On the Rain and Drainage-Waters at Rothamsted. 47 
quantities of both were, upon the whole, very exceptionally large. 
The mean annual rainfall of the two seasons was 34"8 inches, or 
nearly 6^ inches more than the average of the thirty years. The 
mean annual drainage through the 60-inch soil-drain-gauge was 
more than 18 inches, or more than 8 inches above the average 
of eleven years. These conditions would obviously, on the 
other hand, tend to excessive loss by drainage. There were, 
therefore, during the period selected, exceptional circumstances 
operating in opposite directions. Still, as illustrations of loss 
bv drainage, the results given are extremely effective. 
Referring to the record of the quantity of drainage passing 
through the 60-inch drain-gauge during each period, as given at 
the bottom of the Table, it will be seen that in the first season, 
1879-80, there was a very large amount of drainage from spring 
sowing to harvest, that is during the active growth and ripening 
of the crop, and a comparatively small amount from the time 
of its removal to the commencement of active growth in the suc- 
ceeding spring. Accordingly we have even a much greater loss 
of nitrogen by drainage during the active growth and ripening of 
the crop than during the six or seven months afterwards. 
In the second season, 1880-1881, on the other hand, there 
was a very small amount of drainage during the growth and 
ripening of the crop, and a very excessive amount from the time 
of its removal to the commencement of growth in the following 
spring. Under these very different conditions the amounts of 
nitrogen lost by drainage within the two periods of the second 
season are very different from those of the first. There is now 
scarcely any loss from spring sowing to harvest, but very great 
loss from harvest to the next spring sowing. 
It should be stated, however, that as in the dry winter of 
1879-80 the drains only ran in February, and then gave water 
rich in nitrates, the result of accumulation, the estimates for that 
period, calculated on the composition of those waters alone, are 
doubtless above the truth. The losses during the very wet 
winter of 1880-81 are, on the other hand, probably under- 
estimated. 
Comparing the one whole season with the other there is not 
a very great difference in the amount of total loss in the two 
cases ; but it is, upon the whole, greater in the first year, with 
the excessive drainage soon after the application of the nitro- 
genous manures, notwithstanding the counteracting effect of the 
growth of the crop during that period, both upon the amount of 
drainage, and upon the amount of nitrogen as nitrates the soil 
would contain. 
The great loss of nitrogen as nitrates from the soil itself, in 
wet seasons, is strikingly shown in the results for Plot 36c4 con- 
