Grafts et al. — 46 — Water in Plants 



in B rising, then lowering until it finally reached the initial state as all hydrogen was 

 replaced by air in the thimble. 



If, instead of a beaker of water, a water manometer is placed at the lower end of 

 C, both positive pressure (corresponding to the bubbling) and negative pressure (as 

 noted by the rise of water in B) could be registered. With a manometer in place, if 

 a coarse-pored thimble is used, response to change of gas around A is more rapid but 

 the maximum and minimum pressures measured in B would be less than with a fine- 

 pored thimble. Finally, if a differentially permeable membrane (popularly termed 

 semi-permeable) such as heated palladium or silica, which is permeable to hydrogen 

 but not to oxygen or nitrogen, is used in place of a porous thimble at A, the pressure 

 in B will rise to a high value, depending upon the partial pressure of hydrogen inside 

 the apparatus and this pressure will remain at its maximum height so long as hydrogen 

 is kept in contact with the differentially permeable membrane and no leaks develop. 



Although the case ©f the porous membrane would be difficult to analyse because 

 of the different velocities of the gases involved, that of the differentially permeable 

 membrane seems clear. When the pressure within this gas osmometer reaches a maxi- 

 mum the partial pressure of hydrogen inside equals the pressure outside and movement 

 of the molecules in the two directions is equal. The excess gas pressure within the 

 apparatus must therefore be equal to the partial pressure of the molecules of air that 

 are unable to penetrate the membrane when the system is placed under the total external 

 applied pressure of the manometer. 



Osmosis in Liquid Systems: — The problems of osmosis w^here a 

 liquid solvent and liquid or solid solutes are involved differ from the above 

 principally in the differences that exist between the gaseous state and the 

 liquid state. With liquids the vapor pressure is a manifestation of the 

 ability of the molecules to escape from the body of the liquid against at- 

 tractive forces within the liquid and surface tension forces at the surface. 

 It is the actual pressure (or partial pressure) of the vapor phase in equilib- 

 rium with the liquid and it reflects the kinetic energy of the molecules as 

 influenced by temperature and, to a slight degree, total pressure. Diffusion 

 pressure, on the other hand, is a measure of the average intrinsic energy 

 of all the molecules and is a function of internal pressure, the force that 

 renders all liquids relatively incompressible. 



A gas under increasing pressure is readily compressed and eventually 

 changes state to become a liquid under proper temperature conditions. A 

 gas under reduced pressure will expand indefinitely. The repulsive forces 

 between molecules are much greater than the attractive ; both however are 

 relatively small, and the latter almost disappear as a gas is rarified. A 

 liquid represents a state of equilibrium between attractive forces and re- 

 pulsive forces. As the molecules are forced closer together the repulsive 

 forces increase very rapidly so that the pressure rises tremendously with 

 small decrease in volume. Under reduced pressure a liquid expands very 

 little and tensions of 72 atmospheres for ether and greater than 100 atmos- 

 pheres for water have been attained before the forces of cohesion are over- 

 come and a vapor phase appears. Hildebrand (1924, p. 102) has pic- 

 tured the relation of internal pressure and cohesion in the following man- 

 ner (Figure 12). 



While the volume changes involved in the gas osmometer are different 

 in magnitude than those of the liquid, the underlying kinetic principles are 

 the same. As evidenced in (Chapter IV, the improved formulae for cal- 

 culating osmotic pressure from concentrations involve an expression of 

 the relationship between the numbers of solute and solvent molecules pres- 

 ent and corrections to account for the vokmies occupied by the molecules 

 and the attractive force between them. Though the formulae may not meet 

 the requirements for solutions of all concentrations, it is not because of the 

 failure of the basic principles upon which they are founded but because we 



