622 SCIENCE PROGRESS 



manipulative difficulties involved in effecting a satisfactory con- 

 nection between the cell and the manometer, and suggest that 

 the largest single experimental error in their work is that due 

 to the lack of rigidity in the rubber stopper employed to close 

 the cell. 



In the experiments carried out by Lord Berkeley and 

 Mr. Hartley, a porous tube with the semi-permeable membrane 

 deposited close to its outer surface passed tight through the 

 centre of a metal chamber containing the solution. This metal 

 chamber consisted of two halves screwed together, and the 

 joints were rendered tight with the aid of dermatine rings. The 

 metal vessel was connected with an apparatus by means of 

 which a known hydrostatic pressure, sufficient to balance the 

 pressure of the inflowing water, could be applied to the solution. 

 The porous tube, occupying a horizontal position and filled with 

 water, was connected at each end with a vertical glass tube. 

 Observation of the level of the water in one of these tubes 

 showed whether water was entering the solution through the 

 membrane, or whether the applied hydrostatic pressure was 

 squeezing water out of the solution. In the latter case the 

 applied hydrostatic pressure must be greater than the osmotic 

 pressure, in the former case less. By adjusting the apparatus, 

 therefore, so that water is neither entering nor leaving the 

 solution, an equilibrium pressure is found which is equal to 

 the osmotic pressure. 



Magnitude of the Pressures observed with Copper 

 Ferrocyanide Membranes 



V 



The number of substances for which there is no passage 

 through a membrane of copper ferrocyanide is very limited. 

 Recent work on the osmotic pressures observed with such 

 membranes deals almost exclusively with solutions of cane 

 sugar, although the behaviour of dextrose and one or two other 

 sugars has also been the subject of investigation to a limited 

 extent. The statement that a copper ferrocyanide membrane 

 is absolutely non-permeable for cane sugar should perhaps 

 be qualified ; but there seems to be no doubt that when the 

 conditions are favourable and due care has been taken in the 

 setting up of the osmotic cell, the amount of sugar which passes 

 through the membrane, even from a highly concentrated solution, 

 is practically nil. 



