G50 Prof. J. P. Kuenen on the 



Pressure therefore raises the critical point, and after sepa- 

 ration of the liquids by warming, they can be made to mix 

 by applying pressure. It follows that the liquid saturation- 

 curve is closed on the side of the small volumes, as in fig. 4, 

 and has a critical point on that side ; at 18'^'3 it emerges out 

 of the vapour-liquid saturation-curve, below that temperature 

 it is completely hidden in the non-stable part of the diagram. 



In this case the two-liquid curve has the additional peculi- 

 arity of being very broad and flat, the coexisting phases 

 dissolving each other to a ver}^ small degree only * ; the 

 three-phase pressure may therefore be expected to be but 

 little smaller than the sum of the vapour-pressures of the 

 components taken separately. In order to test this conclusion 

 the three-phase pressure was measured at 92*^'^) C and found 

 to be 142-6 cms. mercury; the pressure of triethylamine is 

 not accurately known and I had not got a sufficient quantity 

 of the substance left to measure it, but the boiling-point of 

 triethylamine is about 89° C, and from this we may infer that 

 at 92°' 9 C. the vapour-pressure must be about 86 cms., while 

 the pressure of water-vapour at that temperature amounts to 

 58*6 cms. Hence the sum of the pressures comes out at 

 144*6 cms. as compared with 142-6 cms. for the mixtures ; the 

 difference as expected is very slight. 



i. Diethylamine and Wafer. 



Several vain attempts were made to confirm Guthrie's 

 observation t of a lower critical temperature for these 

 mixtures at about 120°('. The diethylamine which was used 

 in the experiments had been kept for a considerable time, 

 probably some years, and I have not been in a position to 

 repeat the experiments with a fresh sample, but I know no 

 reason why this fact should have completel}- prevented the 

 phenomenon from showing itself. It must be added that 

 the description by Guthrie of the behaviour of the mixture 

 seems to make the whole phenomenon doubtful; Guthrie^s 

 words are the following : "It is to be noted that, ov\ing in 

 part to the refractive index of diethylamine being nearer 

 that of water than is the refractive index of triethylamine, 

 the above separation is not marked by the milkiness which 

 characterizes the separation of the latter body. If it were 

 not for accidental dust particles floating on the surface of the 

 lower liquid, it would be, in some cases, almost impossible to 

 distinguish the two layers." Now it seems impossible for a 

 homogeneous liquid to separate into two layers without 



* Eothinund, /. c. p. 461. • 

 t Cuithrie, /. c. p. oOO. 



