80 SOME RECENT RESEARCHES IN PLANT PHYSIOLOGY 



position of the membrane permits of this condition being 

 realized, for the solution can be subjected to an additional 

 pressure, to which the pure solvent is not exposed. In 

 the natural course of events, this excess pressure is pro- 

 duced by the entrance of the solvent into the solution 

 by the process of diffusion which is here termed 

 osmosis. 



The foregoing considerations may perhaps be rendered 

 more easily intelligible by reference to the accompanying 

 figure, which represents a tube in the form of a rectangle 

 with two sides vertical. In it a solution of a non- volatile 



.__ : , substance is separated from the 



pure solvent by a semi-permeable 

 membrane below, and by a space 

 P. . containing only the vapour of the 

 solvent above. It is evident that 

 passage of the solvent into the solu- 

 tion will take place till a position of 

 equilibrium is attained. At this 

 stage it is clear that the vapour 

 pressure of the solvent at s must be 

 greater than that of the solution at 

 s r by an amount equal to the pres- 

 FlG 4 sure exerted by the column of vapour 



of height h, which is the difference 



in level between the liquid in the two limbs. Under these 

 conditions the vapour pressure at s f and at t, a point at the 

 same level in the other limb, will be equal, and no isother- 

 mal distillation will take place. 



Turning now to the semi-permeable membrane m, separa- 

 ting the liquids, since a condition of equilibrium has been 

 reached, the net amount of transference of solvent across 

 the membrane must be zero. This result is, of course, 

 quite independent of the nature of the membrane, pro- 



