142 
J. HOPKIXSOX-WATER ASTB WATER-SUPPLY 
that there is always a breeze.”^ Again, the altered position of 
Mr. Griffith’s gauge, from open country at Strathfield Turgiss to a 
garden partially sheltered from the wind by houses, and with the 
less clear atmosphere of London, fully accounts for the smaller 
amount of evaporation arrived at by Mr. Symons. If we assume, 
and I think we may do so, that one inch, at least, of the lessened 
evaporation is due to the difference in the clearness of the atmo¬ 
sphere, we have the following results for the South of England with 
an average rainfall of about 25£ inches :—Annual evaporation from 
Stillwater on land in a sheltered situation, 15^-inches; from the 
same in the open country, 18 inches; from still water on the surface 
of a flowing stream, 20£ inches; while from a flowing river itself 
the evaporation is probably rather greater than this. 
These figures will, I think, represent approximately the evapora¬ 
tion from our sheltered ponds, our lakes and reservoirs, and our 
flowing streams. 
The average evaporation from the surface of the ground can only 
be deduced from the difference between the rainfall and the perco¬ 
lation. Erom observations of a percolation-gauge -roVo of an acre 
in area and 3 feet in depth, at E-othamsted, during the 20 years 
September, 1870, to August, 1890, Messrs. Lawes and Gilbert 
determined the average annual percolation to be 13*61 inches out 
of a rainfall of 30 29 inches, showing an evaporation of 16*68 inches, 
the surface of the ground being kept free from vegetation. At Lea 
Bridge, during the 22 years Oct. 1851 to Sept. 1873, Mr. Greaves 
determined the average annual percolation through a gauge 3 
feet square and 3 feet deep to be 6’85 inches out of a rainfall of 
25-76 inches, leaving 18-91 inches to be accounted for by evapora¬ 
tion and absorption by the grass growing on the surface. At Nash 
Mills during the 44 years, October, 1840, to September, 1884, 
Messrs. Dickinson and Co.’s percolation-gauge, f 18 inches in 
diameter and 3 feet in depth, gave an average annual percola¬ 
tion of 6-94 inches out of 28-18 inches of rain, 21*24 inches being 
evaporated and absorbed by the grass growing on the surface. 
In each instance I give the results only of the percolation through 
gauges filled with the ordinary soil of the district. Through chalk 
and sand the percolation is greater. 
Taking into consideration the difference in the rainfall, in the 
situation of the gauges, and in their construction, the results do not, 
I think, show any greater divergence than should be expected.J 
Much depends upon how the rain falls. A heavy fall after a dry 
period will not percolate more than a few inches, and if another dry 
period succeed, the moisture will be rapidly drawn up to the surface 
and evaporated. On the other hand, a continued drizzling rain, 
* ‘ British Rainfall,’ 1889, p. 40. 
f Still in operation, hut no more-recent results have been published. 
+ The results of experiments on evaporation at Strathfield Turgiss and Camden 
Square are taken from ‘ British Rainfall,’ 1886-89, and those of experiments on 
evaporation and percolation at Lea Bridge, and on percolation at Nash Mills 
and Rothamsted, from several papers and discussions in the ‘ Proceedings of the 
Institution of Civil Engineers,’ 1843-1891. 
