SECT. 2] SUBMABINE CANYONS 501 



have been much larger during the Pleistocene than had commonly been con- 

 sidered to have been the case. As studies of the oceans continued, Pleistocene 

 sea-level lowering was definitely excluded because the dredging from the 

 oceanic seamounts refuted such a relatively recent lowering. On the other hand, 

 these seamounts, and especially the flat-topped guyots, have yielded ample 

 evidence of subsidence over a long period of time. Similarly, the continental 

 margins in many places contain sediment thousands of feet thick which is 

 now definitely far below the level at which it was deposited (see, for example, 

 U.S. Geological Survey's Tectonic Map of the United States, 2nd edn.). Therefore, 

 it should not be a matter of surprise if coastal canyons had been deeply 

 submerged, at least in many places. 



Proceeding from the premise that coastal canyons have probably been 

 submerged, at least locally, we run into some obstacles in using this as an 

 explanation for submarine canyons. First: why are the canyons world-wide? 

 This would indeed be a problem, if we were certain that they are world-wide, 

 but this we do not know. As has been explained previously there are numerous 

 types of submarine valleys, only a few of which have been sufficiently surveyed 

 so that they can be shown to possess true river-canyon characteristics. Thus 

 the Corsican canyons, those of the French Riviera, Carmel Canyon, the head of 

 La JoUa Canyon in California i, the canyons at the tip of Cape San Lucas in 

 Baja California, Tokyo Canyon, and Trincomalee Canyon of Ceylon can be 

 considered as definitely within the fold. All but one of these is cut into hard 

 rock. A number of others, particularly a group of canyons along the California 

 coast can be considered as probable, but most of the reported but unexplored 

 marine valleys could be due to such other causes as faulting, turbidity-current 

 channels kept open during the deposition of deltaic masses, or even as scars 

 left by giant landslides. Therefore, at the time of A\'riting, there seems to be 

 notliing particularly disturbing about the idea that some coasts have sub- 

 merged, bringing river canyons deep below the sea. 



The next obstacle offered to the idea of a remote submergence of river 

 canyons is that the canyons should have been filled. The strength of this 

 objection is, however, considerably weakened by the discoveries of landslides 

 or of some tjrpes of sand flow or mud flow operating in the heads of many 

 canyons. These slides undoubtedly set up turbidity currents carrying the 

 sediment out of the canyon and depositing it as a fan on the outside. The 

 writer knows of no reason why such periodic flushing of the canyons may not 

 have operated for periods of millions of years. In fact the huge size of the fans 

 outside the canyons indicates that this has been going on for extended periods 

 (Menard, 1960). Therefore, this objection does not seem to be very valid. 



Perhaps more troublesome is the question of why the submarine canyons, if 

 due to ancient erosion, still have continuous outward slopes, particularly in 

 such unstable areas as California. Why has there been no faulting or folding 

 action which has cut off sections of the canyons? The head of Delgada Canyon 



1 Recent evidence reported by R. F. Dill shows that the head of this valley is retreat- 

 ing owing to submarine erosion. 



