles. It has cut a can 
yon from 100 to 200 feet deep through over 100 feet of glacial drift 
and terrace material and a varying thickness of shales and grits 
During this part of its course the fall is 100 feet. As shown by the 
arrows in the figure, some parts of its meanders are parallel to the 
strike of the shales cut by the river, but the outline of the edge of the 
gorge shows that the course of the stream was fixed long before the 
rock was reached and was of post-Glacial date. The post-Glacial 
work of streams is also shown by the fact that the ravines on the Green 
Mountain Range, ancient as they must be, are usually choked with 
huge bowlders. For several miles north and south of the mouth oil 
the Hoosic the lowest terrace of the Hudson is cut by small ramifying 
east-west ravines, from 1 to 3 'miles long and up to 100 feet deep,j 
which reach down to the shales in parts of their course and cross 
their strike (PI. VII, B). These must all be of latei date than the! 
post-Glacial uplift of the Lake Champlain region. 
GLACIAL EROSION. 
Although glacial deposits are outside the province of this paper,, 
yet. inasmuch as glaciation of the rock surface was a factor in the 
topography, it will be here briefly considered. As is well known, the 
prevailing course of the glacier in this region was southeast or south- 
southeast. The striae on Mount Grey lock are S. 40° E. The course 
of the well-known Richmond bowlder, one of the largest in New 
England, from its source at the Knob in Canaan, N. Y., to its site 
about 6 miles distant at the eastern foot of the Taconic Range, was 
southeast. That course was traced about a half century ago by 
Edward Hitchcock and Charles Lyell. & A bowlder of metamorphic 
grit recently observed near Pittsfield, Mass., lies over 11 miles, south- 
east, from its source in Rensselaer County, N. Y. Bowlders from 
" Warren Upham gives the vertical amount of marine submergence along the basin of 
that lake, as shown by fossiliferous beds of modified drift supplied from the melting ice 
sheet and resting on till, as from 300 to 400 feet, increasing from south to north. 
i Upham, Warren, The Champlain submergence : Bull. Geol. Soc. America, vol. 3, 1892, 
pp. 508—511.) Baldwin shows that the valley in pre-Glacial time was occupied by a 
river, the level of which was far below the present lake surface. The valley was then i 
glaciated, depressed, filled with a glacial lake, then submerged below sea level so that sea i 
beaches were formed in its northern part at what is now the 500-foot level, and finally 
elevated to its present position. (Baldwin, S. Prentiss, The Pleistocene history of the 
Lake Champlain Valley : Am. Geologist, vol. 13, 1894, pp. 170, 184, PI. V. See also Peet, 
Charles E., Glacial and post-Glacial history of the Hudson and Champlain valleys : Jour. 
Geol., vol. 12, 1904, pp. 415-469, 617-660.) 
b Hitchcock, Edward, Am. Jour. Sci., vol. 49, 1845, pp. 258-265. Lyell, Charles, Proc. 
Royal Inst., 1855; also, in "Geological Evidences of the Antiquity of Man," Chapter 
XVIII. 
