( 184 ) 



these lines possess two liorizontal and two vertical tangents in normal 

 cases, and with regard to tig. 1 we maj now observe that as we 

 have assumed to be near the concentration of a critical end-point, 

 Vgf is positive not only on the connodal line, but also far outside 

 it, so that the liquid branch starting from L and the vapour branch 

 starting from G possess vertical tangents in the stable region. 



If we now go to greater concentrations of anthraquinone, the 

 border curve moves more to the right with respect to the line solid- 

 fluid, i.e. towards higher temperature. Moreover both curves move 

 upwards, so towards higher pressure, but the left-side branch of 

 the line solid-fluid moves quickest in this respect, so that the left- 



hand meeting changes into contact at a certain concentration, after 

 which this contact is quite broken off. 



Before this takes place, however, the discussed critical end-point 

 appears, for the plaitpoint moves to the right, and when this at 

 last has coincided with the point G, the end-point of the first three- 

 phase pressure line (Fig. 1), this implies that the liquid- and vapour 

 phase coexisting with solid anthraquinone, have become identical. 



This case, which is indicated by fig. 2, presents also this peculiarity 

 that the three-phase line HLK and the line for solid-fluid Z)Zil/i 6^16^2 

 touch in K, for as van der Waals proved : 



dTsLG \dpsf. 



holds for this point. 



