182 SCIENCE PROGRESS. 



molecular and the dissolved substance is associating (Case 

 III.) lower values than the normal are obtained. This 

 is not what one would at first expect, for there seems no 

 immediate reason why, in an associating solvent, a dissolved 

 substance which is also associating should give values 

 which are different to those obtained when the solvent is 

 monomolecular. It is evident in fact that the difference in 

 the solvent exercises the determining influence in these 

 cases and not the character of the dissolved substance, 

 as otherwise an associating compound would in all solvents 

 give lower values for the osmotic pressure than the normal. 

 The difficulty has been usually explained by assuming that 

 an associating solvent brings about a dissociation in any 

 dissolved compound of an associating type, and reduces 

 it to the monomolecular condition. The dissociation here 

 spoken of is not electrolytic dissociation, but a breaking 

 down of complex molecules into simple ones. 



It is difficult to admit the universal dissociating influence 

 thus assigned to associating solvents, without any explana- 

 tion of the why and wherefore of its existence. For 

 according to this view every compound, whatever its 

 character, when dissolved in any associating solvent, must 

 be regarded as undergoing dissociation. If the compound 

 is associating, the action merely proceeds as far as the 

 breaking up of the complex molecules into simple ones, 

 and the dissociation is therefore of the ordinary kind. 

 But if the compound is monomolecular, the action is of 

 a more complex character, for the molecules themselves 

 are broken up, and so-called electrolytic dissociation occurs. 

 A far simpler view of the matter is obtained if it is 

 allowed that the abnormal behaviour of solutions con- 

 taining an associating solvent is not due to any change in 

 the dissolved compound, but due solely to the character of 

 the solvent itself, and its influence on the osmotic pressure. 

 It has been already pointed out that monomolecular and 

 associating liquids are not directly comparable one with 

 another. Admitting the validity of the osmotic pressure 

 formula? when the solvent is a monomolecular one, it does 

 not therefore follow that these formulae will necessarily 



