184 Pickering . — The Effect of one Plant on Another. 
need for imagining such an occurrence : all plants in growing leave much 
root-detritus in the soil, and such disjecta may account for toxic properties, 
just as well as ejecta. That some forms of organic matter may be highly 
poisonous to plant- life has been established : soil which has been heated to 
125 0 is very toxic, and there is evidence that toxicity may be produced in it 
by heating to much lower temperatures (‘ Journal of Agricultural Science,’ iii. 
277). In the case of heated soils, the chemical changes are complicated by 
changes in bacterial character ; this may or may not be the case with soil 
which is growing crops ; but there is one feature in common between the 
two, namely that in both cases the toxin is easily oxidized, and that after 
oxidation, it acts as a plant nutrient, increasing the fertility of the soil. 
When in pot experiments such as those mentioned, the teachings from the 
crop in the trays are kept exposed to the air for about twenty-four hours 
before being given to the plant, their toxic property is found to have entirely 
disappeared, and in some cases, indeed, they act beneficially : even a two-inch 
layer of pumice-stone interposed between the tray and the earth in the pot 
will admit of sufficient oxidation for a reduction in the toxic effect to be 
discernible. 
A reversal of the effect of grass may also be recognized in field 
experiments, for in a case where apple-trees were planted and kept clear 
from grass to a distance of 3 feet from the stems, the trees flourished better at 
first than those without any grass near them (and, of course, much better 
than those which had been entirely grassed). But as they grew, and 
their roots approached the grassed ground, the toxin affected them before it 
had time to become oxidized, and they began to suffer. Though there 
are other reasons why land under grass gradually becomes more fertile, the 
accumulation of the oxidized products of the toxin must constitute an 
important factor in this enrichment. In certain experiments with apple- 
trees it was found that soil which had been under grass for ten years induced 
double as much growth as similar soil which had been under tillage, though, 
when the turf was replaced on the soil, the trees showed all the bad effects 
of grass. 
From the general character of the action of one crop on another, 
it follows that the tables may be turned on the grass, and, even in pot 
experiments, it has been proved that grass in the pots will be adversely 
affected by apple seedlings in the trays. In practice, of course, it is 
known that grass and other surface crops are adversely affected by trees. 
This is generally attributed to the shading effect, and to the robbing of the 
soil of its nourishment. Doubtless the shading produces bad results in 
many cases, but it may be questioned whether any serious robbing of the 
soil occurs, for there is good evidence for believing that ground under trees, 
even when worked regularly for timber, increases in fertility, just as does 
ground under grass. At any rate, it has been found that a surface crop 
