64 ESSAYS BIOGRAPHICAL AND CHEMICAL 



and zinc sulphates, take equal times to diffuse. More- 

 over, some salts may be decomposed into their constitu- 

 ents by diffusion ; among these are ordinary alum, where 

 the more easily diffusible potassium sulphate passes away 

 from the less quickly diffusing aluminium sulphate. 

 And even potassium sulphate itself shows signs of 

 yielding potassium hydroxide and sulphuric acid on 

 diffusion. 



It was known that a solution, placed on the outside of 

 a porous diaphragm, on the inside of which was pure 

 water, tended to pass through the septum; and if the 

 inner vessel, containing the water, were fitted with a 

 pressure-gauge, the pressure would rise in the interior. 

 This pressure had been named ' osmotic pressure/ 

 Graham attempted to connect this phenomenon with 

 diffusion, but found that ordinary salts, as well as sugar, 

 tannin, alcohol, urea, and similar bodies, had little effect 

 in raising pressure. On the other hand, osmotic pheno- 

 mena were well marked when strong acids, or tartaric, 

 citric, or acetic acids, were present in the cell. In all 

 cases of osmotic pressure, it was found that the porous cell 

 was strongly attacked, and Graham was inclined to ascribe 

 the phenomenon to chemical action. It is in all pro- 

 bability due to the fact that such diaphragms present 

 very little of what we now term 'semi-permeability' to 

 the salts in question. 



From the year 1852 to the year 1861, Graham's duties 

 at the Mint absorbed nearly all his time, so that there is 

 a long gap in the series of his publications. But in the 

 latter years he published the results of experiments on 

 the transpiration of liquids, a subject which has lately 

 been successfully treated by numerous investigators. 

 And with his practical bias, Graham devised a plan of 

 applying osmotic phenomena to the separation of 

 crystalline substances, which easily pass through a 



