Part III.] OR POISONED BY VEGETABLE GROWTH ? 
But in this continual contest, by comparison 
with their adjacents, the “lines” of rivers and 
valleys are far from being denuded. On the 
contrary, they are the favoured receptacles, the 
permanent, rich reservoirs of soil. But, accord¬ 
ing to Professor Sedgwick, they ought to be the 
denuded parts. 
This universal portage of soil by rain, the eternal 
effect of eternal causes, which in huge spaces of 
time results in such vast geological changes, 
would to some be incredible were it invisible. 
But this I think may be made visible, oculis 
jidelibus . It may be seen wherever a fence runs 
horizontally along the side of a hill. A terrace 
is then formed, for aqueous denudation goes on 
below the fence, and in chalk countries the 
ground becomes white; and not only does aque¬ 
ous denudation cease above the fence, but aque¬ 
ous deposit takes place, and the good soil which 
was on its way to the valley is arrested. Even 
a slight dead wattle, if kept up, will produce 
this effect, and, though the hedge is dead, the 
ground on which it is placed will grow; and a 
gateway wrongly placed will often let consider¬ 
able quantities of this collection of the best soil 
escape, which might otherwise have accumulated 
for what man might call for ever. So an inju- 
