82 SOME RECENT RESEARCHES IN PLANT PHYSIOLOGY 



molecules by union with those of the solute, but this could 

 not be a satisfactory explanation in the case of an ideal 

 solution in which there is no such combination. 



THE NATURE OF SEMI-PERMEABLE MEMBRANES. 



A semi-permeable membrane is one which is permeable 

 to the solvent only, and does not permit of the passage 

 of solutes. Some membranes show selective permeability 

 in that they allow certain solutes to pass through, but 

 restrain others. It has frequently been urged that no 

 membrane is absolutely semi-permeable, but that some 

 small quantity of solute always penetrates. In view of 

 the perfection to which the preparation of copper ferro- 

 cyanide membranes has been brought by Morse and 

 Frazer (1911), the Earl of Berkeley and Hartley (1906), 

 and others, this objection can be no longer maintained 

 except as a purely theoretical belief beyond the limits of 

 experimental verification. Morse (1911), for instance, 

 working with a sucrose solution, found that a pressure of 

 over 12 atmospheres was sustained without any leakage 

 for sixty days at the temperature of 15. 



The earliest view of the structure of semi-permeable 

 membranes was that they acted as sieves, allowing the 

 passage of molecules which were below a certain size. 

 This conception is certainly too crude. 



H. E. Armstrong (1906, 1909, 1912), using as an illustra- 

 tion A. J. Brown's (1912) experiments on the selective 

 permeability of the seed coats of Hordeum vulgare, has 

 sustained his theory that such membranes permit the 

 passage of substances which are hydrated feebly or not 

 at all, whereas those which form stable hydrates are un- 

 able to penetrate. This view is in many respects a very 

 tempting one, as in solvents other than water quite similar 



