( 187 ) 



of the llii-ee-phase line, so tliat now the pressure-minimum has got 

 into tlie stable region ; the pressure-maximum., on the other hand, 

 has disappeared, and now llic meeting with the three-phase line takes 

 place in a point where iü,f is positive, from which follows that 

 iVsf has passed throngh zero on the three-phase line; and if we now 

 consider that this holds for an anthraquinone concentration which is 

 smaller than that corresponding to the second critical end-point q, 

 it follows from this that the reversal of sign of w..j takes place on 

 the vapour-branch of the second three-phase region. 



If we then proceed to the concentration 2.6 mol. "/o' ^'^^ see that 

 the line for solid-fiuid has mainly the same shape as that of the 

 concentration 1,9 mol. Vo, b"t it has moved pretty much upwards 

 and to the right, the pressure-ma.vimum having become less deep. 



For the following concentration of 3,6 mol. 7o the latter takes place 

 in a far greater degree, and for 4,2 mol. "/„ the minimum is only 

 just present, and for 5,7 mol. " „ i^ h^s quite disappeared. So we 

 see from this that the meeting with the three-phase line for this 

 concentration takes place in a point where lUg/ is negative, so that 

 we come to the conclusion that the quantity lUgf has passed through 

 zero for the second time, and as the concentration of the second 

 critical end-point has not been reached as yet, it follows from this 

 that all this takes place on the vapour branch of the second part of 

 the three-phase region. 



That this is a necessity is at once to be seen when we consider 

 that the line for solid-liquid possesses horizontal tangents in two 

 points, so two places where zys/=0, which will lie inside the coniiodal 

 line for systems without critical end-points, but which here have got 

 outside it for a series of concentrations lying between p and q. 



Now we know that for the concentration of the second critical 

 end-point q : 



rj. ^ ^ f^P_\ _ !V 



dTsLG \dTsfJx Vsj ' 



dp 

 and 1 — — being negative for the system ether-anthraquinone, 



d-l SLG 



lUsf will also be negative in the point q. 



We know further that this point q must lie on the upper branch 

 of the line solid-liquid, from which follows that before we have 

 reached the second critical end-point, first the maximum, and then 

 the minimum will have entered the metastable resp. unstable region, 

 in consequence of which Wgf must twice pass through zero on the 

 vapour branch of the border line. 



