NV. S. Shaler—F luviatile Swamps of New England. 215 
gram (see figure 2) we often find on either side of the present 
stream bed the remains of a terrace-like mass of sand and gravel 
which is the product of peculiar conditions of the Glacial period. 
In some rare cases these masses of tabular drift are found as 
fragments of an old table-land left in the middle of the valley. 
These lateral and central masses of terraced sands have in all 
cases steep escarpments which by their slope indicate the erosive 
action of the stream. It is often evident that these drift plains 
at the close of the Glacial period extended across the valleys in 
which they lie. 
In some cases the mass excavated is very great; on the 
average it seems to me as great as that which has been removed 
by some of the larger streams of the valleys which flow from 
north to south. The excavation in the valleys of the rivers 
which flow from south to north is entirely beyond the power of 
their streams to accomplish with their present slopes. In simi- 
lar streams flowing in the reverse direction the process of ero- 
sion of their drift plains is still going on. In the Merrimac, for 
instance, the river is rapidly wearing away its banks at many 
points. On the western side of the elevated plateau of kame 
plains, which lies on the east side of the river near Concord, 
N. H., the process of erosion keeps a fresh face on a cliff nearly 
one hundred feet high and several hundred yards in length. 
This escarpment is wearing away at the rate of somewhere near 
one foot perannum. The same is the case with many other 
bluffs on this stream. This activity of erosion by the rivers is 
seen in many other streams of this class in this part of New 
England, but in the valleys which slope to the south the streams 
no longer attack their banks in this manner though it is evi- 
dent that they at one time were capable of such work. 
I have at present no data sufficient to determine the depth 
to which the valleys have been encumbered by the swamp 
accumulations which have been formed in them or how much 
their beds have been raised from their original level. It seems 
to me likely that they originally possessed alluvial plains such 
as those which now characterize the Merrimac and other nor- 
mal valleys. This would seem almost necessarily the case from 
the extensive erosion of the drift terraces, for the reason that 
this erosion by the stream would necessarily lead to the oscilla- 
tion of the channel from side to side, which would bring about 
the production of fluviatile plains. If such plains existed the 
filling up of the valleys has evidently been sufficiently great to 
suppose their surface beneath the existing swamp deposits. 
As before remarked I am disposed to believe that the great 
area of wet-woods on the banks of the Concord river in Bedford, 
Mass., is in its nature a supposed fluviatile terrace. 
It thus appears probable that after the streams which flow to 
Am. Jour. Sci.—TuirpD Surizs, Vou. XXXII, No. 195.—Marog, 1887. 
14 
