Rev. E. Hill — Permanence of River Valleys. 71 



part of their channels ... the valleys at first almost obliterated 

 are now reassuming their old appearance." 



A similar effect may be produced in a different manner.^ Such 

 superincumbent drift will generally be porous, and more porous 

 than the lost land-surface. Water sinking through this porous drift 

 will sink to that old surface and find its way along that to the old 

 hidden channels. Then flowing along these it will tend to erode 

 subterraneously, and by the consequent land-sinkings it may produce 

 a system of surface channels which will follow the lines of the 

 former drainage system. I suppose the subsidences in the Cheshire 

 salt-working district indicate hidden hollows of Triassic times : 

 similar causes may lead to similar results. 



Man makes reservoirs by dams across valleys; so Nature with 

 landslips, moraines, lava-flows. When the water has risen above its 

 dam, the river will commence to cut a channel through, and will 

 resume its previous course. Sometimes, however, it is able to over- 

 flow across a lateral col : then the whole drainage system is altered ; 

 a lake replaces the old valley, and this begins to be silted up. Yet 

 even now the valley may not be obliterated for ever. Allow 

 sufficient time, the new channel will be cut down to the lake-floor, 

 the lake will be empty : its bed will now seem a pair of opposite 

 valleys meeting at a common outlet. What was depression before 

 will be depression still. The result will be a physical feature such as 

 we now see on the south-east side of the Mont Blanc chain. A vast 

 moat extends below that mighty wall, from the Col de Ferret to the 

 Col de Seigne, collects the whole drainage of the south-east face, and 

 pours it out, opposite the central pass of the Col du Geant, through 

 the sluice-like cleft where Courmayer stands. I do not affirm 

 that the Allee Blanche was so produced ; but an uprising of the 

 Col de Seigne would have been able to produce it. 



A lava stream down a valley would produce most serious 

 alteration, even if it did not completely fill the trough, for the 

 materials of the cast would be more resistant than those of the 

 mould. Sometimes, indeed, erosion might go on along its edge, and 

 create a new valley roughly pai-allel to the old. But lava-flows are 

 exceptional agencies. 



The depressions and re-elevations hitherto considered have been 

 regarded as simple drops or lifts without tilt. It was convenient 

 to keep separate the effects of deposition and the effects of altered 

 inclination. Similarly, in considering the consequences of tilts, let 

 us separate those which only alter the fall of the valley-bed from 

 those transverse to them which alter the slope of the valley-sides. 

 The former, if they increase the fall of the river, will increase 

 erosion and deepen the valley ; if they decrease it, will check 

 deepening, may cause deposition to begin : they may reverse the 

 flow, with erosion continued ; they may create lakes, which may fill 

 up parts of the valley. But very rarely can a valley by this means 

 be so obliterated that every trace is lost. 



1 Suggested by a friend in conversation. 



