500 SHEPARD [chap. 20 



into San Clemente Basin. Similarly, where the Cascadia Channel off Oregon 

 flows across a deep ridge, it appears to have cut a deep gorge (Hurley, 1959). 

 The same was rejwrted by Laugliton (1959) for a channel crossing the fan off 

 the Bay of Biscay. The size of some of these outer channels is so large and the 

 dejjth below the natin-al levees so great that it seems quite jjossible that 

 the currents which nuist fill them from time to time^ to form the natural levees 

 have relatively high velocities, provided there is any appreciable density 

 difference with the overlying water. These channels are far larger than any 

 stream channels on land and have higher gradients than most of the large rivers 

 (Hurley, 1959). 



The channel found in the Rhone Delta front in Lake Geneva must also be 

 due to turbidity currents. This is indicated by the presence of coarser sediments 

 in the channel than on the levees. However, here the channel is probably more 

 iion-depositional than erosional since the floor is somewhat higher than it is out 

 beyond the natural levees. The gullies in the fronts of marine deltas are prob- 

 ably not due to turbidity currents because, so far as is known, they do not have 

 coarser sediment than the adjacent slopes. An exception may exist on the 

 Rhone-Mediterranean Delta since small natural levees exist here. 



C. Turbidity Currents and Delta- Front Troughs 



The troughs which extend across the shelf off" the Ganges, Indus, and some 

 other deltas are not easily exjilained. Most of them have gradients of only 

 al)out 1%, far less than is typical of canyons. Little is known about their sedi- 

 ments except in the partly filled trough off the Mississippi. Apparently the 

 latter has had turbidity currents flowing along it because there are indications 

 of displaced faunas (Phleger, 1956) and there is considerable likelihood that 

 sediments formerly were being transported into the Sigsbee Deep along its 

 course (Ewing et al., 1958; Curray, 1960). All of these troughs appear to have 

 walls of unconsolidated sediment. It is possible that they have been maintained 

 by turbidity currents during slow submergence as the delta grew u])ward on 

 either side. The turbidity currents may be provided by the abundance of 

 sediment brought in by the rivers. This could account for the location of the 

 troughs off the Indus and Ganges near the present mouths. 



9. Subaerial Erosion 



When only a few submarine canyons w^ere known most geologists had little 

 hesitation in ascribing them to the drowning of river canyons. When, however, 

 reports of the canyons began to come in from all parts of the world it was only 

 natural that skepticism should grow despite the fact that the newly discovered 

 canyons appeared to have many of the characteristics of river canyons. At first, 

 it seemed to the writer that the situation could only be explained by a great 

 sea-level lowering and in turn this seemed to indicate that the glaciers might 



1 More recent study suggests that these channels are being deepened ^ff La Tolla and 

 that the levees were formed before the deepening. 



