224 



SCIENCE. 



LN. S. Vol. XVII. No. 423. 



kames and eskers; confirmatory evidence 

 in these cases; combined evidences; the 

 cumulative value of repetition; the inter- 

 pretation of imbedded and striated relics; 

 evidences from the interglacial deposits; 

 evidences from assorted drift lying upon 

 or outside of the true glacial series, river 

 deposits especially; scour-and-fiU ; unequal 

 effect on aggrading and degrading rivers; 

 breadth of action in the great streams of 

 the glacial area; adjustment plains as 

 sources of deception ; amount of the errors ; 

 flood-plain deposits ; bluff -border accumula- 

 tions; derivative formations; decomposed 

 secondaries; the meaning of a principal 

 river as a cause of alternate erosion and 

 deposition in the mouth of its tributary 

 valleys, and association with certain ex- 

 tinct animals not a criterion. 



Glacial Cirques and Bock-terraces on 

 Mount Toby, Massachusetts: B. K. Em- 

 erson, Amherst College. 

 Mount Toby is composed of coarse Tri- 

 assic conglomerates, and, being placed in 

 the lee of the great ridge of the Deerfield 

 trap-sheet, has been especially protected 

 from the erosive action of the inland ice 

 of the glacial period. The mountain is 

 now cut into deep amphitheaters with 

 strong, often vertical walls which head 

 against an extremely narrow ridge, and 

 are separated by ridges which are crossed 

 by many vertical rock-terraces from ten 

 to a hundred feet high, most of which seem 

 to be formed by the plucking action of the 

 ice. They do not cross the cirques, which 

 are deeply filled with foreign glacial ma- 

 terial. They seem to have been formed 

 by small glaciers, and then overridden and 

 filled by the main ice-sheet. 



Protection of Terraces in the Upper Con- 

 necticut Biver: C. H. Hitchcock, Dart- 

 mouth College. 

 Recent papers by Professor W. M. Davis 



propose the theory that many alluvial ter- 



races are kept in place by underlying 

 ledges. First, the high flood-plain was 

 deposited by the river, enormously devel- 

 oped through melting of the ice-sheet. 

 Next, as the waters diminished in volume, 

 a large excavation was made in the high 

 plain, and the steep slopes resulting are 

 the escarpments of terraces. These ter- 

 races are variable in bulk and number, and 

 it has been a difficult matter to explain 

 these variations. As the river swings back 

 and forth over the low ground, it meets 

 ledges which it is unable to remove ; neither 

 can it remove the earth superimposed upon 

 them. In other words, the ledges protect 

 the terraces behind them from destruction. 

 It is common to see a ledge at the extreme 

 pointed end of a terrace, and the terrace 

 broadens as you follow it back. The outer 

 edge below the ledge will ordinarily be 

 curved. The localities described by Pro- 

 fessor Davis are at Westfield, Mass., and 

 Bellows Falls, Vt. I have extended the 

 observations above that point, especially 

 between White River Junction and Wells 

 River. 



Glacial Features of Lower Michigan: 

 Frank Leveeett, Ann Arbor, Mich. 

 The paper was illustrated by maps and 

 diagrams setting forth the relation of the 

 Michigan, Saginaw and Huron-Erie ice- 

 lobes in lower Michigan during the Wis- 

 consin stage of glaciation, and represents 

 three years of field work for the United 

 States Geological Survey. Especial atten- 

 tion was given to eskers, and their bearing 

 upon the question of superglacial or sub- 

 glacial origin is discussed. An extensive 

 drumlin area was briefly described in its 

 relation to ice movements and moraines. 

 The paper closed with reciting the evidences 

 of earlier ice invasions than the Wisconsin 

 which have been found within the limits 

 of the Wisconsin drift. 



