22 6 Prof. L. Kahlenberg on the 



considerable ranges of* alterations of temperature, pressure, 

 and nature of co-existing pbases, is not at all affected by tbe 

 above considerations. Tbe discovery of these laws has been 

 of inestimable value in analytical and synthetic work and in 

 systematizing our knowledge ; but it is a mistake to think that 

 they necessitate the conclusion that the processes of solution 

 and chemical action are different in character and are caused 

 by different agencies. The fact that the quantities of similar 

 solutes which must be added to equal quantities of a given 

 solvent to produce solutions of the same A r apour-tension, are to 

 each other as the quantities in which such solutes unite with 

 other substances to form stereotyped chemical compounds, 

 constitutes a strong argument for regarding the processes of 

 solution and chemical action as identical in character. 



It thus becomes evident that in investigating solutions we 

 must begin with the most concentrated and end with the most 

 dilute ; the latter will appear simply as a limiting case. 80, 

 for instance, the change of the vapour-tension of solutions 

 with the concentration must be studied from the strongest 

 solutions obtainable to the dilutest that can still be measured 

 throughout the range of temperatures at which the solutions 

 can exist at all. And this work must be done for a large 

 number of solutes in a large number of solvents. Such data 

 being at hand, the equations expressing the changes of vapour- 

 tension with temperature and concentration may be written. 

 Work of this character has only barely begun ; but judging 

 from the results at hand, analogous substances will exhibit 

 similar behaviour ; and though it is not to be expected that 

 one equation will serve for all solutions, similarities between 

 the equations holding for different solutions will not be 

 lacking. 



According to the views here advanced — or, I had better 

 say, revived — the molecular weights of, for instance, sodium 

 and potassium salts in aqueous solutions as determined by the 

 diminution of the vapour-tension, elevation of the boiling- 

 point, and lowering of the freezing-point, are not abnormally 

 low because these salts are dissociated, but rather because 

 they have great affinity for water. On the other hand, these 

 salts have less affinity for liquid sulphur dioxide, as is indi- 

 cated by the relatively high vapour- tension of the solutions 

 and the correspondingly high molecular weights, though the 

 solutions conduct well nevertheless. As an example of the 

 other extreme, the molecular weights of colloids in water arc 

 not enormous ; the results come out high simply because 

 the affinity between water and these colloids is relatively 

 slight. Furthermore, it is also easy to see why colloids are 



