F. FLETCHER. ES 
The amount of substance given out by the roots is not 
inconsiderable. For instance, the precipitate obtained by 
adding potassium sulphate to a solution containing the excre- 
ment of 10 cotton plants growing until their combined air-dry 
weight was ‘4 gramme, weighed, when dry, ‘21 gramme. 
Sesamum in its early stages of growth, appears actually to 
excrete a greater amount of material than it builds up in its 
own substance. 
ConcLuston. 
The bearing of the phenomena described in this article on 
the question of rotation of crops is obvious. 
The question may, however, be put why cotton, for instance, 
which grows so feebly near sorghum (Table IV) grows at least as 
well ifnot better, after sorghum than after cotton. From experi- 
ments now in progress it appears that this is explicable as follows :— 
When cotton is growing near sorghum the roots of the 
latter exude the toxic substance into the soil in large quantities. 
This spreads rapidly through the soil into the subsoil especially 
during the rainy season, and neighbouring cotton plants are not 
protected by the fact that their tap roots go down far below the 
zone in which the sorghum roots are situated. When cotton 
follows sorghum, however, the condition of affairs is different ; the 
toxic substance remaining, at the time of harvesting, in the roots 
of the previous sorghum crop is now being given out slowly in 
the course of the decay of those roots,* and is held entangled in 
the organic matter of the roots, largely in the zone of soil in 
which the roots of sorghum spread. Each crop thus fouls the 
soil for a crop of the same variety, whose roots will take the 
same course as the previous crop, more than for a crop whose 
roots spread in another layer of the soil. 
The precipitation of the toxic substance by most of the 
mineral manures in common use indicates the manner in which 
many manures act in increasing crop yields. 
* That the roots of sorghum and other crops exert an extraordinarily toxic effect when 
mixed with soil in which plants are then grown has been proved by the writer in a set of pot 
experiments. 
