nKSKARrUKS ON KVAPORATION AND DISSOCIATION. 327 



his view, altliougli tiiey are in direct contradiction to tlie 

 experimental evidence adduced by Bineau {AnnaUn, 60, 

 p. 157) and by us. But in bin calcubitions, lie ouiployed 

 Landolt's vapour-pressvxrcs, whicb, at the lowest tempera- 

 tures, n,ro nearly double those obtained by E,egnault and 

 by us. On recalculating his results with true vapour- 

 prossures, it is foujid that there is a tendency for the 

 density of the saturated vapour to rise, instead of to fall, 

 with decrease of temperature, though the results are not 

 suflicieutly concordant to bo iilottcd in the form of a curve. 

 (For details, see Trans. Ghom. Boc, 188(5, p. 809.) 



Whereas, with the alcohols, and with ether, as soon as 

 condensation i'rom gas to litpiid occurs, pressure remains 

 constant until condensation is complete, with acetic acid, 

 and with chloral ethyl-alcoholate the pressure continues 

 to rise, implying the condensation of the more complex 

 niolocuUw before tlie simpler ones; in fact, the substance 

 behaves like the mixture of alcohol and ether, of which it 

 was impossible to determine the true vapour-pressures. 



A soniowhat similar phenomenon was observed with 

 water, but in this case it was most noticeable before visible 

 condensation took place, and it was proved to be due to the 

 hygroscopic nature of the glass. Owing to this source of 

 error it was impossible to obtain direct determinations 

 of the density of the saturated vapour of water, but the 

 accuracy of the determinations of vapour-pressures was not 

 aifocted. 



Many attempts have been made to discover an equation 

 which shall represent accurately the relations of tem- 

 perature, pressure, and volume of a substance in both the 

 liquid and gaseous state. The most important advance 

 was made by Van dor Waals, who proposed the equa- 

 tion (p+'\)(v~h)='llil+at) in place of the ordinary 



