for Twenty Years in succession on the same Land. 3C9 
manured with ammonia-salts in the autumn. Owing, however, 
to the much less capacity of a given surface of soil for the 
absorption of nitrate of soda, or other nitrates arising from its 
decomposition, than of the ammonia of ammonia-salts, heavy 
rains, soon after sowing, would carry off more of the nitrogen from 
nitrate of soda than from a corresponding dressing of ammonia- 
salts. In one case Dr. Voelcker found, in the drainage collected 
from the nitrated plot soon after a dressing of 550 lbs. of nitrate per 
acre ( = 400 lbs. ammonia-salts), applied in the spring, 5 ■ 83 parts 
of nitrogen per 100,000 parts of water, corresponding to a loss 
of about 13 lbs. of nitrogen per acre per inch of rain passing. 
These facts, showing how great may be the loss of the 
nitrogen of manure by drainage, are obviously of the greatest 
practical importance, and demand very serious consideration. 
Owing to the difficulty of determining with certainty, either 
the total amount of nitrogen retained by the soil within the reach 
of the roots, the proportion of the total rain which would pass 
beyond the reach of the roots, or the average composition of the 
drainage-water, absolute proof whether the whole of the supplied 
nitrogen which is not recovered in the crop is either retained by 
the soil, or lost by drainage, is not at command. Still, a 
consideration of such data as are available in reference to the 
points here indicated, points to the conclusion that the whole 
of the nitrogen which was applied as ammonia-salts or nitrate of 
soda to the wheat was either recovered in the increase of crop, 
accumulated within the soil, or lost hy drainage. 
As already said, as the proportion of the nitrogen of ammonia- 
salts Avhich was recovered in the increase of produce was much 
greater in the experiments with barley than in those with wheat, 
there remained of course much less, in its case, to be accounted 
for by accumulation in the soil, and by drainage. 
Only few determinations of nitrogen have as yet been made 
in the soils of the barley plots ; but, so far as can be judged from 
the results obtained hitherto, it seems probable that there is less 
accumulation than in the case of the wheat, especially in the 
lower layers. It seems pretty certain, too, that there must be 
much less loss by drainage ; but, as the experimental barley-field 
is not artificially drained, no direct evidence can be adduced on 
the point. It may be observed, however, that as the ammonia- 
salts are sown for the barley in the spring, when the soil is in a 
porous condition, when there is comparatively little risk of wash- 
ing out, and when growth almost immediately succeeds, there 
will be a less immediate and wide distribution of the ammonia, 
or of the nitrate resulting from its oxidation, a larger proportion 
at once taken up by the growing crop, and, probably, a larger 
proportion fixed near the surface before the winter-rains, and 
remaining available there for succeeding crops. 
