NV. S. Shaler—F luviatile Swamps of New England. 211 
It is a fact, one that so faras Iam aware has not been noticed, 
that the uppermost of these terraces in the Merrimac and Con- 
necticut, as well as some other rivers, generally exhibits fea- 
tures not reconcilable with the supposition that it was formed 
by river action. Its surface is frequently cast into the peculiar 
forms characteristic of the deposits which were made at the 
front of the ice-sheet when the base of the glacier lay below the 
level of the sea. It is my belief that these kame-bearing ter- 
races were formed while the valleys in which they lie were 
depressed. Mr. Warren Upham in his essay on the New 
Hampshire drift, repeatedly refers to the fact that these plains 
differ essentially from those forming at present, still he con- 
ceives them as formed by river action. 
Although we must exclude the upper ‘terraces, those having 
kame ridges and the crater-like hollows which accompany kame 
deposits, from the category of terraces formed by the action of 
rivers, there remain enough of the elevated benches, distinctly 
marked by the action of fluviatile forces, to make it clear that, 
in the valleys of the rivers flowing from north to south, the con- 
ditions have been such that the streams have had no difficulty 
in constantly cutting deeper and deeper into the detrital depos- 
its which hindered their flow at the close of the Glacial period. 
The accompanying diagram (figure 1) represents the general 
condition of the terraces in the valleys of the Merrimac, Connec- 
ticut, and other New England rivers of considerable size, which 
flow from north to south. It is intended to give the typical 
form of these terraces and not to show the actual conditions at 
any one point, though those who are familiar with the aspect of 
the Merrimac or the Connecticut will recall many sections which 
it closely resembles. 
1 
Tervaces. 
Channet 
ngland when the streams flow from north to 
south. 
It is evident that in a valley characterized by the presence 
of the terrace debris the conditions are extremely unfavorable 
for the formation of swamps, for even the lowest of the alluvial 
plains is drained, except in times of flood. 
Turning now to the streams which flow from south to north 
we find conditions in marked contrast to those which are found 
in the rivers flowing in the opposite direction. The number 
of these northward flowing streams is small and none of them 
have drainage areas to be compared with those of the greater 
