been explained in several
ways. The oldest view is that of Decandolle, who founded his theory on
the fact that the plants excrete certain substances from their roots. He
found that when plants are grown in water, a peculiar matter is thrown
off by the roots; and he believed that this extrementitious substance is
eliminated _because_ it is injurious to the plant, and that, remaining
in the soil, it acts as a poison to those of the same species, and so
prevents the growth of another crop. But this excretion, though
poisonous to the plants from which it is excreted, he believed to be
nutritive to those of another species which is thus enabled to grow
luxuriantly where the others failed. Nothing can be more simple than
this explanation, and it was readily embraced at the time it was
propounded and considered fully satisfactory. But when more minutely
examined, it becomes apparent that the facts on which it is founded are
of a very uncertain character. Decandolle's observations regarding the
radical excretions of plants have not been confirmed by subsequent
observers. On the contrary, it has been shewn that though some plants,
when growing in water, do excrete a particular substance in small
quantity, nothing of the sort appears when they are grown in a siliceous
sand. And hence the inference is, that the peculiar excretion of plants
growing in water is to be viewed as the result of the abnormal method of
their growth rather than as a natural product of vegetation. But even
admitting the existence of these matters, it would be impossible to
accept the explanation founded upon them, because it is a familiar fact
that, on some soils, the repeated growth of particular crops is
perfectly possible, as, for instance, on the virgin soils of America,
from which many successive crops of wheat have been taken; and in these
cases the alleged excretion must have taken place without producing any
deleterious effect on the crop. Besides, it is in the last degree
improbable that these excretions, consisting of soluble organic matters,
should remain in the soil without undergoing decomposition, as all
similar substances do; and even if they did, we cannot, with our present
knowledge of the food of plants, admit the possibility of the direct
absorption of any organic substance whatever. Indeed, the idea of
radical excretions, as an explanation of the rotation of crops, must be
considered as being entirely abandoned.

The necessity for a rota