a4G 



Prof. J. P. Kuciien on the 



Table III.— Propane and Methyl-alcohol. 



Critical points of liquids. 



Temp. 



1 

 Press. ! 



21]o 



10 



20-85 



11 



20-8 



13 



2005 



23 I 



19-85 



20 [ 



19-4 



34-5 I 



19-2 



39 \ 



190 



40 



Temp. 



18-6 

 18 05 



70 

 79 

 82 



85 

 93 

 95 



(100) 



Comparing this case with that of ethane we find the chief 

 difference to be, that there the critical point between vapour 

 and upper layer was reached before the liquid-curve had 

 contracted into the vapour-liquid curve, here on the other 

 hand the liquid-curve disappears far below the critical tem- 

 perature of the top layers, otherwise the two cases seem very 

 similar in character. 



It was mentioned above that a consistent application of 

 van der Waals^ theory enables us to consider the character 

 of the saturation-curves or branches of these in the non-stable 

 conditions, t. e. inside other curves. This may be illustrated 

 by some observations made with the mixtures under con- 

 sideration. The liquid-curve disappears from the stable part 

 of the diagram at 21°*15, but as explained it may still be 

 considered to exist inside the vapour-liquid curve (fig. 4). 

 With some diificulty I succeeded in showing this view to be 

 legitimate. At a temperature of ?1^*4, i. e. just above the 

 critical point of the liquids, the mixture was completely 

 liquefied and then made to expand very slowly. It is w^ell 

 known that in those circumstances, by what is called tbermo- 

 dynamical retardation, the vapour does not ahvays appear 

 when the saturation-pressure is reached, but that the mixture 

 often remains homogeneous in metastable equilibrium. The 

 same phenomenon took place in this instance, but it was 

 observed that under those circumstances the liquid became 

 and remained turbid, indicating the formation of a second 

 liquid layer ; by a shock or by further expansion the vapour, 

 was suddenly formed, the mixture stirred up and the turbidity 

 disappeared. The phenomenon can be explained by assuming 

 that in the metastable condition the completely metastable 

 liquid-curve had been reached and two metastable liquids 

 formed, entirely in accordance with the nature of the diagram. 



It follows from the fact that the three-phase pressure is 



