them to have been subglacial and others seeing proofs that they 
were formed in channels on the ice surface. 
Professor Stone, in the articles before cited, concludes that the 
many and very long series of osars in Maine were mostly depos- 
ited in narrow, canon-like river-courses on the surface of the ice- 
sheet where its melting had advanced so far that it had a thick- 
ness of only a few hundreds of feet, while its motion hd nearly 
ceased. My own opinion of the osars or eskers of the Saco, Mer- 
rimack, and Connecticut valleys, examined for the New Hamp- 
shire Geological Survey under the direction of Prof. C. H. 
Hitchcock, is that they were progressively deposited near the 
ice-front in channels which were cut backward into the retreat- 
ing edge of the ice by such superglacial streams. 1 Prof. G. F. 
Wright has carefully studied the Haverhill and Andover osar 
series in northeastern Massachusetts, and, after seeing the sub- 
glacial and englacial rivers of the Muir glacier in Alaska, he 
refers the formation of our osars to streams produced by the final 
melting of the ice-sheet, and seems to agree with Professor Stone 
and myself that this drainage was mainly in superficial channels. 2 
Numerous osars observed in Minnesota by Prof. N. H. Winchell 
and the present writer have been attributed by us to streams 
flowing in channels on the wasting border of the ice-sheet ; 8 and 
in Manitoba I have found the same explanation applicable to 
osars near Winnipeg, where the gravel and sand of the glacial 
river were evidently underlain at the time of their deposition by 
a thickness of about 500 feet of ice. 4 
A different view is taken by Professor Shaler, whose studies of 
osars and kames on Martha’s Vineyard, Nantucket, and Cape 
Ann, in other parts of New England, and in New York, lead him 
to interpret them as deposits formed by subglacial streams flow- 
ing in arched tunnels beneath the ice-sheet during its recession ; 
and he further believes that their accumulation took place be- 
neath the level of the sea upon the coastal region, or beneath the 
^roc. A. A. A. S., vol. xxv, 1876, pp. 216-225. Geology of N. H., vol. iii, 1878, 
chapter i. 
2 Proceedings of this Society, vol. xix, 1876, pp. 47-63, with three maps; vol. xx, 
1879, pp. 210-220. The Ice Age in North America, 1889, chapter xiv. 
3 Geology of Minn., Final report, vols. i and ii. 
4 Geol. and Nat. Hist. Survey of Canada, Annual report, vol. iv, pp. 36-42 E. 
