for Twcntij Years in succession on the same Land. 351 
4. Analysis of tlie drainage waters from the experimental 
wheat plots showed that they contained a large amount of nitrogen 
in the form of nitrates ; that the quantity of nitrates in the drain- 
age Avas the g-reater the greater the amount of ammonia-salts 
applied as manure; and that (after autumn sowing), the quantity 
was very much greater in the winter, than subsequently in the 
spring and summer. 
5. The analysis of the drainage waters further showed — that 
the winter drainage, after sowing ammonia-salts in the autumn, 
may often contain from two to three parts (and sometimes much 
more) of nitrogen (as nitrates and nitrites) per 100,000 parts of 
water. Calculation showed that, for every one part of nitrogen 
per 100,000 parts of drainage, there will be a loss of 2\ lbs. of 
nitrogen per acre for every inch of rain passing beyond the reach 
of the roots. In one case Dr. Frankland's analysis showed 
7"841 parts of nitrogen per 100,000 parts of drainage, corre- 
sponding to a loss of 17f lbs. of nitrogen per acre, provided an 
inch of rain passed as drainage of that strength. 
6. A given surface of soil possesses much less capacity of 
absorption for nitrate of soda, or its products of decomposition, 
than for the ammonia of ammonia-salts. Consequently, heavy 
rains soon after sowing would carry off in the drainage water 
more nitrogen from a dressing of nitrate of soda, than from a 
corresponding dressing of ammonia-salts. In one case, after a 
heavy dressing of nitrate of soda in the spring. Dr. Voelcker 
found the drainage-water to contain 5*83 parts of nitrogen per 
100,000 of water, corresponding to a loss of 13 lbs. of nitrogen 
per acre, per inch of rain so passing. 
7. Owing to the much less loss by drainage in the case of 
spring than of winter sowing, there was not only more increase 
in the immediate crop from a given amount of nitrogen 
applied in the spring for barley (or oats) than in the autumn 
for wheat, but there was also much more effect upon succeeding 
crops, from the at first unrecovered amount, in the case of the 
barley than in that of the wheat. 
8. It is probable tha't the whole of the nitrogen supplied as 
manure in ammonia-salts, or nitrate of soda, is either recovered 
in the immediate increase of crop, retained in the soil in a very 
slowly available condition, or drained away and lost. 
9. Owing to the slow decomposition of the nitrogenous 
organic matter of rape-cake and farmyard manure, their nitrogen 
is less rapidly available than that of ammonia-salts or nitrate 
of soda ; but, so far as can be judged from the direct experiments 
on the point, it would appar to be, at the same time, less subject 
to loss by drainage. 
10. Certain important mineral or ash-constituents of manures 
