On the Rain and Drainage - Waters at Rothamstcd. 315 
Excepting these extremes, the variation in eight years has been 
only from 11 023 to 12-535 inches. 
This comparative constancy of the evaporation under very 
different conditions of climate is certainly remarkable ; it must 
be greatly due to the fact that the two principal conditions 
which determine a large evaporation, namely excessive heat 
and abundant rain, very rarely occur together. In a wet season, 
when the soil is kept well supplied with water, there is at the 
same time a more or less saturated atmosphere, with an absence 
of sunshine, conditions unfavourable to a considerable evapora- 
tion. In a hot season there is, on the other hand, usually a 
scarcity of rain ; and after the surface of the soil has dried 
evaporation must proceed very slowly. 
The figures representing the evaporation from the soil during 
the winter half of the year are by no means so regular as those 
relating to the summer half, or to the whole year. This is 
partly due to the small amount of the evaporation, which is 
naturally, therefore, considerably affected by any disturbing 
cause. A nearer study of the results shows, however, that the 
evaporation is considerably more constant than at first appears. 
Both in 1870-1 and in 1872-3 the high calculated figure for 
evaporation is certainly in excess of the truth ; in each case the 
soil was unusually dry just before the commencement of the 
winter period, and in the latter winter it was also some- 
what unusually wet at the end of this period : rain has thus 
been reckoned as evaporation which really was simply retained 
by the soil. Errors of reckoning of this description are natu- 
rally greater in the case of the deeper drain-gauges, and least in 
the case of the 20-inch gauge. With this gauge the range of 
the evaporation calculated lor the winter months is much smaller 
than with the other gauges ; omitting 1878-9 and 1879-80, the 
range with this gauge is but from 5*182 to 6-924 inches. 
The winters of 1878-9 and 1879-80 show a remarkably small 
calculated evaporation, and considerably diminish the total 
evaporation of the years of which they form part. The 
winter of 1879-80 is the most striking of these exceptions ; it 
succeeded the wettest summer in the whole ten years, while the 
winter itself was the driest and coldest in the same period. The 
rainfall of October and January being insufficient to provide for 
a normal evaporation, and the low temperature tending also to 
a reduced rate of evaporation, we should expect the amount of 
evaporation during this winter to be somewhat below the average. 
A more potent cause of the extremely low figure found by cal- 
culation is, however, the abnormally high drainage. The soil 
commenced October in a saturated condition, while the winter 
concluded with a fairly dry March ; a part of the winter drain- 
age thus belonged to the summer rainfall — a most unusual cir- 
