148 
J. HOPKINSON—WATER AND WATER-SUPPLY 
means the plane of saturation is lowered, and the Chalk is gradually 
dried throughout its whole extent. At each deep boring from 
which a large quantity of water is being pumped, this plane forms 
ah inverted cone which gradually enlarges, and in course of time 
the present sources of our rivers must be affected by the water 
flowing underground to supply the void thus caused, instead of over 
the surface. Every million gallons of water thus drawn from the 
chalk carries with it in solution, on the average, a ton and a quarter 
of the chalk through which it has percolated, giving an additional 
storage-room for 110 gallons of water. This accounts for the yield 
of a well increasing for a certain time until the maximum of 
permeability is reached, but it implies a more rapid lowering of the 
plane of saturation, and thus increases the evil effects of exhaustive 
pumping. 
That there are channels in the Chalk through which water flows 
with comparative freedom, apart from the more or less vertical 
“pipes” and slightly-inclined layers of flints, does not admit of 
doubt. They are frequently met with in well-sinking, etc., when 
tapped the water in them flooding the works, and the determination 
of the probable direction in which they convey the water is of much 
importance. In most cases we cannot ascertain it, but I think we 
have some grounds for inferring the usual direction of such channels 
near the surface. The plane of saturation conforms pretty nearly 
to the form of the ground, sloping down under our hills transversely 
to the rivers which flow between them, but at a less inclination 
than that of the surface of the ground. At the bottom of the 
valleys in the Chalk, when rivers flow through them, the longi¬ 
tudinal inclination must be as great as or greater than that of 
the rivers. The general direction of the flow of water underground 
is therefore the same as that above ground, for the water must flow 
from the higher to the lower level. This is well shown in bournes, 
or streams which only flow after very wet winters, the water in 
them sometimes disappearing for some distance, and re-appearing 
with undiminished flow. Even in the River Colne I have seen a 
considerable volume of water entirely disappear, and re-appear lower 
down the valley. Such cases show that there must be underground 
channels which follow the general course of the streams on the 
surface. This, we may infer, is most probably the usual direction 
in which water flows immediately below the surface of the ground, 
but this flow is not from the bottom but from near the top of our 
underground reservoir, and therefore is rather an underground 
overflow than a leakage, and could not unduly drain it. 
But, as it has already been pointed out, there must also be a 
considerable flow of water in the layers of flints in the Upper Chalk, 
and in Hertfordshire the inclination of these layers is from north¬ 
west to south-east at a comparatively high (though really low) angle 
in the north-west, and gradually approaching horizontality in the 
south-east. Part of the catchment-basin of the Colne is directly 
north-west of the lower portion of the valley of the Lea north of 
London, and therefore much of the rain which falls within the 
