( 829 ) 
ae . dp p 
and 1 coincide, there is again a point where aan O and Pees) 
Av & 
in the righthand part, but in contrast to what took place before this 
point now lies on the lower branch. At a temperature somewhat 
higher than that of the branches getting detached, there are now 
again values of we, for which a line normal to the a-axis has four 
points of intersection in common with the two detached branches. 
At lower pressure we then meet with phenomena of condensation 
for the equilibrium 1,2 and at higher pressure for the equilibrium 
3,2. Then there is homogeneity between these pressures. 
If we also inquire into the course of the 7'v-projection of the 
plaitpoint line and the three-pbase-line in this case, a difference can 
immediately be pointed out compared with fig. 40 that on the left 
side the vapour branch does not belong to smaller but to greater 
value of w than the liquid branch. I have drawn a schematical 
representation of the two lines in fig. 45, assuming that the three- 
phase-pressure does not continue to exist as far as 7'=0. The full 
line which begins in A, descends to Q,, rises again to #7, then 
descends to PP, after which it rises to the critical point of the 
second component, is the plaitpoint line. So compared with fig. 40 
the ascending part AP of this figure has still a minimum, and 
further it has been considered as possible that the descending branch 
of this figure has a maximum and a minimum value of x. The part 
AQ, contains the realisable plaitpoints of the branch which retracts 
into the w-axis, after it has got detached from the righthand part 
of the p‚v-lines. At the temperature indicated by Q, the two parts 
join, which, however, may not be called contact. Then the spinodal 
curves and the binodal curves intersect in one point, whereas above 
Tg, these lines remain at a distance from each other. So here the 
same circumstance is met with which occurs for mixtures with 
minimum 7, but for another value of 7’ and w. With rise of the 
temperature from below 7’, to above it two realisable plaitpoints 
appear. One of them was mentioned before, but the other lies in the 
considerably larger righthand part. Though we have called it realisable, 
it does not show itself but remains covered under the more stable 
equilibrium 3.2. If phenomena of retardation could appear, it would 
be realisable a circumstance which always occurs if splitting 
up of the spinodal line takes place for three-phase-equilibria. These 
plaitpoints lie on the branch Q, P.a. 
For the discussion of the remaining part of the plaitpoint line we 
shall begin with P,,. At the temperature of this point an heteroge- 
neous double plaitpoint arises and with rising temperature these plait- 
56 
Proceedings Royal Acad. Amsterdam. Voi. XL. 
