( 516 ) 



t^ig. 4. 



We now imagine temperature or pressure 

 changed a little in such a direction that the 

 two curves of fig. 3 move away from each 

 other. We then obtain fig, 4, in which the 

 line of saturation has been represented but 

 parti}'. It is clear that it now likewise has 

 to show two points of inflection. If both curves 

 move still farther from each other, then of 

 course both points of inflection can vanish. 



If we change temperature or pressure in opposite direction, then 

 we cause both curves of fig. 3 to overlap somewhat. We then find 

 fig. 5, in which haa'h' represents the binodal line and raa'r' the 

 line of saturation of F. 



On the part aa' not represented of the binodal 

 line lies the plaitpoint; the part aa' of the 

 line of saturation lies between the part aa' of 

 the binodal line and the straight line aa' . The line 

 of saturation of P is only partly drawn. p' 



As long as a and a' lie but close enough to 

 each other, ar and a'r' must lie as in fig. 5, 

 they must run namely from a and a' to that 

 side of line aa' , where the stable part of the ^'S- ^• 



binodal line lies. In their further course the lines ar and a'r' can of 

 course intersect the line aa' . 



We now have besides a series of solutions saturated with P {ar 

 and a'r') and a series of conjugate solutions {ah and a'b') also a 

 conjugate pair of fluids La-\-La' saturated with solid P. 



As the pieces aa' left out of the binodal line and of the line of 

 saturation lie inside the three-phase-triangle Paa' , the fluids repre- 

 sented by them separate into P -[- La -^ La' . 



Examples of lines of saturation with two points of inflection, 

 between which a curvature in the same direction as the binodal 

 line, we find e. g. in the system ^) : water-AgNOg ethylene cyanide. 

 At ± 11° the line of saturation of 2 C,H, (CN), . AgNO, . H,0 

 touches the binodal line in its plaitpoint and two points of inflection 

 appear as in fig. 3. 



With a rise of temperature both lines move away from each 

 other; the lines of saturation determined experimentally at 12°, 20° 

 and 25° show distinctly the type of fig. 4. 



1) W. MiDDELBERG. Z. f. Phys. Chem. 43. 305 (1903). 



