Theory of Electrolytic Dissociation. 229 



and then the solution extracts solvent from the membrane. 

 The latter part of the process can only take place when the 

 affinity of the solution for additional solvent exceeds that of 

 the saturated membrane for solvent. It is clear that, as the 

 solution is diluted, a strength will finally be reached where 

 the affinity of the solution for additional solvent is just equal 

 to that between membrane and solvent, and then osmosis will 

 stop. /. r.. the osmotic pressure will become nil. This critical 

 strength of the solution varies with the nature of the mem- 

 brane and that of the solution ; but in general it is far from 

 infinitely dilute, being quite within the pale of what is 

 readily measurable. The classical copper ferrocyanide mem- 

 brane, being a colloid, has but slight affinity for water, 

 which is readily extracted from it by a fairly dilute solution of 

 sugar. Thus osmotic pressure so-called is due to the same 

 affinity which causes adhesion, imbibition, absorption, ad- 

 sorption, solution, and chemical action. 



Of late years work in physical chemistry has been largely 

 directed to the study of the effects of temperature, pressure, 

 and concentration on the progress of chemical reactions; and 

 properly so. It has at times been forgotten, however, 

 that affinity must exist before union can take place ; and 

 that temperature, pressure, and concentration are simply 

 to be regarded as modifying factors aidino- or retarding 

 the tendency of affinity. By some it is even deemed as 

 somewhat old-fashioned to talk about affinity. But upon 

 the distinct recognition of affinity, and a careful inves- 

 tigation of the laws governing it, clearly depends the future 

 progress of chemistry, physical chemistry, and physiology. 

 Wo must learn to measure affinity quantitatively; but in 

 order to take into account the phenomena of morphology 

 as they confront us in crystals, but particularly in living 

 being-, we need to learn to study the direction as well as 

 the strength with which affinity acts under given conditions. 

 Finally, the question why certain solutions, molten salts, &c., 

 conduct electricity and others do not, will probably not be 

 answered until we can tell why a stick of silver conducts 

 electricity and a stick of sulphur does not. These questions 

 really involve a better understanding of the relation between 

 electricity and gros- matter, a problem which is apparently 

 being attacked with promise by J. J. Thomson and his 

 co-workers. Until we have more light on this subject, we 

 can hardly hope for very material improvements of our views 

 of the nature of electrolytic processes. 



Laboratory of Physical Chemistry, 

 University of Wisconsin, Madison, 



\ _ i~t. 1904. 



