GEOLOGY 
to flow at first over rock of uniform composition at an early stage in 
the process of elevation. As the land gently rose and a larger area was 
exposed, the streams enlarged their channels, and probably maintained 
their original courses without any important deviations all through the 
period while the channels lay exclusively through the uppermost rock. 
By degrees, as the upheaval slowly proceeded, and the waste of the 
surface went on, the outer coating (if one may so express it) of the 
dome, was worn through, and the head waters of the rivers flowed sea- 
wards across rocks of quite a different kind from that in which their 
channels commenced. For, assuming that the outer envelope consisted 
of Cretaceous Rocks, it has been shown that these lay upon a floor 
consisting of the denuded ends of various strata which comprised rocks 
of the most diverse powers of resistance to subaerial denudation. These 
embraced representatives of nearly the whole of the Pre-Cretaceous rocks 
of Cumberland. In some few cases the efforts of the river to maintain 
its primitive course were more or less successful, and where that was 
the case the streams made their way across every rock, durable or not 
durable, that their channels happened to intersect. In many instances the 
rivers continued to cope with and to overcome all the difficulties, one 
after another, that arose in their way, and succeeded in maintaining 
their original courses without much change. In all cases of that kind 
the erosive power of the river in carving its own channel has exceeded 
that of ordinary atmospheric waste in lowering the surface adjoining the 
river channels. As a rule this is so; in other words, a river usually 
lowers the torrential part of its channel faster than subaerial erosion 
lowers any adjoining part of its basin. In a few cases the two processes 
go on side by side and at nearly equal rates. In some exceptional in- 
stances a part of the basin of a river adjacent to its channel may be 
lowered by subaerial erosion at a faster rate than the river lowers the 
adjoining part of the channel itself. Sooner or later this results in a 
diversion of the stream into the new course, which the river unavoidably 
follows as far as the new channel offers the easiest route seawards. 
This factor in the evolution of land surfaces has brought about 
many important changes in the initial direction of rivers in Cumberland, 
as elsewhere ; and is answerable for many of the inosculating valleys 
which characterize so much of the scenery. 
. Another case of a nature analogous to the last, and which has led 
to many important modifications of the river-courses in the Lake district, 
may next be considered. It has frequently happened that a sheet of soft 
rock laid down upon a floor of hard has been bent by earth movements 
into a dome. If we think of the inner mass as the core of the dome, it 
may help to simplify the description that follows. A river-course well 
established in the envelope of softer rock, cuts its way after a time down 
to the core. Where the difference in destructibility of the outer mantle 
is much greater than that of the core (as where a mantle of soft lime- 
stones enwraps a core of tough greywacke), the river gradually tends to 
wander away from its primary course, and as the extent of exposure of 
47 
