TOXIC EXUDATES AND SOIL TOXINS. 149 



toxin is rapidly oxidized by the action of air and moisture, and is destroyed 

 under cultivation in a few weeks. In soil which has been heated, therefore, 

 plants will not thrive at once, especially if the supply of air is restricted, 

 though after a time they grow better in it than in soil which has not been 

 heated at all. Thus plants may behave in diametrically opposite ways in 

 heated soil, according to the conditions under which they are grown. This 

 has been found to be the case with apple trees, as well as with grasses and 

 other plants. 



"The toxic substance produced by heating soils was found to be toxic 

 toward the germination of seeds as well as toward the growth of plants, 

 retarding the germination and reducing the percentage of seeds which germi- 

 nate. In extreme cases seeds may take five or six times as long to germinate 

 in heated as in unheated soil. As experiments on seed-germination can be 

 carried out in a day or two, whereas those on plant-growth require many 

 weeks, during which the character of the soil may become materially altered, 

 the former offered a promising means for searching for the presence of toxic 

 matter in grassed soil. A considerable number of instances were taken in 

 which grassed and tilled soils within a few feet of each other were examined 

 as to their behaviour toward germinating seeds, and the examination was 

 conducted at three different seasons in the year; but the results in every case 

 showed, contrary to expectation, that the soil from the grassed ground was 

 slightly more favourable toward germination than the tilled soil. These 

 results, of course, afford no direct evidence in favour of the presence of a 

 toxic substance in grassed soils, though they are quite consistent with such 

 a view, for a toxic substance, if present, might, just as in the case of heated 

 soil, give rise, on decomposition, to conditions specially favourable toward 

 germination. It was noticed also that in most cases the soil which had been 

 under grass absorbed water much less readily than the neighbouring tilled 

 soil, a behaviour which is highly suggestive, inasmuch as the same character 

 is observed in heated soils, in contrast with unheated ones. 



"Strong evidence of a positive character as to the formation of a toxic 

 substance during the growth of grass was finally obtained from various 

 series of experiments with trees grown in pots. It was found that such 

 trees, when watered with the leachings obtained from trays containing grasss 

 growing in sand, flourished more than when water alone was supplied; but 

 when the trays were placed on the surface of the soil (or sand) in which the 

 trees were growing, so that the washings from the grass reached the tree- 

 roots with practically no exposure to the air, they then had a very deleterious 

 effect, nearly, if not quite, as great as when the grass was grown above the 

 roots of the trees in the ordinary way. The trays containing the grass were 

 movable, and the sand in them, with the. grass growing in it, was separated 

 from the medium in which the trees were growing by the perforated iron 

 bottoms of the trays and a sheet of wire gauze; moreover, the contact between 

 the bottoms of the trays and the sand or soil beneath would be, at the best, 

 very imperfect, so that it is impossible to explain the action of grass in such 

 a case by the abstraction by the grass of anything from the soil (or sand) 

 below the trays, and it must be due to the passage of something from the 

 trays down to the trees. The experiments on this subject were numerous, 

 and the grass-effect was uniformly shown in all of them; and, it should be 

 mentioned, the trees without grass, with which the grassed trees were com- 

 pared, were grown with trays of sand above their roots, so as to exclude 

 the possibility of explaining the results by the mere presence of the trays. 



"The ready oxidisability of the toxic matter formed by grass into some 

 substance which favours plant-growth will explain the previously observed 

 beneficial effect of grass-leachings in cases where these had been exposed 



