EXPERIMENTS ON THE ROOTS OF PL-ANTS. 519 



clear, and scarcely coloured by the kidney bean, but turns more yellow 

 with the pea and common bean ; it precipitates the acetate of lead, and 

 nitric acid re-dissolves the precipitated gum without effervescence; 

 nitrate of silver gives a slight precipitate soluble in acids (carbonic 

 acid) ; oxalate of ammonia renders it turbid ; the other reactives cause 

 no change. By slow evaporation a yellowish or brownish residuum is 

 obtained, more or less abundant, according to the plant under experi- 

 ment, increasing in this order : kidney beans, peas, beans. In all other 

 respects these residua are similar to each other. Ether separates an 

 oily substance ; alcohol nothing, and a substance remains analogous to 

 gum and a little carbonate of lime. 



In the course of the experiments on these plants, I perceived that 

 when the water in which they had been kept was charged with much 

 excrementitious matter, the fresh flowers of the same species that were 

 put into it faded quickly, and did not live well in it. To ascertain if 

 this resulted from the want of carbonic acid, although they might draw 

 it from the air, or from the effect of the matter excreted, which these 

 plants refused to absorb, I replaced the leguminous plants by those of 

 another family, especially that of corn. The latter lived in it, and 

 the yellow colour of the liquid diminished in intensity ; the residuum 

 was less considerable, and it was evident that the new plants absorbed 

 a part of the matter excreted by the former. It was a kind of rotation 

 of crops in a bottle, and the result tends to confirm the theory of M. 

 De Candolle, of whom I spoke at the commencement of this memoir. 

 It is not impossible that by trying this experiment on a great number 

 of plants, we may arrive at some results which may be applicable to 

 the practice of agriculture: for example, by supposing, as I feel disposed 

 to believe by my trial, that the exudation from the roots of cultivated 

 legumes contributes to the nourishment of corn, I should be disposed 

 to conjecture, according to the relative quantity of these exudations, 

 that the bean will produce the finest wheat, then the pea, next to that 

 the kidney bean. I am not sufficiently a practical agriculturist myself 

 to know if experience has confirmed this view of the fact. 



Graminece. The plants examined were wheat, rye and barley. 



These plants do not thrive so well in rain water as the Leguminosce, 

 and I suppose that this difference arises from the great quantity of 

 mineral substances, especially silex, which they contain, and which 

 they do not imbibe from pure water. The water in which they have 

 vegetated is very clear, transparent, without colour, smell, or taste. 

 The reactives demonstrate the presence of salts, muriates and carbonates, 

 alkaline and earthy ; and the residuum from evaporation is scanty and 



