216 WV. S. Shaler—F luviatile Swamps of New England. 
the northward had in good part done their work of excavating 
their drift-encumbered channels, bringing them into essentially 
the condition of those occupied by the rivers flowing from north 
to south, a change came over them which led to a lowering 
of their slopes and a consequent diminution of their fall. It is 
evident that the only change which could produce this would 
arise from a tilting of the land area in which they lie. This 
tilting might be accomplished by a positive down sinking of 
their head waters or by a positive elevation of the lands about 
their mouths; either of these movements might be local in 
their nature or they might be a part of a general movement of 
the continent: ; 
It is perhaps too soon to interpret the movements which have 
led to these peculiarities in the drainage of the two classes of 
our New England rivers, but in order to secure a basis for the 
further study of the facts I venture to give the succession of 
events which my observations appear to indicate. 
First as to the origin of the kame terraces which are by far 
the most conspicuous as well as the most inexplicable features 
of our river valleys. I have already incidentally given the 
principal reasons why they are not to be considered as in any 
way the result of river action. I have stated that the character 
of their surfaces as well as the order of stratification of their 
materials excludes them from the class of fluviatile deposits. 
A further study of their features shows us that there are other 
reasons why they cannot be regarded as the product of river 
action. Deposits of essentially similar nature are found along 
the shore line at many points between Portland and New York, 
especially on Cape Cod and the islands of Nantucket and Mar- 
tha’s Vineyard where the supposition of river action is quite 
out of the question. Again they are found in the valleys of 
our smaller lakes in positions where they could never have 
been exposed to river currents. Still further they are, when 
found in the river valleys, in no very definite relation to the 
plane of the stream, they lie at all heights above its level and 
only for short distances and accidentally correspond in slope 
with the surface of true fluviatile plains. I am, therefore, 
driven to the opinion that these kame terraces were formed be- 
neath the level of the sea during the submergence which at- 
tended and followed the stage of the Glacial period when this 
surface was deeply covered with ice. During the period of 
submergence which endured for a time after the disappearance 
of the ice the surface of the sea floor was swept by strong cur- 
rents which tended to move the detrital matter from the eleva- 
tions and accumulate it in the hollows of the bottom. The cur- 
rents which bore the materials to the sea and arranged them on 
its floor were in the main produced by the sub-glacial streams 
