Properties of a Mixture of Liquids. 405 



several mixtures of different composition in each case. But 

 the results, as well as those of Wullner on alcohol and water, 

 are not expressible by equation (18) ; that equation is repre- 

 sented by a simply-shaped curve, not possessing the most 

 characteristic features of some of Konowalow's curves — 

 notably those of propyl alcohol and of formic acid. 



Brown, as previously mentioned, gives both vapour-pressure 

 and composition of vapour ; the two quantities should, of 

 course, give the same value of r. He used, however, only 

 one mixture at any particular temperature — viz., that whose 

 vapour-pressure was 760 millim. Moreover, he states that 

 the vapour was cooled in the upper part of his distilling-flask, 

 so that the vapour which came over was presumably in equili- 

 brium with a liquid of somewhat lower boiling-point (and 

 therefore containing less benzene) than that in the flask. 



It will be seen from the form of the equations, that when 

 t is given r can at once be found. For 



-© 'OS)- 



V A ) 7T B 



giving . . 7Ta 



6 6 logs-log — 



-—if (23) 



But when only^> is given equation (18) must be used, and 

 this cannot be solved directly. It is, however, fairly amenable 

 to logarithmic computation, so that by assuming values of r 

 and making successive approximations, the true value may be 

 found without much trouble. An actual example will best 

 show how the approximation proceeds : — 



Benzene and Carbon 



Bisulphide at 50° (F. D. Brown) . 



A (Benzene) = 



78. 



7r A = 269 millim. 



B 



76. 



7r B = 857 „ 



With q = 1 





p (obs.) = 760. 



With r = 1 





p (calc.) = 712-8, 



•9 





742-8, 



•87 





751-7, 



•845 



o In 



759-3. 



nlrarl fnr IipKvppti nncprvorl ' 



calculated vapour-pressures is not of a high order : Wullner 



