212 SCIENCE PROGRESS. 



the liquid and vapour, in the order in which they lie on the 

 isothermal. It is impossible to explain here how the con- 

 dition of equilibrium is arrived at. Suffice it to say that 

 the theory is thermodynamical. Though among these 

 phases there is a group of phases necessarily unstable when 

 supposed to exist over a finite space, V^an der Waals proves 

 that the layer as conceived by him, in which each of the 

 phases exists over an infinitely small distance, is stable as a 

 whole. According to this theory the continuity of the two 

 states within the border-curve would not only be an imagi- 

 nary, a theoretical one, but a real one, and in a seemingly 

 discontinuous surface between two phases we should have 

 a special kind of continuous transition. This is the widest 

 possible generalisation of the dogma of the continuity of the 

 two states. The more one considers the matter the more 

 one is struck by the beauty of the theory, and the remarkable 

 simplicity of the conception of the continuity which it contains. 



The application of the theory to a curved surface leads 

 to some interesting results. The condensation-pressure 

 appears to be different in that case to what it is at a plane 

 surface, a fact which a direct application of thermodynamics 

 had already revealed. The pressure is higher in one of the 

 phases than in the other, and both differ from the ordinary 

 condensation-pressure given by the height of the straight 

 lines in Andrews' diaaram, being- either both hiMier or both 

 lower. 



It is also found that possibly a continuous transition 



exists above a certain temperature only, below which the 



transition would have to be discontinuous ; and a number of 



other things as well. The theory leads to a law giving the 



change of the capillary constant a with temperature ^ near 



/ / \ ?". 

 the critical temperature 4) ^^-S"-, cr = f (i- — j " in which 



^ is a constant and m would be i^ close to the critical 

 point. This law was tested with the experimental results 

 obtained by De Vries and Ramsay and Shields ; it appears 

 that at some distance from the critical temperature w is a 

 constant quantity, but smaller than i^, about 1*25 for most 

 substances. The only exceptions are the same substances 



