44 On the Rain and Drainage - Waters at Rothamsted. 
manure, and the time at which it is sown, the period of the 
year, and the character of the seasons, on the composition of 
the waters. 
Confining attention to the loss of nitrogen by drainage 
which has been indicated, it has been seen that its amount has 
been very directly connected with the amount supplied in the 
manure. The practical question obviously suggests itself — 
whether, in the case of the experimental wheat-field, in which 
known quantities of nitrogenous manure ha ve been applied, and 
known quantities of nitrogen removed in the crops, for many 
years in succession, the facts at command are sufficient to enable 
us to estimate how much nitrogen has been lost by the drainage 
from the different plots?- — and whether the whole of the nitrogen 
of the manure which has not been recovered in the increase of 
crop, may be accounted for by the ascertained loss by drainage ? 
It is obvious that, to be able to give an exact answer to 
this question, it is essential to know, not only the total amount 
of drainage which has passed from the land, and the amount of 
nitrogen it has carried with it, but also how much nitrogen has 
been supplied to the soil, or the crop, from the atmosphere by 
rain or condensation, and how much has been yielded to the 
crop, or to the drainage, by the soil itself? — or whether, on the 
other hand, some of the unrecovered amount is retained by the soil 
or subsoil, possibly to be slowly recovered in succeeding crops ? 
Unfortunately, we have no means of gauging the total amount 
of drainage passing from the land of the experimental wheat- 
field. We can only form some judgment of it from the quan- 
tities determined in the case of the 20-inch, 40-inch, and GO-inch 
soil-drain-gauges, the results obtained by which during between 
ten and eleven successive seasons have been fully described in 
Part II. of this Paper. It is assumed that the results of the 
GO-inch drain-gauge will probably afford the best basis for esti- 
mating the amount of drainage in the experimental wheat-field. 
It would seem probable that during the late autumn, the winter, 
and the early spring, the amount of drainage would not differ 
widely in the two cases ; but that, during the active growth of the 
crop, and for some time afterwards, the loss would be less in the 
experimental wheat-field than through the drain-gauge, owing 
to the drying of the soil under the influence of the growing crop. 
Using such data as we possess, we propose to give, in the 
first place, an estimate of the quantity of nitrogen lost by 
drainage from most of the plots of the experimental wheat-field, 
during two recent years, for which we have analyses of every 
running from the drain-pipes. We shall afterwards attempt to 
estimate the loss for a much longer period in the history of the 
experimental field. 
