( 5fi9 ) 



tliis case occurs, the entectic ini'I(iiu/-jioint caire, furnished bv the system 

 A -\- B will generally present the same course. This case is rare. 



As, however, Bakhuis Roozeboom already ()bser\ed "), a negative 

 eutectir melting-point curve is also possible, when oidy the melting- 

 point curve of one siilislance is negative, provided the negative course 

 of one melting-point curve be stronger than llie ])osiiive course of the 

 other. To this lielong all cryo-hydrate lines. 



In the Z', 7'-prqjection fig. J it has been assumed (which, however, 

 is of minor importance here; that the negative course of the entectic 

 melting-iioini curve results from negative melting-point cur\es of the 

 substances ^4 and B. 



The particularity attencbug the negative course of the entectic 

 melting-point curve, is this, that a y/,.r-section corresponding with a 

 temperature below the euleclic point, will contain a region for ,S'^i -)- L 

 and a region for N/; -|- L, separated by a li(pnd r(\i;ioii L. The 

 limits of tins liquid reuion are given In soluhilily isdilicrms, whicli 

 according to van uku Waai.s' theory, are [xirlidus nf twd coulinudus 

 curves indicating the Huid phases wiiich can coexist willi the solid 

 substance A respectively A', and wliicii lia\e been called de .s()//^A/7//// 

 isotlierms. 



The regions for .b'.i -|- (i and N.i 4- L resp. S]j -\- ^' and Sn -f- /> 

 below the entectic point being separated by a region for ,S.i -\-Sjj, 

 the question which I wislied to solve came to this: "what is the 

 course of the t\vo solubility isotherms in the i-egioii foi' ,S'.i -|~'^/.' "• 



In order to answer this (piestion we lirst exanune what is tlie 

 y>„i'-seclion which corresponds with a tenqierature above the euteclic 

 point, but below the melting points of the two components. The 

 temperature whicii 1 have chosen for this purpose, is denoted bv 

 ?, in the /'. 7 -projection. The y^-.c-section corresponding with this is 

 re[)resente<l in fig. 2. As van uer Waals has proved that the s(tlu- 

 bility isolherni has two xcrtical tangents foi' the case r., <^ Vf . lint 

 only one vertical langeni for die case r» ^ r ^ two continuous soln- 

 bibty isotherms with one \erlical tangent have been drawn in lliis 

 /y-.r-section : for llie one solnliility isotherm this \ertical langent lies 

 at Ihc li(|iiid point A. an<i I'oi- the other at die vapour point (t. 

 We see further that the branches which separate the licpdd region 

 L from the i-egions feu- .S'^ -^ L and Sn -\- L diverge towards higher 

 pressure. The porlion of the licpiid-vapour-reiiion L -f- (r. which iiiav 

 be realized in slaltle condition, lies between the two three phase 

 pressure lines Sa^^L and S/; /. (_;. If we now examine a j(>-./-section, 



1) Loc. cit. p. 418, 



