THEORY OF ITS USE. 163 



Of all the views which have been adopted regard- 

 ing the cause of the favorable effects of the alter- 

 nations of crops, that proposed by M. Decandolle 

 alone deserves to be mentioned as resting on a firm 

 basis. 



Decandolle supposes, that the roots of plants 

 imbibe soluble matter of every kind from the soil, 

 and thus necessarily absorb a number of substances 

 which are not adapted to the purposes of nutrition, 

 and must subsequently be expelled by the roots, and 

 returned to the soil as excrements. Now as excre- 

 ments cannot be assimilated by the plant which eject- 

 ed them, the more of these matters which the soil 

 contains, the more unfertile must it be for the plants 

 of the same species. These excrementitious matters 

 may, however, still be capable of assimilation by 

 another kind of plants, which would thus remove 

 them from the soil, and render it again fertile for 

 the first. And if the plants last grown also expel 

 substances from their roots, which can be appropri- 

 ated as food by the former, they w^ill improve the 

 soil in two ways. 



Now a great number of facts appear at first sight 

 to give &, high degree of probability to this view. 

 Every gardener knows, that a fruit-tree cannot be 

 made to grow on the same spot where another of the 

 same species has stood ; at least not until after a 

 lapse of several years. Before new vine-stocks are 

 planted in a vineyard from which the old have been 

 rooted out, other plants are cultivated on the soil 

 for several years. In connexion with this it has 

 been observed, that several plants thrive best when 

 growing beside one another; and, on the contrary, 

 that others mutually prevent each other's develop- 

 ment. Whence it was concluded, that the beneficial 

 influence in the former case depended on a mutual 

 interchange of nutriment between the plants, and 

 the injurious one in the latter on a poisonous action 

 of the excrements of each on the other respectively.* 



* That these supposed exudations are uniformly more or less injuri- 



