528 GEOLOGY. 



These changes of slope and of gravity obviously tended to cause these 

 soft beds to creep back toward the abysmal basin. This tendency 

 may well have been greatest at the edge of the continental shelf, where 

 the newer and softer beds may naturally have been thickest. This 

 creep may therefore have carried the outer ends of the channels pre- 

 viously formed, down to depths much below the relative horizons 

 at which they were eroded. Adjacent to the deep channels off the 

 mouths of the Congo, the Indus, and the Ganges, the edge of the lower 

 part of the continental shelves is observed to be somewhat protrusive. 

 This may, of course, be due to greater building-out at these points; 

 but the fact is at least consistent with the conception here entertained 

 and the contours are observed to be spread apart on the base of the 

 slope, instead of being crowded together as might be expected from 

 normal delta-building. 



Cooperative water-displacement. — The basal deformative move- 

 ment, by deepening and extending the great basins, .tended to draw 

 down the waters on the borders of the continents and hence aided in 

 the emergence. The postulated reversed movement of the shell and 

 the continental platforms tended in the opposite direction and aided 

 in the subsequent advance of the sea on the continental border. So, 

 too, the accelerated stream-erosion resulting from the increased pro- 

 trusion of the land tended slowly to lift the sea-level by the transfer 

 of sediment from land to sea. 



Tidal cooperation. — Under any hypothesis it seems remarkable 

 that river-channels could be submerged without being filled in the 

 process, for the rivers must have been carrying detritus, and coastwise 

 currents must have swept drift into the channels. River waters can 

 scarcely be supposed to have been very efficient in erosion after they 

 reached the coast, for they were fresh and relatively light, and should 

 have spread out on the surface of the salt waters. The efficient agent 

 in the case was probably the tides. Their entrance and exit, par- 

 ticularly where the river-mouth broadened to an estuary, as it was 

 likely to do at the beginning of a submergence after a period of active 

 erosion, doubtless scoured the channel, and not improbably enlarged 

 and deepened it where the coast configuration was favorable. This 

 was not improbably true of some of the channels at all subsequent 

 stages of submergence, where they were favorably situated relative 

 to tidal movements, and such channels may owe not a little of their 



