( 897 ) 
such a way that not only minimum 7%, but also the intersection of 
2 
dx? 
found. If for such a case we draw a p‚v-curve for given 7, we are 
already past the maximum pressure at «= 0, and the pressure already 
decreases at v= 0. On a cursory examination we might think that 
continually increases. So the p,v-curve has the same shape and 
the same properties as for the case of fig. 40. But a difference appears 
in the neigbourhood of the minimum plaitpoint temperature; then 
the figures 41” etc. must be modified. Thus e.g. fig. 41/ is the shape 
of the righthand part at the moment of this temperature, but then 
there still exists a small closed branch starting from «=Q, and only 
at 7, which is higher than (77,,) minimum, this branch has entirely 
retracted into the axis. Then the 7,v-projection of the plaitpoint line 
has, indeed, the descending portion AQ,, which belongs to the little 
branch which retracts into the axis «—0O, but the three-phase-pressure 
has been modified in so far that the vapour branch, just as in fig. 40, 
possesses the smallest value of x, and that, therefore, the point of 
intersection of vapour and liquid branch does not occur at lower 
temperature. As for «=O we approach nearer to the z of the 
minimum plaitpoint temperature, the line AQ, is smaller and the 
value of 2, at which the splitting up of the spinodal line takes place, 
is smaller. Not until the initial value of # lies beyond the point with 
minimum plaitpoint temperature, we get back fig. 41¢ etc. The obser- 
vations made for the mixture CO, and urethane suggest the question 
if for this mixture the initial value perhaps coincides with (7,1) 
minimum. 
In the discussed cases the spinodal line always splits up into a 
righthand part and a lefthand part, and the splitting up takes place 
; ‘d'v av 
in that point of intersection of ( | = 0 and € = 0, which is 
an" Jy ® Jg 
denoted by A in fig. 44. As mentioned above the intersection 
‘of these curves in the point B never gives rise to splitting up 
of the spinodal curve, because the latter cannot pass through in 5. 
But the intersection in the point C can give rise to this, but then 
into what we may call an upper and a lower part. After the 
: 
d 
= 0 and (=) =0 takes place, and so incomplete mixture is 
v/a 
splitting up the lower part surrounds the space inside which zen is 
2 
d 
negative, and the upper part the space, inside which “is negative. 
But as a spinodal line can never intersect the curve 5 eeh 
