106 
The leading features connected with the mode of occurrence of 
our lakes may be stated thus. They occur in greatest force in the 
most mountainous parts, and they usually occur in such a manner 
in relation to the stream that enters their upper end and emerges 
again at their point of outflow, as to suggest that they represent 
little else than a local dilation, or enlargement, of the banks of the 
river itself. I would refer, as familiar illustrations of this feature, 
to the course of the Eamont in relation to Ullswater; to that of 
the Derwent in relation to Derwentwater and Bassenthwaite; to 
Wyburn Beck in relation to Thirlmere; the Liza to Ennerdale 
Water, and to many other instances of the same general nature 
that will immediately occur to everyone in the least degree familiar 
with the physical geography of Cumberland and Westmorland. 
_In each and every one of these a little reflection will convince 
any one that the occurrence of the lake is but a temporary and 
accidental condition of the course of the river, brought about by 
exceptional events at a late period in the history of the valley where 
they occur. 
Another point that will be noticed is that there is a marked 
tendency in all grouped lakes to occur in chains, separated either 
by ordinary narrows, such as close in upon the sides of Thirlmere 
near Armboth, or else by long stretches of alluvial land, such as 
separate Buttermere from Crummock Water, or Derwentwater from 
Bassenthwaite. It will also be noticed in many cases, and it can 
be proved by soundings in many more, that the lowest parts of the 
bed of each lake are several feet below the rock occurring in the 
bed of the river at or near the spot where the lower end of the 
lake merges into the river. In other words, a line drawn from the 
lowest part of almost any one of our lakes to the first mass of solid 
rock that occupies the bed of the river at the outlet of the lake, 
would be found to run uphill, instead of downhill, as is the case in 
the bed of a river in its normal condition. Very commonly, also, 
this rock barrier at the foot of the lake rises on each side of the 
outlet in such a way as to suggest that it once extended across the 
outlet of the lake at a much higher level, which level has since 
been lowered by the erosive action of the river. This rock barrier 
