156 GEOLOGY. 



towards the Red Sea is due to marine action is a very- 

 difficult question. If the sea ever exerted any great 

 influence "on the denudation of the plains at the base of 

 the range, it must have been at a comparatively remote 

 period ; for so far as they were examined, the hills pre- 

 sented a very different aspect from that of a sea cliff. 

 Still it is highly probable that, before the commencement 

 of the volcanic outbursts which have left their traces 

 along the whole southern portion of its shores, the sea 

 extended farther west, and very probably it reached the 

 foot of the hills in places. But tropical seas, and espe- 

 cially calm land-locked basins, like the Red Sea, are 

 trifling agents of denudation when compared with such 

 oceans as the Atlantic. Independently of the local 

 causes which determine the size of the breakers, most 

 tropical coasts are more or less protected either by coral 

 reefs, where no sediment is brought down by rivers, or 

 by the deposition of that sediment itself, which, owing 

 to the more violent rainfall (a far more important item 

 than the absolute amount, although the latter is also 

 usually excessive in the tropics), and to the more rapid 

 disintegration of rock, is far larger in quantity than in 

 temperate regions.^ 



' I believe that the comparative importance of fresh- water and marine 

 denudation is to this day misunderstood by many of the best geologists of 

 Western Europe, and is only beginning to be appreciated by any of them, in 

 consequence of the exceptional conditions there occurring. The whole of the 

 circumstances attending the rainfall in tropical countries are far more favour- 

 able to denudation than in the temperate zone. Compare more especially 

 the case of the British islands and the neighbouring portions of the Con- 

 tinent — ^the part of the world, that is, in which by far the greatest number of 

 careful and continuous observations upon such phenomena as denudation have 



