DIFFUSION AND OSMOSIS. 915 



cules will pass through from one side as from the other; the two streams, 

 in fact, will neutralize each other, and the volumes of the two masses of 

 water will remain unchanged. The movement of the water molecules in 

 this case is not actually observed, but it is assumed to take place on the 

 theory that the liquid molecules are continually in motion and that the 

 membrane, being permeable, offers no obstacle to their movements. If, 

 now, on one side of the membrane we place a solution of some crystalloid 

 substance, such as common salt, and on the other side pure water, then it 

 will be found that' an excess of water will pass from the water side to the 

 side containing the solution. In the older terminology it was said that the 

 salt attracted this water, but in the newer theories the same fact is expressed 

 by saying that the salt in solution exerts a certain osmotic pressure, in conse- 

 quence of which more water flows from the water side to the side of the 

 solution than in the reverse direction. As a matter of experiment it is found 

 that the osmotic pressure varies with the amount of the substance in solu- 

 tion. If in experiments of this kind a semipermeable membrane is chosen 

 that is, a membrane that is permeable to the water molecules, but not to 

 the molecules of the substance in solution the stream of water to the side 

 of the crystalloid will continue until the hydrostatic pressure on this side 

 reaches a certain point, and the hydrostatic pressure thus caused may be 

 taken as a measure of the osmotic pressure exerted by the substance in solu- 

 tion. Under these conditions it can be shown that the osmotic pressure 

 is proportional to the concentration of the solution, or, in other words, to 

 the number of molecules (and ions) of the crystalloid in solution. As a 

 matter of fact, most of the membranes that we have to deal with in the 

 body are only approximately semipermeable that is, while they are readily 

 permeable to water molecules, they are also permeable, although with more 

 or less difficulty, to the substances in solution. In such cases we get an 

 osmotic stream of water to the side of the dissolved crystalloid, but at the 

 same time the molecules of the latter pass to some extent through the mem- 

 brane, by diffusion, to the other side. In course of time, therefore, the 

 dissolved crystalloid will be equally distributed on the two sides of the mem- 

 brane, the osmotic pressure on both sides will become equal, and osmosis 

 of the water will cease to be apparent, since it is equal in the two directions. 

 All substances hi true solution are capable of exerting osmotic pressure, 

 and the important discovery has been made that the osmotic pressure, meas- 

 ured in terms of atmospheres or the pressure of a column of water or mer- 

 cury, is equal to the gas pressure that would be exerted by a number of 

 molecules of gas equal to that of the crystalloid in solution, if confined within 

 the same space and kept at the same temperature.* A perfectly satisfactory 

 explanation of the nature of osmotic pressure has not been furnished. We 

 must be content to use the term to express the fact described. It is a matter 

 of great importance to measure the osmotic pressures of various solutions. 

 As was stated above, this measurement can be made for any solution pro- 

 vided a really semipermeable membrane is constructed. As a matter of 

 fact, however, the use of such membranes has not been general. In actual 

 experiments other methods have been employed, and a brief statement 

 of a theoretical and a practical method of arriving at the value of osmotic 

 pressures may be of service in further illustrating the meaning of the term. 

 Before stating these methods it becomes necessary to define two terms 

 namely, electrolytes and gram-molecular solutions that are much used 

 in this connection. 



Electrolytes. The molecules of many substances when brought into 



* The interesting researches of Morse and Frazer (" The American Chemi- 

 cal Journal," 34, 1, 1905), who have succeeded in making semipermeable 

 membranes in such a form as may be used for determining directly the os- 

 motic pressures of concentrated (normal) solutions, have shown that this 

 law is not accurately stated. The actual pressure is that which would be 

 exerted if the particles in solution were gasified at the same temperature and 

 kept to the volume of the pure solvent used (water), instead of the volume 

 of the entire solution. 



