60 
On the Rain and Drainage - Waters at Rothamstcd. 
30 years of continuous experiment ; the amount of nitrogen in 
the surface-soil of Plot 5 without ammonium-salt is lower in 
1881 than in 18G5 ; and the difference in the amount on the 
different ammonium plots, compared with Plot 5, is, in most 
cases, approximately, twice as great after the 30 as after the 
14 years. The order of excess of nitrogen on the different plots 
is again, in 1881, almost parallel with that of the increased 
yield of nitrogen in the crops, and it is almost the converse to 
that of the amount estimated to be lost by drainage. 
From these facts it is obviously to be concluded that the 
relative excess of nitrogen in the soil of the different plots re- 
ceiving nitrogen in manure, is much more closely connected 
with the amount of growth, than with direct retention of the 
nitrogen of the manure. In other words, the difference is 
mainly due to the residue of the crops — to the stubble and 
roots, and perhaps to weeds. 
Further, the excess on the ammonia plots compared with 
Plot 5, is much more due to the reduction in the nitrogen of 
Plot 5 than to any increase on the ammonia plots ; for the 
percentage in these is very nearly the same at the two periods, 
being, as a rule, slightly higher in 1881 than in 1865 on the 
plots of the best growth, and slightly lower on those of deficient 
growth. The indication is, therefore, not that the ammonia- 
plots have gained in the degree which the excess in the amounts 
over Plot 5 would show ; but only that the loss which the soil 
itself has sustained in all cases, has been more or less com- 
pletely compensated by the retention of nitrogenous crop residue, 
which has taken place much in proportion to the amount o4' 
crop grown and removed. 
It is obvious that we must suppose a. certain amount of 
nitrogen has been supplied to the crop, or to the drainage, of 
to both, from the soil itself. Whether this amount is greater 
or less where there is a liberal supply of nitrogen by manure 
than where there is not, the data do not enable us to determine 
with any certainty. On this point it may be stated that, taking 
the average of the whole 30 years, the estimates show an 
annual yield of nitrogen on the unmanured plot of 18"6 lbs. in 
the crop, and 10-3 lbs. in the drainage, or in all 28-9 lbs. per 
acre per annum. In like manner Plot 5, receiving nitrogenous 
manure during the first 8 years, but mineral manure only 
during the last 30 years, has yielded 20'3 lbs. in the crop, and 
12 lbs. in the drainage, or in all 32-3 lbs. per acre per annum. 
So far as can be estimated, it would appear that the soil of each 
of these two plots has on the average of 30 years lost about 
two-thirds of these amounts annually, to the depth of 27 inches. 
There would remain, therefore, about one-third — say 10 lbs., 
