522 



GEOLOGY. 



those on the east side of the Atlantic by E. Hull, 1 by Nansen, 2 and 

 others; and those of the Pacific by Geo. Davidson. 3 Many channels 

 are so connected at the coast-lines with existing rivers as to leave 

 no reasonable doubt that they are but submerged portions of the 

 seaward extremities of the former channels of these rivers. Others, 

 notably some on the California coast, are not so connected; but even 

 these are usually interpreted as old drainage-valleys cut while the 

 border of the continent was above sea-level. Besides these there 

 are channels of more doubtful interpretation. Fjords, and the sub- 

 merged shelf- valleys connected with them, are very numerous in the 

 glaciated regions of both hemispheres, and undoubtedly owe some 

 of their features, and perhaps some of their abundance, to glaciation; 

 but Spencer, 4 Hull, Davidson, and others have shown that such sub- 

 merged valleys are not confined to high latitudes or to glaciated regions. 

 They appear to be phenomena common to essentially all coasts. Some 

 of the best examples are the deep channels off the mouths of the Congo, 

 the Indus, and the Ganges in low latitudes. Not only do channels 

 cross the continental shelves, but troughs interpreted by Spencer and 

 Hull as their continuations descend the abysmal slope on the outer 

 edge of the continental platforms, to depths ranging from 7000 to 12,000 

 and even 14,000 feet; in other words, practically to the bed of the 

 ocean. On the edge of the continental shelves, deep canyons havo 

 been identified, as the Hudsonian Channel, about 3800 feet deep. 5 

 These channels have usually been interpreted as evidence of vertical 

 elevations of the continents whose borders they affect. The inter- 

 pretation has usually been extended to the bodies of the continents, 

 or at least to large portions of them. With the present evidence that 

 essentially all continental borders are thus affected, and that the depths 

 are in some cases nearly equal to those of the average ocean itself, a severe 

 strain is put upon this interpretation, not only because of dynamical 

 and faunal 6 objections, but because of the difficulty of disposing of the 



1 Trans. Victoria Inst., Vol. XXX, 1897, pp. 305-324; also idem, 1900 and 1902 

 and Geog. Jour., 1899. 



2 Rep. Arc. Expl., 1904, pp. 232. 



3 Proc. Cai. Acad. Sci., Vol. I, 1897, pp. 73-103. 



4 Loc. cit. 



5 Spencer, The Submarine Great Canyon of the Hudson River, Am. Jour. Sci. 

 Vol. XIX, 1905, pp. 1-15. 



6 Dall, Tertiary Fauna of Florida; Wagner Free Inst. Ser.. Vol. Ill, 1904, p. 1544 



