Young — Vapour-Pressures, ^-c, of Thirty Pure Substances. 409 



reappears on cooling ; (2) that tlie temperature depends on the position of 

 the meniscus in the tube at the moment of disappearance — that is to say, 

 on the mean density of tlie substance ; (3) that the temperature at which 

 the meniscus disappears is not the same as that at which the densities of 

 liquid and saturated vapour become equal, or at which the heat of vaporization 

 vanishes. These statements have not been verified. 



The Critical Pressure. 



This has in every case been determined by means of the modified Andrews 

 apparatus. Tlie pressure is read when the meniscus has just disappeared, but 

 can be made to reappear momentarily by suddenly very slightly increasing 

 the volume. Strictly speaking, the position of the meniscus before it 

 disappears should be midway between the upper surface of the mercury and 

 the top of the tube; but the fluid is so compressible that its volume may be 

 altered to a moderate extent without altering the pressure more than two or 

 three millimetres. 



21ie Griticdl Volume. 



It was thought for a time that the critical volume could be directly 

 determined by causing the disappearance of the meniscus to take place just 

 at the top of the tube. It was, however, pointed out by Grouy' that, owing 

 to the extreme compressibility of a substance at its critical point, the density 

 of the column of fluid varies very considerably at different levels. The 

 small hydrostatic pressure of the fluid above a given level is quite sufficient 

 to cause a considerable compression of tiie fluid below that level. The true 

 critical density is the density of the fluid at the level where the meniscus has 

 disappeared. Above that level the density is lower ; below, it is higher. 

 The mean density is equal to the true critical density when the meniscus 

 disappears at about the centre of the column. 



For this reason the earlier determinations of the critical density^ are 

 incorrect. The critical densities of the substances already investigated were 

 afterwards recalculated by the method of Cailletet and Mathias,^ on the 

 assumption that the diameter is always strictly rectilinear — that is to say, by 

 means of the formula 



St = S(^+ at; 



and this method was adopted for the majority of the substances subsequently 

 investigated. 



1 Compt. rend., cxv., p. 720, 1892. 



-Phil. Mag., (5) xxxiii., p. 182, 1892. 



3 Compt. rend, cii., p. 1202, 1886 ; civ., p. 1.563, 1887. 



