Geoscience Reference
In-Depth Information
The three elements, then, that are most likely to limit crop production are nitro-
gen N, phosphorus P and potassium K, and of these nitrogen is the key to high yields.
Most of the nitrogen in soils is fixed and cannot be used by plants. The nitrogen in
organic matter is also unavailable until it is converted to the inorganic nitrate (NO 3 - )
and ammonium (NH 4 + ) ions ( chapter 6 ). A hundred tonnes of organic matter per hec-
tare, accumulated under a grass ley, contains substantial quantities of N, about 200 kg
of which will be gradually released in the 2-3 years after ploughing. Primitive agri-
culturalists practised shifting cultivation as a way of cashing the store of N and other
nutrients accumulated under natural plant communities without re-investing anything;
they simply moved on when the soil became too poor to sustain further crops. Settled
agriculture, on the other hand, has to balance the nutrients extracted and lost from the
land with the rate of replenishment. Natural replenishment comes from the weather-
ing of rock particles and the input from rain. Nitrogen is also captured from the air
by soil algae and some bacteria, especially those associated with the root nodules of
legumes as described in chapter 3 . The use of clover was, therefore, an important step
forward. Farmyard manure (FYM) was carefully conserved and recycled, and gradu-
ally other organic and inorganic substances were introduced, such as ground bones,
guano and saltpetre.
By the early 19th century, the chemical composition of these 'natural' and 'artifi-
cial' fertilizers was known: saltpetre, for instance, is potassium nitrate (KNO 3 ); guano
from fish-eating birds contains mainly phosphorus and ammonium. The problem, in
the early days of scientific agriculture, was to derive unambiguous results from their
use. During the 1840s, various experiments were done with combinations of manures
and mineral salts on single plots, but the results were inconsistent and even contra-
dictory. J.S.Henslow at Cambridge, J.F.W.Johnston at Durham, and C.G.B.Daubeney
at Oxford all saw the need for organized experiments with single substances. Such
experiments, they felt, should also include the use of two untreated plots in order to
judge the variation due to natural causes - the beginnings of statistical design! If bar-
ley produced a heavier yield from one plot receiving substance X than from another
plot without it, could the difference be attributed to substance X?
In hindsight, it is easy to see the need for replication and for 'control' plots to
take account of other factors such as local variations in soil fertility, drainage, pests or
disease. At the time, such influences could only be guessed at, and it was J.B.Lawes
in 1847 who most clearly defined the problem for agriculture by asking “What sub-
stances is it necessary to supply to the soil in order to maintain a remunerative fer-
tility?” And it was he who, with J.H.Gilbert, decided “to make experiments at once
more systematic and on a larger scale on some of the most important crops of our
rotations.” Thus was laid down the most famous of all long term agricultural exper-
Search WWH ::




Custom Search