On the Rain and Drainage - Waters at Rothamsted. 
43 
All the plots mentioned in this Table receive the same quan- 
tity both of nitrogen and chlorine, but with different supplies of 
ash constituents. Where the principal ash constituents required 
by the crop are supplied (Plots 7, 17 or 18, 13) there a large 
assimilation of nitrogen takes place during the summer months, 
and the proportion of nitrogen to chlorine in the drainage-water 
becomes verv low. Where potash has never been applied 
(Plot 111, or not for many years (Plots 12, 14"), a larger propor- 
tion of nitric acid escapes assimilation. Where neither phos- 
phoric acid nor potash is applied (Plot 10), the proportion of 
nitric acid left untouched by the crop and removed in the 
drainage-water is much increased. In winter time the pro- 
portion of nitrogen to chlorine in the drainage-water is in all 
cases hiffh, the chlorides of the manure having: bv this time 
been washed out of the soil to a considerable extent, while a 
new formation of nitric acid is continually in progress. 
We must not leave the subject of the amount of nitrogen as 
nitrates present in the drainage-waters without referring to the 
quantities shown bv Voelcker's and Frankland's early analyses 
(Table XLV.). The amount of nitrogen per million shown 
for the unmanured Plot 3 & 4 is identical with the average 
found for this plot in the later Rothamsted determinations 
(Table XLVII.) : the amount found for Plot 5, receiving no 
nitrogenous manure, is also very similar to the later results ; 
all the remaining determinations are, however, much higher 
than those obtained during the last three vears. This difference 
is partly due to the extreme wetness of recent seasons, resulting 
in weak drainage-waters : but in the case of plots receiving 
ammonium- salts it is chiefly determined by the fact that, during 
the years to which the earlier analyses refer, the ammonium- 
salts were in everv case applied to the land in autumn. The 
results obtained by Voelcker and Frankland for Plots 7, 10, 11, 
12, 13, 14, 15, are thus, so far as time of sowing is concerned, 
comparable with those now yielded by Plot 15. 
The greater loss of nitric acid in spring and summer drain- 
age, where the ash constituents required bv the crop were 
deficient, is equally shown by the earlier analyses (Tables XLVI. 
and XL^ II.), as by our own later determinations. 
3. The Quantity of Xitkogex lost by Drainage. 
We have now discussed in considerable detail the very 
numerous results obtained relating to the composition of the 
drainage-waters collected from the differently manured plots in 
the experimental wheat-field. We have shown the influence of 
the amount and stage of the running, the description of the 
