514 REPOET— 1888. 



ficetic acid on the score of its saturation vapour-densities at the higher 

 temperatures any essentially different molecular arrangement than we do 

 to alcohol and ether. Now, for these latter bodies, the vapour-densities 

 at saturation are at low temperatures noi-mal, i.e., normal under condi- 

 tions of temperature more favourable to complex molecules : Seeing that 

 complex molecules of alcohol are not formed at low temperatures, a fortiori 

 they are not at the higher temperatures ; there is no reason, seeing that 

 at the higher temperatures alcohol and acetic acid behave similarly, for 

 supposing that acetic acid has other than a normal molecular constitution 

 at hicrher temperatures : But there remains the peculiarity of acetic acid, 

 that its saturation vapour-density below 150° increases continually. The 

 pressure being relieved, we should expect, as in the case of alcohol, that 

 as the pressure diminishes with the temperature the vapour-density would 

 get smaller till it is normal. There must, therefore, for temperatures below 

 150°, be, in the case of acetic acid (also in that of nitric peroxide, whicb 

 is S'milar), some cause (different from the mere approximation of the 

 molecules by pressure, which at high temperatures accounts for the high, 

 vapour- densities at saturation) operating to increase the density. This 

 can be no other than a chemical cause, and is probably due to the com- 

 bination of some of the molecules by pairs, or, perhaps, partly into more 

 complex molecules. 



There are other arguments besides the one put forward here ; and for 

 these, and for the light thrown by the analogy of the behaviour of the 

 dissociating compounds, the original paper must be seen. The general 

 result from all the considerations adduced is, that for stable substances 

 the difference between the liquid and the gaseous state is not one of kind 

 but of degree ; that the phenomena accompanying the vaporisation of a 

 single stable body in the liquid state presents no evidence, from the study 

 of the vapour, in favour of the existence of complex molecules in the 

 liquid ; that, on the other hand, in the cases of acetic acid and nitric 

 peroxide, there is evidence of the formation of complex molecules in the 

 vaporous and in the liquid forms of these bodies ; the complex molecules 

 corresponding in some respects to molecules of an unstable, dissociable 

 body. 



The difference between the gaseous and the liquid condition of a stable 

 body is thus referred to the greater approximation to each other of the 

 gaseous molecules, and to the consequent increase in their attraction for 

 one another, which is known as cohesion. 



There is no doubt of the wonderful accuracy with, which the equation 

 p='bT — a expresses the results of experiment for four of the bodies 

 examined, viz., ethei", methyl- and ethyl- alcohols, and CO.2 ; and of the 

 considerable deviation of the results of experiments from the predictions 

 of this formula in the cases of nitric peroxide and acetic acid, and prob- 

 ably for formic acid and other bodies. The explanation given above of 

 the differences between these two sets of cases involves the separation 

 of bodies into two classes, to one of which the equation applies accurately, 

 and to the other not at all. 



Perhaps it will be found that even here it is a question of degree and 

 not of kind ; that bodies differ from one another in reference to the 

 degree to which they conform to the rule expressed by the above equa- 

 tion in the gas-liquid state ; that there is, in fact, a jierfect gas-liquid 

 state, as there is a perfectly gaseous state ; and that many bodies can be 



