for Ttoenty Years in succession on the same Land. 347 
To the foreffoing: illustrations of the effects of the unexhausted 
residue from previously supplied nitrogen, must be added sfjme 
evidence as to the effects on succeeding crops of previously sup- 
plied mineral manures, or ash-constituents. The experiments 
on barley do not furnish absolutely unexceptionable comparative 
evidence on the point ; though there can be little doubt that the 
superphosphate and sulphate of potass applied in the first year, 
1852, on Plots 1 N and 2 N, have materially increased the effects 
of the nitrate of soda afterwards annually applied up to the 
present time. The experiments on wheat do, however, afford 
very conclusive evidence on the subject, and as we are now able 
to give the results of eight more seasons than when writing on 
the question in 1864, we append the following Table (pp. 
348-9) relating to that crop. 
For the crop of 1844, both plots, 10a and 10^, received a 
mineral manure, consisting of silicate of potass and superphos- 
phate of lime. Every year since, 10a has been manured with 
ammonia-salts alone. 106 has been manured exactly simi- 
larly in every year excepting the third, fifth, and seventh (1846, 
1848, and 1850); in 1846 it was left unmanured ; in 184H 
it received, in addition to the ammonia-salts, a mineral manure 
containing salts of potass, soda, and magnesia, and superphos- 
phate of lime ; and in 1850 the same mineral manure without 
the ammonia-salts. That is to say, during the first six years of 
the twenty-seven, the application of ammonia-salts was twice 
omitted on lOi, but it twice received mineral manure when 
10a did not. 
The Table shows that during the 6 years, 1845-50, \0h, with 
less ammonia-salts, but more mineral manure, yielded, in the 
aggregate, 14|- bushels less corn, and 11|- cwts. less straw, 
or 2^ bushels corn, and 1^ cwt. straw, less per acre per 
annum than 10a. On the other hand, in almost every year 
since up to the present time, a period of 21 years since the last 
application of mineral manure, 10Z» has yielded more of both 
corn and straw than 10a ; in all 69^ bushels more corn, and 
61^ cwts. more straw, or an average annual excess of 3^ bushels 
of corn, and cwts. of straw. 
It is obvious that the excess of produce on lOZ*, over that on 
10a, during the last 21 years, may be partly due to the less 
exhaustion of the mineral constituents of the soil on lOb during 
the first 6 of the 27 years, owing to the less supply of ammonia- 
salts to it during that period. But, if we deduct the difference 
between the produce on the two plots during these 6 years, 
from the excess of produce on ]0/> during the last 21 years, we 
still have, during the latter period, an aggregate excess of 54f- 
bushels of corn, and 50|- cwts, of straw, or an average annual 
excess of 2i bushels of corn, and 2^ cwts. of straw, on IQb^ 
