654 REPORT— 1897. 



the Cliamplain and Hudson valleys to a point eighty miles out to sea. The study 

 of the striae shows a series directed southerly through the lowest line of this 

 depression, nowhere much elevated above the sea-level. 



On the west the striae point S.W., and stones have been transported in the 

 same direction. Thus fragments of Potsdam sandstone are strewed over the 

 Adirondack mountains even to their very summits, as proved in 1896 by the 

 writer. All through middle New York and into Pennsylvania, boulders of the 

 Adirondack granites may be seen. 



On the east of the central line the striae point S.E. on the summits of all the 

 Green and White mountains, and boulders from the N.W. have everywhere been 

 carried up to and beyond these summits. Laurentian boulders are found in 

 northern Vermont and New Hampshire, and, in one place at least, over the height 

 of land into Maine. 



On examining this area it seems to be a broad lobe, with striae diverging from 

 the central line, much like the barbs of a feather from the central shaft. 



Studies of the Erie, Michigan, and Superior lobes show a similar arrangement 

 of striiB, but the lobes themselves are more acuminate. 



This southern lobe is remarkable for its movement from a plain near the sea- 

 level over the highest mountains in New England and New York, 6,000 and 

 5,000 feet. 



The terminal moraines of this great glacial lobe correspond to the two sets of 

 striae, being rudely at right angles to the direction of the movement in both cases. 

 Those of central New York run meridionally, and then follow down the west side 

 of the Hudson valley. Those in New England are parallel to the margin in the 

 outer portions, and those in N. H. and Vt. run more nearly N.E. and S.W. 



As portrayed on the map, the line of junction between this southern lobe and 

 the one coming from Lake Ontario is near Salamanca, N.Y. An angle is made 

 there, which is the most northern part of the unglaciated country outside of the 

 limits traversed by the ice known in the United States. 



The moraines of the Ontario lobe are arranged in parallel looped lines, and 

 those in the immediate vicinity of Toronto belong to this series. 



If this great lobe had its origin in the Laurentian hills, it is difficult to under- 

 stand how the ice can have been accumulated at a lower level sufficiently abun- 

 dantly to move over a higher level, probably three thousand feet. It is easy to 

 see how the Ontario lobe could have made its way, as the greater altitude of the 

 rim of the basin in Ohio is comparatively slight. 



The fact that the area of the southern lobe is greater than that of any other, 

 reminds one of the map of the Great Baltic glacier given us by Professor James 

 Geikie. 



12. On tli.e Origin of Driimlins. 

 By N. S. Shaler, Professor of Geology in Harvard Universiiy. 



History of previous studies of drumlins — The question of their origin still 

 undetermined— Method of inquiry— Geographic distribution of phenomena in 

 relation to ice-sheets — Distribution of phenomena by series of forms — Importance 

 of studying drumloidal forms occurring in bed-rock and in morainal hills — Rela- 

 tion of drumlins to moraines — Evidence that drumlins are due to locally intense 

 deposition of detritus — Evidence that they have been subjected in most, if not all, 

 cases to glacial erosion — ^Analysis of the conditions of local deposition — Reasons 

 for believing that pressure-melting occurring at the base of a glacier induces the 

 formation of drumlins — Relation of drumlins to moraines formed upon previously 

 existing ridges — Phenomena of disappearance of drumlins towards the margin of 

 the ice-sheet — Probable history of drumlin growth as shown by an analysis of the 

 phenomena — Revision of the evidence in relation to the theory. 



