98 OSMOSIS AND OSMOTIC PRESSURE 



in the form of water-vapor. Callendar (1908) who first proposed such a 

 "vapor pressure theory" of osmosis, assumed the existence in differentially 

 permeable membranes of minute capillary pores too small to be wetted by 

 liquids, but freely permeable to molecules in the vapor state. Warrick and 

 Mack (1933) have described an example of an inorganic membrane through 

 which water passes in the vapor state, but it is impossible to say whether or 

 not this is also true of the differentially permeable membranes of plant cells. 



The essential concepts in the vapor pressure theory of osmosis can be 

 presented in terms of two simple experiments. Figure 20, A represents an 

 inverted U tube, one arm {x) of which is partially filled with a sucrose 

 solution while the other arm {y) is filled to the same level with pure water. 

 The space above the liquids contains only air plus the water-vapor which 

 enters it from the liquids. The air acts as a membrane which is permeable 

 to water (vapor), but not to sucrose molecules. The presence of the sucrose 

 molecules decreases both the diffusion pressure and the vapor pressure of the 

 water in the solution. In fact, increase in the diffusion pressure deficit 

 (osmotic pressure) of a solvent caused by the introduction of a solute is 

 accompanied by a proportionate decrease in its vapor pressure. As a result 

 the vapor pressure just above the layer of pure water will be greater than 

 that above the laj^er of the solution, and a gradient of vapor pressures will 

 be established, decreasing in magnitude from the water to the solution. Water 

 will therefore move from the region of high vapor pressure to the region of 

 low vapor pressure, and the solution will steadily gain in volume at the 

 expense of the pure water. Ultimately all of the water will be transferred 

 to the sucrose solution. The transfer of water from one arm to the other 

 has resulted from the difference in vapor pressures of the two liquids and is 

 therefore directly correlated with their differences in diffusion pressures. 



If this experiment were modified as illustrated in Fig. 20, Bj where 

 the sucrose solution in x is separated from the water in j by a rigid mem- 

 brane permeable only to water, and evaporation prevented from the open 

 arms of the U-tube by thin layers of oil, there will be a net movement of 

 water from y to x. There will be an increase in the volume of the solution 

 in Xj and a decrease in the volume of the water in y. If we assume that the 

 water moves through the membrane in the form of vapor there is no essential 

 difference between this phenomenon and the one first described. 



In the foregoing discussion the effect of temperature on the osmotic move- 

 ment of M^ater has been disregarded. If different parts of an osmotic system 

 are maintained at different temperatures this will have an effect on the osmotic 

 movement of water. However unless the difference in temperature on the 



