CONDENSATION AND CRITICAL PHENOMENA. 263 



hours or perhaps even longer. In diminishing the pressure 

 the same phenomena are observed in the opposite order 

 {h, g f e). The higher the temperature the smaller is 

 the maximum quantity of liquid, and at about 19° C, a 

 minute quantity of liquid only just appears but disappears 

 again immediately on compression. Though 19' C. may be 

 called the critical temperature, because above 19" C. no two 

 phases are possible, no " critical phenomenon " takes place 

 there. If v^e follow the phenomena at somewhat lower 

 temperature, say 17° and 16°, we find similar results; but 

 the maximum quantity of liquid increases, while the liquid 

 surface becomes less and less distinct. At last the liquid sur- 

 face actually disappears, first near the bottom of the tube, and 

 the lower the temperature the higher is the level at which it 

 does so. At 15*6" C. it disappears in the middle of the tube, 

 and it is below 15° C. that the liquid increases and fills the 

 whole tube before the liquid surface disappears. The 

 critical phenomenon is therefore confined to temperatures 

 between about 15° C. and 16^ C. From the theoretical 

 explanation to be given below it will appear that the 

 critical phenomenon really belongs to one temperature 

 only, in this case 15 "6"' C. ; a temperature which we shall 

 denote as t^ The fact of its occurrence at a range of 

 temperatures (i5°-i6° C.) is due to gravitation : the explan- 

 ation is analogous to that given in part i. of the occurrence 

 of the critical phenomenon for pure substances, at different 

 volumes. In this case gravitation not only makes the 

 density different at different levels but also the composi- 

 tion. As /p depends on the composition x, i.e., is different 

 for different mixtures and, therefore, not the same for 

 the mixtures at different levels in the tube, the critical 

 phenomenon does not take place at the temperature t^ 

 belonging to the mean composition x (here 56) only, but 

 also at temperatures slightly below and above /p_ each time 

 at a different level in the tube. The longer the tube the 

 wider the range of temperatures within which the critical 

 phenomenon occurs. All this follows from the theory and 

 is confirmed by experiment. If gravity did not act we 

 should find normal condensation below 15 "6" C, the critical 



