TORRENTS, ETC., BY COSTA DE BASTELICA. . 121 



natural stable declivity ; and from that time vegetation could begin to take 

 hold and complete the extinction. 



" This remark is important in this way, that if the disappearance of a 

 forest always gives birth to torrential disturbances, it does not always hold 

 true that one can put a stop to them by the planting of a forest alone. 



" Much as an unstable ground is protected by being wooded — though it 

 maintains itself and behaves in a hydrological point of view as do the most 

 solid lands, if the wood come to disappear, if the ground be deeply ravined, 

 if the bottom of the ravine continues to be easily undermined and washed 

 away — it becomes extremely dif&cult to establish vegetation on the moun- 

 tains, which continually crumble away, and which with this instability no 

 longer retain any trace of vegetable soil. 



" In the Alps there are numerous cases of old mountains which crumble 

 away when the foot of the slope is undermined by the water. And one is 

 thus left, if he desire to effect a radical and prompt extinction of a torrent, 

 to give, artificially, to the bottom of the ravine a power of resistance to 

 undermining and washing away, by appropriate works of consolidation. 



" But be this as it may, the potent action of forests is beyond aU question. 

 Whatever be the character of the woods — timber forest, coppice-wood, or 

 simple shrubbery — all contribute to give firmness to the soil, to retard and 

 to regulate the flow of the water drained off. 



" In comparing the different kinds of woods, it may be said that lofty 

 timber forests, with their vast apparatus of foliage at a great elevation above 

 the soil, are of most use with a view to meteorological and hydrological effects; 

 and that young trees serve perhaps better to insure the consolidation 

 of the soil on steep declivities. But as generally, on poor land, the soil of 

 timber forests covers itself with branches, &c., it follows that a mixture 

 of the two kinds of woods accomplishes best the end which it is sought to 

 effect." 



In regard to gazonnennewt, he says,—" To report efficiently the influence 

 exerted by a bed of turf or herbage covering the soil, it is necessary to 

 follow the very interesting natural process which goes on when pasturing is 

 suppressed on land, till then, given up to the abuse of pasturage. 



" The facts which I am about to state are not exclusively theoretical, they 

 are confirmed by numerous experiments of enclosures which have been 

 made during a great many years. 



" I ought first to make an important remark on the subject of the 

 different disintegrations of the soil which occur. Some are simply superficial, 

 and in no way compromise the stability and the solidity of the bulk of the 

 ground. The surface is more or less disintegrated, but the subsoil is 

 unmoved. With others, on the contrary, if they occiir on unstable grounds, 

 or on grounds badly poised, breaking up the mass, they tear it deep and 

 move it to its very foundation. 



" It is apparent from what is said that the influence of turf or herbage, 

 even if it could be produced there, can be of alipost no effect in this latter 

 case. There is nothing but woods, with their strong and deep roots, which 

 can render firm and protect a soil so unstable, and often on such the wood 

 itself requires the assistance of artificial works of consolidation to sustain it. 

 But the action of turf or herbage is, on the contrary, very powerful if it 

 be employed only to repair a superficial evil by removing traces of erosion. 

 Let us take up, then, the description of the work of nature. 



" When the soil is no longer trampled, and the few herbs which it is 



