43 2 NEW YORK STATE MUSEUM 



orous abrasion of hard rocks would scarcely be consistent with 

 drumlins in the same locality. 



The building of drumlins by the plastering process was coincident 

 with a rubbing off and shaping effect. As masses or hills the drum- 

 lins were produced by accretion of the drift, but their peculiar form is 

 due to the erosional factor. The whole process may be compared to 

 the work of the sculptor on a clay model: a plastering on and rubbing 

 away. The accretion was due to the greater friction between clay 

 and clay than between the clay and ice. The hills of accretionary 

 drift resisted the ice impact and rasping effect just as did the hills 

 of shale. The form possessed by both classes of hills is that which 

 opposed successful resistance to the ice erosion", and the least resist- 

 ance to the ice movement. 



Dynamics. It seems evident that in the typical drumlin area the 

 ice did not move as a solid mass or even in wide sections, for such 

 motion should produce a planing or leveling effect, such as is illus- 

 trated in the Niagara-Genesee prairie [pi. 19]. The drumlins are 

 proof of a plastic flow or yielding of the ice; while the long, straight 

 ridges suggest that the ice was pushed in comparatively rigid bolts 

 or prisms that wavered and shifted. 



In the balancing and adjustment of the several dynamic factors in 

 the drift-burdened ice the two opposing forces of rigidity and plas- 

 ticity seem to be the most important. The amassing of the drift into 

 drumlin form, or at least the nonremoval of the hills, implies that the 

 depth of ice and the vertical pressure were so moderate as to allow 

 the plastic ice to override and adapt itself to the hills, while at the 

 same time the whole sheet of ice was sufficiently rigid to move 

 under horizontal thrust. 



Judging from the facts and theoretic mechanics noted above it 

 would seem that the drumlins represent the short lines of temporarily 

 diminished pressure and of lagging flow. These lines of variable 

 pressure and motion, though close set in the dominant drumlin area, 

 must have been discontinuous, short and constantly shifting. Drum- 

 lins could not have been determined, as regards location at least, by 

 external influences, as atmospheric agencies above or topographic 

 and geologic features beneath, but must have been produced by the 

 interaction of the mechanical factors resident within the ice itself, 

 the latter moving as a plastic solid. Their initiation may have been 



