On the Rain and Drainage - Waters at Rothamstcd. 317 
that up to the point at which the rainfall equals the evaporation 
of the season no drainage will take place, but thai bejond that 
point every inch of rain will produce an inch of drainage- 
water. This general statement assumes of course ordinary and 
not extraordinary conditions, and it is true of long periods 
rather than short ones ; in short periods the immediate distribu- 
tion of the rain will certainly have a preponderating effect. 
The cause of the immense excess of winter over summer drain- 
age becomes now apparent. The rainfall in summer is actually 
in excess of that in winter, but the drainage during the summer 
months is only half that experienced in winter, owing to the 
far greater amount of evaporation during the warmer half of the 
year. 
As to the various rates of evaporation from soils of different 
depths, the drain-gauges unfortunately give, as we have already 
seen, no certain information. The average evaporation from the 
20, 40, and 60-inch gauges are separately given at the foot of 
Table XXII. (p. 318) ; the figures we have discussed have been 
in all cases a mean of the three results. With soils more open 
than that at Rothamsted, of less water-holding capacity, and of 
lower capillary power, the amount of evaporation will be much 
less, and the proportion of the rainfall which appears as drainage 
far greater. The amount of evaporation in such soils is also 
more variable, the surface possessing no regular supply of w ater. 
The behaviour of soils of this description is shown in an ex- 
aggerated manner by Mr. Greaves's experiment on a mass of 
pure sand, the general results of which will be found in Table 
XXV. (p. 325). On the other hand, with soils of greater water- 
holding power, or with a heavier summer rainfall, the amount 
of evaporation during the summer months, and consequently 
for the whole year, might be somewhat increased ; but the 
possible range of variation in this direction is not large, as 
Mr. Greaves has shown that the average annual evaporation 
from a water surface is but 20'658 inches, or only about 3 inches 
greater than that obtained from the Rothamsted soil. 
To complete our view of the results yielded by the drain- 
gauges we must now proceed a step further, and see what has 
been the average amount of drainage and evaporation for each 
month in the year. 
It appears from Table XXII. that drainage has on an average 
commenced in earnest in September, and remained at a high 
point till March, that is, during the periods usually designated as 
"Autumn" and "Winter"; the maximum drainage being reached 
in November or January. From March to August (" Spring " 
and "Summer") the drainage has been comparatively small, 
the minimum having occurred in May. In considering these 
figures it must, however, be borne in mind that the rainfall 
