274 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1937 



prove the glacial-excavation hypothesis. It is well known that in 

 areas of continental glaciation numerous glacial striations cross the 

 general direction of ice flow. Also, as has been pointed out by 

 Bretz," the present continental glacier of Greenland has a concen- 

 trated flow along the fiords which are buried beneath the ice of the 

 interior. This observation helps explain why there is such clear 

 evidence of excavation along certain valleys in formerly glaciated 

 regions whereas the uplands between the same valleys were not sub- 

 ject to much glacial erosion. 



That the ice did move down the Great Lakes basins is clearly in- 

 dicated. The map compiled by Leverett '^ shows the various lobes 

 of ice which occupied each of the basins. These lobes have been 

 demonstrated by the glacial striations and by the morainic belts 

 which surround the outer termini of the Great Lakes basins. 



COMPARISON WITH FIORDS AND OTHER GLACIAL FEATURES 



It is quite generally conceded that fiords are due largely to glacial 

 erosion. Also, the large lakes of Switzerland and other glaciated 

 mountain ranges are certainly glacial excavations.'^ The writer 

 called attention to the complete gradation which exists between 

 fiords and submarine troughs out on the open continental shelves 

 oflF glaciated coasts.'* The Great Lakes, likewase, can be shown to 

 be related to the mountain lakes of S^\^tzcrland with all gradations 

 between the two. All four of these features have many points in 

 common. Thus, while the Great Lakes are not strictly U-shaped, 

 their transverse profiles are trough-shaped with relatively steep walls 

 on either side (fig. 2). Also, the sides of the lakes are comparatively 

 straight as are the sides of many glacial valleys in the mountains 

 and the fiords of glaciated coasts. The unconnected depressions 

 which characterize Lake Superior (fig. 5) have their counterpart in 

 most fiords and in many large mountain lakes. The rocky islands 

 mthin Lake Superior, which were cited by Thwaites as an argument 

 against ice erosion, are matched by the rocky islands of most fiords 

 and are found also in some of the Swiss lakes. 



In 1870 Andrews '* called attention to certain subaqueous terraces 

 along the margin of Lake Micliigan which he thought extended to 

 depths of 60 feet. Johnson '^ subsequently examined the evidence 



" Bretz, J. H., The flord region of East Greenland. Geog. Soc Spec. Publ. 18, p. 239, 1935. 



» Leverett, F., and Taylor, F. B., The Pleistocene of Indiana and Michigan. U. 8. Geol. Surv. Monogr. 

 63, pi. 5, opp. p. 62. 



'» The attempts which have been made to explain fiords and mountain lakes as diastrophic are not char- 

 acterized by the clear reasoning which scientific arguments should have. This point is well developed by 

 D. W. Johnson, The origin of fiords. Science, n. s., vol. 41, pp. 537-43, 1916. 



'♦ Shepard, F. P., Glacial trough of the continental shelves. Journ. Geol., vol. 39, pp. 347-57, 1931. 



" Andrews, Edmund, The North American lakes considered as chronometers of post glacial time. 

 Chicago Acad. Sci. Trans., vol. 2, pp. 1-23, 1870. 



'• Johnson, D. W., New England- Acadian shoreline, pp. 406-16, 428, John Wiley <ic Sons, New York, 

 1026. 



