CONDENSATION AND CRITICAL PHENOMENA. 259 



pressure at io°C. and therefore constant, but the pressure of 

 the air becomes higher and higher, and the total pressure 

 must therefore also rise with compression. Taken in this 

 way the problem would be one of the absorption of a gas 

 by a liquid, and of the validity of the laws of Dalton and 

 Henry. But these laws, though capable of describing the 

 phenomena in a few cases and under special conditions 

 which need not be stated here, are as a rule inade- 

 quate, and more especially so the nearer one gets to the 

 critical point. The problem must therefore be taken up 

 in a different manner and the first step is to draw diagrams 

 for mixtures in the same way as this was first done by 

 Andrews for single substances. The changes of pressure 

 and volume for our mixture at io°C, are given diagrammat- 

 ically in fig. i by the isothermal /= io°C. At d the conden- 



Fi6/J 



sation begins, at e it is finished. The curve shows the rise 

 of pressure in the gaseous state, a break at /?, the rise of 

 pressure between d and e, another break at e and a very 

 steep curve beyond e in the liquid state. Curve ^ = 25°C, 

 shows the behaviour of the mixture at 2 5°C. This curve has 

 no breaks. The mixture does not show any condensation 

 and must evidently be above its critical temperature. Ex- 

 periments have shown that the mixture cannot be con- 

 densed at temperatures higher than about I9°C. This 

 temperature may therefore be called the critical temperature 

 of the mixture, and it appears that the addition of one fifth 

 of air has lowered the critical temperature considerably. In 



