396 HOW CROPS GROW. 



chemical action on the membrane must be different on 

 the two sides, and apparently not in degree only, but 

 also in kind, viz., an alkaline action on the albuminoid 

 substance of the membrane on the one side, and an acid 

 action on the other. The water appears always to accu- 

 mulate on the alkaline or basic side of the membrane. 

 Hence, with an alkaline salt, like potassium carbonate, 

 in the osmometer, and water outside, the flow is inwards ; 

 but with an acid in the osmometer, there is negative 

 osmose, or the flow is outwards, the liquid then falling 

 in the tube. 



Osmotic activity is most highly manifested in such 

 salts as easily admit of decomposition with the setting 

 free of a part of their acid, or alkali. 



Hydration of the membrane. It is remarkable 

 that the rapid osmose of potassium carbonate and other 

 alkali-salts is greatly interfered with by common salt, is, 

 in fact, reduced to almost nothing by an equal quantity 

 of this substance. In this case it is probable that the 

 physical effect of the salt, in diminishing the power of 

 the membrane to imbibe water (p. 393), operates in a 

 sense inverse to, and neutralizes the chemical action of, 

 the carbonate. In fact, the osmose of the carbonate, as 

 well as of all other salts, acid or alkaline, may be due to 

 their effect in modifying the hydration* or power of the 

 membrane, to imbibe the liquid, which is the vehicle of 

 /their motion. Graham suggests this view as an explana- 

 tion of the osmotic influence of colloid membranes, and 

 it is not unlikely that in case of earthenware, the chem- 

 ical action may exert its effect indirectly, viz., by pro- 

 ducing hydrated silicates from the burned clay, which 

 are truly colloid and analogous to animal membranes in 

 respect of imbibition. Graham has shown a connection 

 between the hydrating effect of acids and alkalies on 

 colloid membranes and their osmotic rate 



* In case water is employed as the liquid. 



