1909] on Osmotic Phenomena. 493 



difference may be adjusted to balance, and is preferably measured by 

 means of a pair of differential platinum thermometers, which permits 

 a higher order of accuracy to be attained than the thermo-electric 

 method. 



Vapour-Pressure in relation to Molecular 

 Constitution. 



The well-known law of Raoult, according to which the relative 

 lowering of vapour-pressure of a solution is equal to the ratio of the 

 number of molecules n of the solute to the number of molecules of 

 solvent N in the solution, has thrown a great deal of light on the 

 molecular state of the dissolved substance in dilute solutions, but fails 

 notably in many cases when applied to strong solutions. In the case 

 of homogeneous mixtures of two indifferent volatile substances, such 

 as benzol (CgHg) and ethylene chloride (C0H4CI.,), \vhich mix in all 

 proportions wathout mutual action, a slightly different but equally 

 simple law is known to hold very accurately throughout the whole 

 range of concentration from to 100 per cent. The vapour- 

 pressure of each ingredient is simply proportional to its molecular 

 concentration. In other words, the ratio of the partial vapoar-pressure 

 p' of either constituent at any concentration to its vapour-pressure 

 p^ in the pure state at the same temperature is equal to the ratio of 

 the number of its molecules n' in the solution to the whole number 

 of molecules n' + n" of both substances in the solution. Such is 

 evidently the form of the simple mixture law. For substances which 

 form compounds in the solution, or whose molecules are associated or 

 dissociated, this simple law is widely departed from. In a recent 

 paper, " On Vapour-Pressure and Osmotic Pressure of Strong 

 Solutions " (Proc. E.S.A., vol. Ixxx. p. 466, 1908), I have endeavoured 

 to extend this simple relation to more complicated cases by making 

 the obvious assumption that, if compound molecules are formed, 

 they should be counted as single molecules of a separate substance in 

 considering their effect on the vapour-pressure. With this proviso 

 the vapour-pressures of strong solutions are well represented by a 

 natural extension of the simple mixture law% and it becomes possible 

 to investigate the nature of the compounds formed in any case. To 

 take a simple instance, suppose that each of the n molecules of the 

 dissolved substance combines with a molecules of the solvent, the 

 total number of molecules of the solvent being N. The ratio of the 

 vapour pressure p" of the solvent in the solution to the vapour- 

 pressure j?' of the pure solvent at the same temperature will then be 

 the same as the ratio of the number N - an of molecules of free 

 solvent in the solution to the whole number of molecules ^ -an + n 

 in the solution, each compound molecule being counted as a single 

 molecule. 



