Upham.] 
240 
[April 17, 
its final melting. Through the time of its maximum thickness and 
extent the ice-slieet moved south-southeastward across this area, 
and reached to the terminal moraine of Long Island, Block Island, 
Martha’s Vineyard, and Nantucket ; and onward the course of its 
border was probably east-northeast along the submarine plateaus of 
the Fishing Banks. But when a mild climate began to cause the 
glacial boundary to recede northward, the melting probably ad- 
vanced fast r upon the area of the Gulf of Maine, than upon southern 
New England, so that the ice-front became indented by a deep etn- 
bayment east of Massachusetts, toward which the latest glacial cur- 
rents along the coast were deflected. The formation of the drumlins 
about Boston seems to have taken place wholly during this time of 
deflected glacial movements, the ground moraine being massed in 
these hills on account of inequalities in the force and direction of 
the currents of the over-riding ice-sheet, when its receding border 
was probably only a few miles distant. Fifteen to twenty-five miles 
west of Boston, Mr. George H. Barton finds the trends of drumlins 
prevailingly south-southeast, in parallelism with the striation, but 
with occasional exceptions where the longer axes of drumlins vary 
much from this course, perhaps because of small indentations in the 
glacial boundary and consequent divergence of the latest ice-motion 
from its previous course. 
Fragments of interglacial marine shells occur in the till of the 
drumlins of Hull, Boston Harbor and Winthrop, Showing that while 
these hills were being accumulated marine beds on the northwest 
or west-northwest, probably within a distance of a few miles upon 
the harbor area or in the Charles and Mystic valleys, were being 
eroded ; and this erosion was continued through the whole time of 
accumulation of these oval hills of till. That the shell fragments 
occur in the drumlins more plentifully than in the drift south of 
Boston, where none have been observed, may be partly explained 
by the fact that within the valleys and to a less degree in the basin 
of the harbor the interglacial fossiliferous beds had been somewhat 
sheltered from erosion by the ice-sheet while its motion was south- 
ward, but were more fully exposed to its deflected east-southeast 
current. It has seemed to me probable that the border of the ice dur- 
ing its recession, melted by the returning warmer climate, had gen- 
erally a more steeply sloping surface than in its time of greatest 
1 Proceedings of this Society, vol. xxiv, pp. 127-141, Dec., 1888. 
