292 GUILCHER [CHAP. 13 



Foster, 1956; Fairbridge, 1952; Wentworth, 1938-1939; Guilcher and Pont, 

 1957 ; (hiilcher, 1958 ; Nansen. 1922 ; etc.). In the writer's opinion, the result of 

 these numerous processes is probably more efficient than wave attack on solid 

 rock; and below low sea-level, where most of them disappear, tidal currents are 

 able, in some areas, to erode quickly weak rocks, such as the Tertiary strata 

 outcrop]>ing in estuaries of the Netherlands, provided that gravels acting as 

 grinders are carried forward by them. 



On the other hand, many submarine features on shelves are to be ascribed 

 to former subaerial erosion or deposition. The submarine valleys (not the deep 

 canyons) were almost certainly eroded during one or several lowerings of sea- 

 level in Pleistocene times, when the river beds largely extended seawards. 

 Similarly, the submarine ridges in solid rock, for example off West Brittany, 

 must have been shaped by subaerial erosion. The hummocky landscapes, such 

 as those reported from the Gulf of Maine, are drowned glacial deposits 

 (drumlins, eskers or moraines). This is also true for the Baltic Sea, where a 

 beautiful submarine esker 21 miles long, 150 yd wide and 5 to 10 fm high, 

 occurring between Oland Island and Sweden, has been mapped in detail 

 (Dannstedt, 1947). Pratje (1951) has explained stony strips in the northern 

 part of the North Sea as terminal moraines belonging to the Pommern, Frank- 

 furt and Warta stages ; and, farther south in the same area, the Dogger Bank 

 is probably another moraine (Stride, 1959). Many deep depressions extending 

 across the glaciated shelves of Norway and North America off mountainous 

 areas are obviously glacial troughs ; but the longitudinal depressions in these 

 shelves may have another origin and will be discussed below. The recent 

 history is especially intricate in the heavily glaciated areas, since they have 

 been isostatically uplifted after the melting of the Wisconsinian glaciers as much 

 as 500 m in some places in Scandinavia. As a result, the parts of the continent 

 adjacent to the sea have firstly suffered ice action, which is of continental 

 nature ; then they have been drowned by the post-glacial transgression ; and 

 finally they have been uplifted, and became dry land. This shows how much 

 the continental shelf is intimately connected with the continent itself in such 

 regions. Even outside the glaciated areas, the sea-level during the Pleistocene 

 interglacial periods stood at higher levels than now (see, for example, Zeuner, 

 1950), so that the limit between dry land and sea floor also suffered fluctuations 

 in these regions, as a response to glaciations. Innumerable "raised" beaches in 

 South Britain, Normandy, Brittany, Portugal and Morocco (Guilcher, 1958), 

 as well as those reported from Florida and Georgia (Stearns MacNeil, 1950) 

 and elsewhere, point to these fluctuations. 



The deep, discontiiuious depressions in the English Channel and the North 

 Sea (Fig. 2) raise difficult problems. Since some of them are as deep as 100 or 

 120 fm, they are not as easily explained by fluviatilc erosion during the 

 eustatic lowei-ings of sea-level as are the shallower submerged valleys ; nor do 

 they seem to be exjjlained as being overdeepened by glaciers, because no 

 glacier extended into the English Channel during the Pleistocene, and, in the 

 North Sea, the ice had ajjparently no capacity for any imjjortant overdeepening 



