TERRACES AND RIVER VALLEYS OP WESTERN EUROPE. 277 



same distance ; a third off Cape Toriiiana of about 9,750 feet, 

 where the 200 fathom contour is very nearly over a somiding 

 of 1,825 fathoms. These nearly sheer descents have their 

 counterparts in those which bound the submerged terraces 

 off the coast of North America and the West Indian Islands.* 



It is not to be supposed, however, that the descent from 

 the edge of the platform to the base was by any means 

 uniform. The isobathic lines indicate great variations in the 

 amount of slope — with (in some cases) wide terraces inter- 

 vening between steep, perhaps precipitous, descents. This 

 is what might be expected as the result of wave action 

 along emerging or subsiding land, accompanied by occasional 

 pauses in the movement, and amongst rocks varying in 

 character and hardness,! 



Thus off Cape Prior (lat. 44' 10' N. and long. 9° W.) the 

 contours clearly indicate wide terraces ending off in steep 

 descents at the margin of the oceanic floor, and similar 

 features are observable both m the region lying to the north 

 of the grand canon of the Adour in lat. 44° N. and long. 3° 

 W. — and off Cape St. Vincent. A larger number of sound- 

 ings are, however, required in order to arrive at a better 

 knowledge of these details. 



As far as I have been able to judge from the soundings a 

 complete change in the foi-m of the oceanic floor takes place 

 between Cape St. Vin-cent and the Straits of Gibraltar. The 

 Grand Declivity appears to widen out, and to give place 

 to a gradual slope descending from the margin of the land 

 to the abyssal regions of the ocean, probably in a succession 

 of steps or terraces. 



IV. Submerged Riyer-Chanxels or Canons. — It is owing 

 to the existence of river-channels, sometimes traceable up 

 to those now entering the ocean from the land, that we 

 are enabled to arrive at the conclusion that the ocean-bed 

 was formerly, to some extent, a land surface ; and it is also 

 by means of these features tiiat we can ascertain the extent 

 of the former emergence. It is only on emergent lands 

 that rivei'S can wear down their channels; for once thev 

 have entered the sea their currents are checked, or 



* See map to Pi^ofessor Spencer's " Reconstruction of the AntiJlean 

 Continent," Bull. Geol. Soc. Amer., vol. vi, 1895 ; Geol. Mag., Nov. 1898, 

 Plate, p. 515. 



t It must be recollected also, that the effect of wave-action on the 

 subsiding escarpment would have been in the direction of reducing the 

 amount of the original slope which after the maximum emergence uiiv 

 have been much greater than at present. 



X 



