236 ACTION OF 



water above the level of the sea, unless on the ab- 

 surd supposition that the water ran up one side of 

 the mountain for no other imaginable purpose than 

 that it might run down the other that it acted con- 

 trary to the law of gravitation on one side of the 

 summit, just in order that it might the more readily 

 and effectually obey that law on the other side. 



Even in plainer cases than that those in which 

 there is only a gradually inclining dell, with a rivulet 

 meandering along we cannot give the rivulet the 

 merit of making the dell, at least not out of the hard 

 strata of primary rock ; because, unless we have the 

 dell at the beginning, we cannot explain why there 

 is a rivulet there ; even rivulets, if they are to be 

 permanent, must have permanent causes ; and, un- 

 less where there is a spring supplied with water 

 from a store farther up, the sloping sides of the dell 

 are necessary to provide the rivulet with water. In 

 rivers of longer course, that is more striking. Take, 

 for instance, the Thames ; the quantity of water that 

 rises up in the springs, though much magnified in 

 the pictorial representations, is in reality consider- 

 able : but look at the distance to the sea, and think 

 whether, instead of that infant stream having exca- 

 vated the goodly valley of which it is now the wealth 

 and the ornament, it must not have been evaporated 

 before it could have reached Windsor or even Ox- 

 ford. 



That a river can cut deeply even into very hard 

 strata is proved by many instances; but in those 

 instances there are always slopes above to send 

 down in a flood, during rains, that water which, if it 

 fell on level ground, would sink into the earth, and 

 not form any flood at all : so that there could be no 

 cutting through even the softest materials. These 

 cuttings are, in general, the secondary strata, or 

 even collections of rubbish ; and there is perhaps no 

 instance of a dell formed by the action of water 

 wholly in the solid granite, though in many places 



