458 REPORTS ON THE STATE OF SCIENCE. 



freezing-points of isomorphous mixtures. He found that Van't Hoff's 

 law for the depression of the freezing-point * was not obeyed in these 

 cases, but that the freezing-points lay in the neighbourhood of the point 

 calculated from the freezing-points of the components of the mixture. 



It was found by Nernst 18 ° that a substance which gives a different 

 value for its molecular weight in one solvent from that found with 

 another solvent gives an abnormal value for its distribution coefficient 

 between these liquids. This he attributed to the fact that one liquid 

 acted better as a simplifying agent than the other; a conclusion sup- 

 ported by experiments made with benzoic acid in water and in benzene. 



During that year Roozeboom 182 expressed views regarding the 

 limitations of Nernst 's law of the relative lowering of the solubility 

 1ftQ „ and the partition law; such limitations were, however, strongly 

 lby ^' contested by Nernst. 1 ** 



The mutual solubility of salts engaged the attention of he 

 ' Chatelier 221 ; he distinguished three general cases: — 



(1) Salts solidify to form isomorphous mixtures of variable com- 

 position. 



(2) Each salt solidifies separately from the mixture. 



(3) The two salts combine and solidify as a compound of definite 

 composition. 



Cases of the first kind having been considered by this author as 

 well as by Kiister 1S3k , in this paper instances of the second class are 

 recorded. 



Kiister 208 investigated the distribution of ether between water and 

 caoutchouc, and found evidence for recognising the existence of a con- 

 siderable proportion of double ether molecules 2 [(C 2 H s )0] at low 

 temperatures. From experiments f on the distribution of iodine 

 between water and starch 208a , carried out in the same year, he came 

 to similar conclusions regarding iodine. 



Jakowkin 2 * 3 determined the distribution coefficient of iodine 

 and of bromine between water and some other solvent, and 

 found, with some solvents, that Kuster's observations were substan- 

 tiated, while with other solvents this was not the case. 



VII. Theoretical Considerations. 



Up to the time of Lavoisier the phenomenon of dissolution was 

 regarded as the result of some purely mechanical process, and, to this 

 end, many ingenious explanations were advanced.]: According to 

 Lavoisier, and indeed nearly all the chemists of that period, § the 



* Zeitschr. f. physik. Chem. (1890), 5, 322. 



f Further references and details may be obtained from ' Der Verteilungssatz,' 

 by W. Herz, Ahren's Sammlung, 15 (i) [1909]. 



X Newton imagined there are spaces in the water, in which the salt takes up 

 position ; and to explain how it was that water saturated with one salt was still 

 capable of dissolving another salt, Gassendi (Chaptal, Elements de Chemie, i. 40 ; 

 Montpellier, 1790) supposed these spaces to be of different shapes. 



§ Berthollet, Statique Chimique, p. 50 ; also Fourcroy, Systeme des Connaissances 

 Ohimiques, ii., p. 360. 



