( 459 ) 



Physics. — ''Contribution to the knowledge of the xp-surface of 

 VAN DER Waals. X[. A gas that sinks in a liquid." By Prof. 

 H. Kamerlingh Onnes. (^omniuiiicatlon N". 96 from the 

 Physical Laboratory ot Leiden. 



If we have an ideal gas and an incompressible liqnid withont 

 vapour tension, in which the gas does not dissolve, the gas will gather 

 above the liquid under the action of gravity, if the pressure is suffi- 

 ciently low, whereas the compressed gas will sink in the liquid if 

 the pressure is made high enough. 



I have observed a phenomenon approaching to this fictitious case 

 in an experiment which roughly came to this, that helium gas was 

 compressed more and more above liquid hydrogen till it sank in the 

 liquid hydrogen. Roughly, for so simple a case as was premised is 

 not to be realized. Everj^ experiment in which a gas is compressed 

 above a liquid, is practically an application of the theory of binary 

 mixtures of van der Waals. Li such an experiment the compressi- 

 bility of the liquid phase and the solubility of gas and liquid inter 

 se may not be neglected, as generally the pressure will even have 

 to be increased considerably before the density of the gas-phase 

 becomes comparable with that of the liquid phase. 



If the theory of van der Waals is applied to suchlike experiments, 

 the question lies at hand whether in the neighbourhood of the plait- 

 point phenomena where gas and liquid approach each other so closely 

 that of the ordinary gas and liquid state they have retained nothing 

 but the name, perhaps on account of a higher proportion of the 

 substance with greater molecular weight ') the phase, which must be 

 called the gas phase, may become specifically heavier than the phase, 

 w-hich must be called the liquid phase. On closer investigation it 

 appears however, to be due to relations between the physical proper- 

 ties and the chemical constitution (so also the molecular weight) of 

 substances, that a liquid phase floating on a gas phase has not been 

 observed even in this favourable region. 



I was the more struck with an irregularity which I came across 

 when experimenting with helium and hydrogen in a closed metal 

 vessel, as I thought that I could explain it by the above mentioned nol 

 yet observed phenomenon, and so the conviction took hold of me, 

 that at — 253° and at a pressure of 60 atmospheres the gaslike phase 

 which chiefly consists of helium, sinks in the liquid phase which 

 chiefly consists of hydrogen. 



1) The limiting case is that in the Jz-surface construed with the unity of weight 

 the projection of the nodal chord on the zf-plane runs parallel to the line v = 0. 



30 

 Proceedings Royal Acad. Amsterdam. Vol. VIII. 



