48 On the Rain and Drainage- Waters at Rothamsted. 
tinuously unilianured, Plot 5 receiving annually for nearly thirty 
years mixed mineral manure alone, but mineral and nitrogenous 
manure previously, and Plot 16 unmanured for sixteen or seven- 
teen years, after excess of ammonium-salts with mixed mineral 
manure, previously. During the first twelve months the loss by 
drainage from these three plots was nearly 15, and 14:^ lbs. 
of nitrogen per acre ; and during the second twelve months it 
was 17J, 18^, and 18i lbs. per acre. Further, notwithstanding 
the comparative dryness of the autumn and winter period in 
the first season, and the very excessive drainage during the 
same period of the second season, nearly the whole of the loss 
is, in both cases, after the removal of the crop ; that is, during 
the long period of the year in which land under cereal culture 
is practically bare of vegetation. It may be here remarked, 
that the close approximation of the quantities estimated to be 
lost on Plots 5 and 16, with very similar manurial history, 
though the one is at one side and the other at the other side 
of the field, affords some evidence of the comparative character 
of the results in the different parts of the field. 
The loss of nitrogen estimated to be derived from that of the 
nitrogenous manures applied, is best studied by reference to the 
columns in the Table which show the excess in the drainage 
from the plots receiving nitrogenous manure over the amount 
from Plot 5 with mineral without nitrogenous manure. 
Comparing Plots 5, 6, 7 and 8, each receiving the same mixed 
mineral manure, but Plot 5 with no ammonium-salts, Plot 6 
with 200 lbs., Plot 7 with 400 lbs., and Plot 8 with 600 lbs. of 
ammonium-salts per acre per annum, there were, during the very 
wet period from spring sowing to harvest in 1879, losses of 8*56, 
16*75, and 23'39 lbs. of nitrogen more from the three ammonium 
plots than from Plot 5 ; that is, quantities increasing closely in 
proportion to the increased supply by manure. Then again, 
taking the series of plots with the same amount of ammonium- 
salts, but with different mineral manures for many years in 
succession, and yielding, accordingly, very different amounts of 
crop, there are very variable amounts of loss of nitrogen, ranging 
from 41*31 lbs. per acre with ammonium-salts alone (Plot 10), 
to only 16*75 lbs. with the same amount of ammonium-salts and 
the most complete mineral manure (Plot 7). The loss on the 
other plots of the series gradually increases with the defect of 
the mineral manure and the coincident defect of growth. Thus, 
with' superphosphate of lime and potassium-salts (Plot 13) it is 
17*45 lbs., or but little more than with the mixed mineral 
manure : with superphosphate and soda, with a residue of potash 
from previous applications (Plot 12), it is 19*67 lbs. ; with 
superphosphate and magnesia, and some residue of potash 
