316 On the Rain and Drainage - Waters at Rothamsted. 
cumstance. The very exceptional high rate of drainage during 
January and February cannot, however, be explained by refer- 
ence to the previous summer's rainfall, and certainly points to 
a condensation of water by the soil, probably occurring at the 
close of the severe frost experienced during those months. 
The winter of 1878-9 was not quite so exceptional as the one 
last mentioned. It followed a wet summer, and concluded with 
a dry March, the temperature was also nearly as low as in the 
winter of 1879-80 ; some of the circumstances tending to pro- 
duce a low 'evaporation and high drainage were thus the same in 
both seasons, though not so marked in 1878-9 as in 1879-80. 
The evidence of condensation of water by the soil during January 
and February was, however, still more distinct during the winter 
now under consideration, the drainage of January and February 
being far above the normal proportion to the rainfall. 
We conclude, therefore, that these winters had probably a 
rate of evaporation rather below the average ; but the serious 
deficiency shown by the figures in the table had probably no 
existence, being simply due to a special increase of the drainage 
from sources independent of the rainfall of the period. 
We now turn to the average amounts of evaporation for each 
season given at the foot of Table XXI. The rate of evapo- 
ration having altered, on the whole, within moderate limits 
during the last ten years, these figures will express with tolerable 
accuracy the amount of evaporation which will ordinarily take 
place from the Rothamsted soil when kept bare of vegetation, in 
a climate having a mean temperature of about 48°. The average 
amount of evaporation during the six summer months will be 
nearly 12 inches, during the six winter months about 5^ inches,* 
and for the whole year 17 to 18 inches. 
We have dwelt thus at length on the amount of evaporation 
from the soil of the drain-gauges because the comparative 
constancy of the evaporation from the surface of a bare clay 
soil in seasons of very different character is probably a novel 
fact, and becau'se this constancy in the amount of evaporation is, 
in the case of our drain-gauges, the law which determines in 
the long run the amount of drainage. Drainage is in fact 
merely the excess of rainfall over evaporation. With this view 
of drainage in our minds, the cause of the great variation in its 
amount becomes readily understood ; for we may almost say 
* The average evaporation from a water surface during the six winter months 
is according to Mr. Greaves only 4 -706 inches. Are we to conclude that under 
some circumstances (he evaporation from a bare soil may he greater than from 
water ? Or that Mr. Grcaves's result is diminished by a condensation of water 
from the atmosjihere greater than is experienced at Rothamsted? Or, on the 
other hand, is the evaporation calculated for our soil too high, from the errors 
known to affect the reckonings for the autumn months (see page 319) ? It is 
impossible at present to give a definite answer to these questions. 
