510 Theory of the Nutrition [Book hi. 



must always belong the merit of having brought into notice 

 this mechanical effect of endosmose and of employing it to 

 explain a number of vital phenomena. Many things in which 

 a mechanical explanation had not been hitherto thought of 

 could now be traced to a mechanical principle, the effects 

 of which could be exhibited and more accurately studied 

 by means of artificial apparatus. Dutrochet rightly attached a 

 special value to the fact, that all states of tension in vegetable 

 tissue could be at once explained by endosmose and exosmose, 

 though, as so often happens in such matters, he may have ex- 

 tended his new principle to cases where it was not applicable, 

 as we shall see below. His account of the nature of endos- 

 mose itself must now be considered to be obsolete, nor did 

 the mathematician Poisson or the physicist Magnus about 

 1830 succeed in framing a satisfactory theory on the subject. 

 It was discovered in the course of the succeeding twenty 

 or thirty years, that the phenomena observed by Dutrochet, 

 and which he called endosmose and exosmose, were only com- 

 plicated cases of hydro-diffusion, which with the diffusion of 

 gas forms an important part of molecular physics. Dutrochet, 

 like his immediate successors, conducted his investigations 

 into osmose with animal and vegetable membranes, the latter 

 being of a complex structure ; with these he always observed 

 in addition to the endosmotic flow of water into the more 

 concentrated solution, an escape of the solution itself, and 

 from this he concluded that there must always be two currents 

 in opposite directions through the membrane which separates 

 the two fluids, that, as he expresses it, the endosmose is always 

 accompanied with exosmose. This error, which was even 

 developed later into a theory of the endosmotic equivalent, has 

 had much to do till recently with making it impossible or 

 difficult to refer certain phenomena of vegetation to the pro- 

 cesses of hydro-diffusion. To mention only one case, Schlei- 

 den rightly observed that if endosmose, as Dutrochet under- 

 stood it, is the sole cause why water is absorbed by the roots, 



