June 21, 1907] 



SCIENCE 



96^ 



the freezing-point and crystal water content of 

 the solute can not be considered as established 

 The well-known fact that salts generally sepa- 

 rate from hot solutions with less water of 

 crystallization than from cold solutions is also 

 regarded as evidence of the existence of 

 hydrates in solution. 



The fact that salts frequently crystallize 

 with water has long been considered as evi- 

 dence in favor of the assumption that hydrates 

 exist in solution. The reasoning has simply 

 been that whatever separates from a solution 

 at a given temperature has also existed in that 

 solution. So in this respect the paper really 

 presents nothing new. Furthermore, it has 

 long been known that the various physical 

 properties of solutions can not be foretold on 

 the assumption that solutions are merely me- 

 chanical mixtures of solvent and solute. The 

 deviations which the properties of solutions 

 have shown from those computed on the as- 

 sumption that they are merely mechanical 

 mixtures have always been regarded as evi- 

 dence in favor of the idea that in the process 

 of solution some form of combination between 

 solvent and solute takes place; and so the 

 argument in favor of the existence of hydrates 

 in solution based upon the occurrence of 

 minima in the freezing-point curves also pre- 

 sents nothing new in principle. However, 

 there has been in the past great difficulty in 

 determining definitely just what the composi- 

 tion of the supposed hydrates in solution 

 really is, and as Professor Jones presumes to 

 compute, at least approximately, the composi- 

 tion of these hydrates, the vital interest of the 

 whole monograph really centers upon the re- 

 sults of these computations. 



Now an inspection of the method adopted by 

 the author for determining the approximate 

 composition of the hydrates in solutions shows 

 that it consists of assuming the approximate 

 validity of the simple gas laws for all solu- 

 tions, concentrated as well as dilute; and that 

 in the case of electrolytes the solutes are elec- 

 trolytieally dissociated, the so-called degree of 

 electrolytic dissociation being calculable ac- 

 cording to the well-known formula of Ar- 

 rhenius even in concentrated solutions, at 

 least approximately. On the basis of these 



assumptions it is computed what the molecular 

 lowering of the freezing-point ought to be; 

 and from this value and that actually found 

 in practise, it is calculated how much water 

 is combined with the solute. In other words, 

 the method simply consists of assuming just 

 so much of solvent as combined with the dis- 

 solved substance in any given case as is neces- 

 sary to make the simple gas laws hold even for 

 concentrated solutions. That the gas laws do 

 not hold for concentrated solutions, even ap- 

 proximately, is a well-recognized fact; and 

 consequently the numerous computations of 

 the approximate composition of the supposed 

 hydrates in the solutions studied and the 

 curves representing the variations of the com- 

 position of these hydrates with change of con- 

 centration need not be further considered here, 

 for they are quite valueless. The republica- 

 tion of the results of the calculations of the 

 composition of these supposed hydrates in this 

 monograph is all the more inexcusable because 

 Van Laar' had already called attention to the 

 gross error of Jones and Getman when a part 

 of the work appeared in the Zeitschrift fur 

 physilcalische Chemie, 49, 385, 1904. 



While then the monograph has failed in 

 what must be regarded as its main objective, 

 namely, the computation of the approximate 

 composition of hydrates in solution, the large 

 bulk of experimental evidence presented, 

 though also in part not new, undoubtedly goes 

 to show that solutions are not merely me- 

 chanical mixtures, but that combination of a 

 specific nature does take place between solvent 

 and solute. The experimental data recorded 

 will no doubt prove useful in the future. 



Part II. of the paper, pages 161 to 259, pre- 

 sents the results of H. S. TJhler of the study 

 of the absorption spectra of aqueous solutions 

 of cobalt chloride, copper chloride and copper 

 bromide and of these solutions mixed in dif- 

 ferent proportions with calcium chloride, cal- 

 ciiun bromide or aluminum bromide. The 

 absorption spectra of cobalt chloride, copper 



^ Chemisch WeekMad, 2, 6-8, 1905. Abstract in 

 Chem. Centralblatt, 76, 491, 1905. Also reprinted 

 in ' Vortrage uber thermodynamisches Potential,' 

 pp. 6-8; published by Vieweg u. Sohn, Braun- 

 schweig, 1906. 



