CONDENSATION AND CRITICAL PHENOMENA. 269 



On the left hand side of the diagram P is to the left of M, 

 on the right hand side P is to the right of M between M 

 and C. But evidently 4 is always higher than /p and /^ ; 

 even when P lies beyond C on the lower branch of the loop 

 as is sometimes the case. 



Fu/'^ 



By means of a diagram like fig. 4 we may predict 

 with some certainty what the influence will be of an 

 admixture of a substance a on another substance d. If 

 the vapour pressure curve for a is above the vapour 

 pressure curve for d the vapour pressure is likely to 

 be raised by the admixture. We should also be able 

 to say something of the influence on the critical con- 

 stants, / and /, if we were able to enunciate a rule about 

 the position and shape of the enveloping curve C^ P C^. 

 This however is not the case. The critical temperature 

 of a mixture is, as a rule, between those for the components, 

 but the critical pressure has a tendency to be high for the 

 mixtures and is often higher than for the pure substances. 

 But both these rules, as also the rule that the loops lie 

 inside the vapour-pressure curves, are far from general. 

 The pressures for the mixture are in some cases higher, 

 and in others lower than the vapour pressures for the 

 components. Mixtures of this kind have been investigated 

 by Konowalow, Guthrie, and others. The writer of this 

 article found a maximum vapour pressure for mixtures of 

 N2O and C2 Hg lying near jr = 1/5 N.O. This maximum 

 was found to exist up to the critical region for these 

 mixtures. Fig. 4 becomes complicated and, therefore, in- 

 convenient in cases of this kind. As to the critical tempera- 



