CHEMICAL THEOKY OF SOLUTIONS. PART I. 11 



the former case the curves of total vapour j)i*essure are convex 

 to the axis of composition, while in the latter case they are 

 concave. He reasoned from these facts that for a mixture in 

 the formation of which neither a change of volume nor of heat 

 takes place the vapour tension curve must be a straight line. 

 He found his anticipation nearly realised in the mixture of 

 ethyl chloride and iodide. 



The results of the experimental studies of Linebaeger, 

 Hartmann, Lehfeld, and Zawidzki have been incorporated in 

 the standard W'Ork of Roozeboom (Heterogene Gleichgewichte, IT, 

 21), hence it is not necessary to discuss them here. In all the 

 cases where the curves of total as well as of partial pressure are 

 straight lines the components are substances known, by the 

 method of surface tension, to be unassociated. Some mixtures 

 of presumably normal liquids give curves which are not quite 

 straight, but the curvatures are mostly quite small. Doubtless 

 many liquids usually supposed to be normal are associated to a 

 slight extent. 



According to van dee Waals the conditions necessary for 

 straight vapour pressure curves are the following : 



{a) The heat of the evaporation of the mixture is the sum 

 of the heats of the evaporation of the components. 



{b) The critical pressures of the tw^o components are equal. 



That (a) is the consequence of (2) is evident, while {b) has 

 no direct relation with (1). Kohnstamm (Zeits. physik. Chem., 

 36, 52 ; 1901) tried to verify the condition {b), but could not 

 reach any satisfactory result owing to the inaccuracy of the data 

 on critical pressure. 



According to Young (Journ. Chem. Soc, 83, 45 ; 1903) 

 the chemical similarity of the components is one of the conditions 



