Taylor. — Preservation of Fish by Salt 123 



and we desire to know what will happen, and how we can 

 influence the process to suit our needs. The quantity and 

 direction of flow through the skin or cell-membrane will de- 

 pend on (1) the nature of the dividing membrane; and (2) 

 the nature and quantity of the substances dissolved in the 

 water on each side. 



The nature of the dividing membrane will be considered 

 first. Almost any substance can be made into a thin film 

 or membrane. Such things as glass, tinfoil and mica may be 

 exceedingly thin, but are totally impermeable and therefore 

 uninteresting in the present connection. But other membranes 

 or films, such as parchment paper, gelatin films, animal blad- 

 ders, and gold beaters' skins, are permeable to a greater or 

 smaller degree. Suppose pure water were on one side of a 

 membrane and water containing dissolved salt on the other. 

 If the membrane is perfectly permeable to all constituents, 

 water will pass through to the salt solution, and salt will pass 

 through to the water, and these movements will continue until 

 the two sides are alike, and then stop. It is always the ten- 

 dency for the two liquids to come to equilibrium, and they 

 would do so if the membrane were perfectly permeable. 

 Nearly all membranes, however, permit a freer flow of the 

 solvent, in this case water, than they do of the solute, that 

 which is dissolved, in this case, salt. If the membrane per- 

 mits the water to flow, but absolutely prevents passage of a 

 dissolved substance, the membrane is said to be semi-perme- 

 able. In the example taken above, of pure water on one side, 

 and salt solution on the other, if the membrane were semi- 

 permeable, then the water would pass through to the salt solu- 

 tion, but the salt could not get through to the water. The 

 level of the pure water would fall and that of the salt would 

 rise; the difference in liquid level would exert a pressure 

 called osmotic pressure. Ideally semi-permeable membranes 

 are not realized in nature, though some of the membranes in 

 plants and animals approach ideal semi-permeability while they 



