TOXIC EXTJDATES AND SOIL TOXINS. 159 



contains toxins, and that these sometimes result, directly or indirectly, from 

 the growth of higher plants. As to the manner in which these poison sub- 

 stances arise in the soil, no definite statement can yet be made, but they are 

 surely not generally excreted as such from the plant roots. That such poisons 

 are present in many soils has now been extablished without question by 

 Schreiner and his coworkers, and also that their deleterious effects may often 

 be removed by oxidation, or by the addition of proper substances." 



Pickering (1917) says: "But though their excretion from the roots 

 is possible, there is no need for imagining such an occurrence; all 

 plants in growing leave much root-detritus in the soil, and such de- 

 jecta may account for toxic properties just as well as ejecta." 



However, if he is right in ruling out deficient aeration and carbon 

 dioxid as causes, then both of these statements appear incorrect, 

 since it has been shown over and over again that under normal con- 

 ditions roots excrete no other toxic substances than carbon dioxid, 

 and the aerobic fermentation of plant material rarely produces toxins. 

 There seems to be no doubt that roots do not excrete other toxins 

 than carbon dioxid, except under anaerobic conditions, and the re- 

 sults drawn from cultures in solutions are either to be explained by 

 deficient aeration or by the limitations of the method itself, as indi- 

 cated by Stiles (1915), Hoagland (1919), Jordan (1920), and Davis 

 (1921). 



The statement of Bedford and Pickering (1914) that every growing 

 crop results in the formation of a substance toxic to the growth of 

 other plants, and still more so to itself, would seem to require that 

 the fruit trees of an orchard or the trees of a grove or forest should 

 be more toxic to each other than to grass or grass to them. However 

 this may be, the results of other investigators warrant Howard's 

 suggestion (1915 : 23) that carbon dioxid should not be finally dis- 

 missed as the toxin concerned, without repeating the Woburn experi- 

 ments dealing with this gas. He finds that the results obtained by 

 Bedford and Pickering with tobacco are exceedingly like those 

 observed at Pusa when tobacco is water-logged or grown on heavy 

 lands that have been green manured. Since tobacco requires a 

 great deal of air and green manuring produces much carbon dioxid 

 in the soil, it seems probable that the Woburn results, in which grass 

 washings injured tobacco, may be due after all to the inhibiting effect 

 of carbon dioxid. This may be the toxin about which so much is 

 written, and it may prove to be the cause of the effect cf grass on 

 trees, as well as of one crop on another. In many parts of England 

 grass is grown under fruit trees without particular damage, but in 

 most of these the soil is very porous, and the carbon dioxid diffuses 

 without doing harm. The soils at Woburn and Pusa are not porous, 

 and in such dense soils the effect of carbon dioxid should be far greater 

 than in the porous soils of Kent. 



Hole (1918 : 439) has also pointed out the probable significance of 

 defective aeration for the problem at Woburn. A dense growth of 



