284 ROOT EXCRETIONS. [BOOK n. 



salts of lime; their absorption appears infinitely less easy 

 than that of several other salts ; but the following experi- 

 ment throws much doubt on the conclusion to be drawn from 

 the facts cited by Theodore de Saussure. 



In a solution in distilled water containing one gramme of 

 sulphate of soda, and one gramme of chloride of sodium to 

 the litre, I planted a Polygonum persicaria, and when half 

 the solution was absorbed, I examined the residue, and found 

 in it, besides the oxalate of ammonia, a notable quantity of 

 lime, which did not exist in it previously, and which had 

 been furnished by the vegetable. 



This then is one capital cause of error which escaped 

 Theodore de Saussure. When a vegetable is immersed in an 

 aqueous solution, there is not a pure and simple absorption 

 of the solution, but a double current is formed. As the salt 

 of the solution passes into the plant, so the salts of the plant 

 arrive in the solution. jt This is the principle which M. Dutro- 

 chet has so well developed in his excellent investigations on 

 Endosmosis. 



There is a strong and a weak current, but always a double 

 current, and not a pure and simple absorption. This cause 

 of error is very important, for Theodore de Saussure operated 

 only upon 637 milligrammes, diminished by the fact of the 

 absorption alone, and he did not at all attempt, in his analysis, 

 as may be seen at page 255 of his Recherches sur la Vegeta- 

 tion, to find any other principles than those which he wished 

 to estimate ; moreover he has not indicated the weight of 

 the plants he employed. 



To avoid, as far as possible, the chances of error caused by 

 the excretions of the roots, I thought that plants should be 

 chosen which, living a considerable time in water, might, by 

 a very long vegetation, be brought into such a condition as 

 no longer to yield any fixed salt to the distilled water, and 

 which would yet possess a marked power of absorption. 

 Mentha aquatica seemed, from numerous previous experi- 

 ments, to fulfil these conditions much better than the Poly- 

 gonum Persicaria and Bid ens cannabina, selected by Theodore 

 de Saussure. The following is the manner in which my 

 experiments were made. Branches of mint, furnished with 



