144 
J. HOPKINSON-WATER AND WATER-SUPPLY 
shown that water will rise in it by capillary action at least 16 feet. 
The roots of many trees and of some more lowly plants descend 
more than 3 feet through the soil, and their tips or spongioles are 
powerful absorbers of moisture. It has been shown that an oak 
tree will absorb during the whole period of its summer growth 
more than eight times the amount of water that falls on the area 
covered by it, the excess being drawn up by the roots from great 
depths.* It must, therefore, I think, be conceded that an appreciable 
quantity of the rain which falls will rise from a greater depth than 
three feet and be evaporated; how much we do not know, but if 
we take it at only a quarter of an inch,f it will leave an available 
supply for our underground reservoir of 6 inches per annum on 
the average of a long series of years. This is the amount the 
Commissioners on the Pollution of Rivers estimated in 1874 might 
be relied upon throughout the exposed chalk area in the river-basin 
of the Thames. But rainfall is exceedingly variable, and percola¬ 
tion is still more so, the results of experiments at Nash Mills and 
Lea Bridge showing that for six years in succession we may have 
to rely upon only 4f inches of percolation per annum, for five years 
upon 4£ inches, for four years upon 4 inches, and for three years 
upon 3 inches. 
This part of my subject has, I fear, already extended to undue 
length; but the erroneous conclusions which have been arrived at 
by neglecting it, or by lack of knowledge of the relation existing 
between rainfall, evaporation, and percolation (and also of the 
amount of available area for percolation), show that it is necessary 
to devote some attention to its consideration. For instance, Mr. 
S. C. Homersham, in a Report to the Directors of the London 
(Watford) Spring-Water Company, in the year 1850, made the 
following astounding statement: — -“ The area of land sloping 
towards Watford, and consisting for the most part of steep chalk 
hills,J embraces more than 1200 square miles, and calculating 
that only a depth of 20 inches of rain per annum falls in this locality, 
at least one-half, or 10 inches, may be considered as reaching the 
lower fissures. . . . This,” he added, “is equal to supply the im¬ 
mense quantity of 408 millions of gallons per day, for every day in 
the year, which at presents finds a vent and is discharged along the 
coast.” 
The fact is that the area of land sloping towards Watford scarcely 
embraces 150 square miles, of which not more than 100 are occupied 
by permeable beds,—either chalk or chalk capped by gravel,—and 
moreover with 20 or even 25 inches of rain not more than 5 inches 
would be likely to percolate into “the lower fissures,” so that instead 
of 408 millions of gallons per day, the amount would probably be 
* See ‘ Nature,’ vol. vii, p. 118. 
f This must be much under the mark. 
t In the same year, in a discussion on one of Mr. Clutterbuck’s papers (‘ Proc. 
Inst. Civ. Eng.,’ vol. ix, p. 162) Mr. Homersham spoke of hills “ from 800 to 
1000 feet in height around and sloping towards Watford.” These hills are 
barely half this height, and only from 100 to 300 feet higher than the valley of 
the Colne. 
