i 2 6 SCIENCE PROGRESS 



and H 2 S0 4 . He therefore concluded that the breaks in the 

 curve were due to the formation of the series of hydrates 

 represented by the above formulae : each section of the density 

 curve showed the behaviour of a mixture of two hydrates, the 

 composition of which was indicated by the breaks at either end, 

 whilst the breaks themselves were the natural consequence of 

 the disappearance of one hydrate from the solution and its 

 replacement by another hydrate of different composition and 

 different density. 



The process of hydrate formation could, of course, be 

 reversed by evaporation or distillation, but it was necessary in 

 applying Mendeleefs method to assume that in any given 

 solution the combination of solvent and solute was essentially 

 complete and non-reversible. Abrupt changes in the properties 

 of the solution were, in fact, only possible if the whole of the 

 solvent entered into combination with the solute, and in such a 

 way that no appreciable quantity of a third hydrate C was 

 found until the first hydrate A had been completely converted 

 into the intermediate hydrate B. 



It is noteworthy that Mendeleefs paper is found along with 

 the classical papers of Arrhenius and van t'Hoff in the first 

 volume of the Zeitschrift filr physikalische Chemie. The pub- 

 lication of this journal marked the inauguration of a new 

 period of study of physical chemistry ; and the present year, 

 which celebrates its coming of age, affords a particularly 

 opportune moment for a review of the progress which has been 

 made by the rival theories during the intervening twenty-one 

 years. During the major part of this period the followers of 

 Arrhenius and van t'Hoff enjoyed almost a monopoly of publica- 

 tion, so far, at least, as the Continental journals were concerned ; 

 the hydrate theory was openly ridiculed by the German school 

 of physical chemists, whilst papers dealing with the theory of 

 electrolytic dissociation increased year by year both in number 

 and in bulk. With the coming of the new century, however, 

 the hydrate theory has come into its own, and in its maturer form 

 has again formed the subject of many important researches, and 

 has attained to almost universal acceptance. There are, indeed, 

 few cases on record in which the elements of truth existing in 

 each of two rival theories has worked more effectively in 

 bringing about a final reconciliation of views which were long 

 thought to be wholly incompatible. 



