310 SCIENCE AND FRUIT GROWING 



side by side, half of each of which was grassed over some years 

 ago; but whilst the grass has killed many of the trees in one 

 of these, it has not very much affected those in the other. This 

 latter plantation occupies one of the best-drained positions in the 

 field, the former one of the worst-drained ones (p. 262). 



Apart from these considerations there are others which may 

 account for differences in the grass-effect in different localities : 

 the nature of the toxin is not yet known, and still less do we 

 know how far its formation and subsequent decomposition may 

 be effected by the chemical and physical nature of the soil : it 

 is possible, too, that trees may, in cases where they are not too 

 much enfeebled at first by the action of the grass, be able 

 to adapt themselves more or less to their altered environment 

 by means other than that of extending their roots downwards. 

 At present, however, this is all speculation, and the formation 

 of a toxin which rapidly becomes converted into a food would 

 appear to supply all the conditions necessary for explaining the 

 existence of great differences in the final results in different 

 localities. 



The problem as to the nature of, and as to the method of 

 formation and decomposition of this toxin is still to be solved. 

 Much of the opposition to the idea of a toxin being formed 

 during plant-growth is due to the fables connected with the 

 subject (see p. 302), and to a misconception of what is necessarily 

 implied by the formation of a toxin. It does not imply that 

 the roots must excrete some substance which acts as a virulent 

 poison to other plants : it is true that instances of excretions 

 from flowers, leaves, stems and seeds of plants, either under 

 normal or abnormal conditions, are known, and hence the 

 possibility of there being excretions from the roots cannot be 

 overlooked : but the view that such occurs in the present case 

 is by no means necessary, for we have an ample explanation of 

 the results in the fact that plants in growing leave much organic 

 detritus in the soil (see p. 2), and that such organic matter in 

 its decomposition under natural conditions forms, first a toxin, 

 and then a plant- food. The importance of the experiments on the 

 action of heated soils on plant-growth and on seed-germination 

 (p. 249), as bearing on this point, cannot be overrated. 



Whether the toxic substance is in a soluble condition to start 

 with, or is rendered soluble only by subsequent chemical change 

 or by bacterial agency, is not yet known, nor is it known how 

 far its subsequent conversion into plant-food is due to spon- 

 taneous decomposition or to oxidation. It is probable, however, 



