108 
streams carrying material to them, whereby the transporting power 
of those streams is reduced almost to 77. 
The silting-up process proceeds with more or less rapidity, in 
proportion to the rate of waste of the drainage area above it. In 
some cases this goes on at a very slow rate, when measured by 
years, as can be shown by the existence of Roman stations on 
some of these old alluvial flats, which are even yet at no great 
distance from the head of the lake. That at the head of Winder- 
mere is an excellent example. If one may judge by this instance, 
it is clear that the Roman invaders of Britain must have seen our 
lakes in very much the same state as we see them at the present 
day ; so imperceptible is the rate of sedimentation in many such 
instances. 
In other cases, the easily-eroded nature of the barrier that 
dammed back the waters of the lakes, has caused the water to be 
drained off before much sediment had time to accumulate. An 
instance of this kind must have occurred in the neighbourhood of 
Threlkeld, where a rock barrier, consisting of Skidda Slates (a 
rock easily eroded), once extended in a south-westerly direction 
right across the Greta to the west of the Penrith-Keswick road. 
This barrier must have shut in a considerable hollow, that certainly 
contained a lake, before the Greta worked its way down through 
the confining ridge to its present level. Threlkeld stands only a 
little above one of the former shores of this old lake. 
In other cases the rate of lowering of the rock barrier has been 
sufficiently slow to admit of the accumulation of a considerable 
mass of sediment, especially in such cases as those where the 
drainage, and, consequently, the subaerial waste, of a large area is 
received by the stream. What I am aiming to show is, that every 
gradation may be traced between a perfect, unmodified lake, with 
little or no alluvium, and the wide-spread holms that (except during 
heavy floods) have nearly or quite ceased to resume the character 
of a lake at all. But, whatever may be the present condition of the 
lake, the features I have just referred to as the essential accompani- 
ments of the lakes are usually traceable. Every intermediate 
gradation may thus be traced between some of our largest sheets 
