(aah) 
The minimum in the critical temperature vanishes again from the 
ry. 
: eke AV > dT, : : 
region which has physical signification, becoming negative for 
adt 
v=d oo. We have now decrease of the critical temperature with 
increase of the size of the molecules. It is again unnecessary to 
give a separate diagram for 7 and pj}, as the required figure is 
obtained by revolving fig. 11 180° round an axis e—C. At low 
: : : ; dp dp 
temperature there is no intersection of —=0O and — ==0; we have 
Av Av 
27a „2 
the course of fig. 10. At the temperature —- is 
Sane 
intersection arises, so we get again fig. 19, which passes again into 
12 
a point of 
ly 
the left part of fig. 13 and fig. 14. The detaching of sand 
at 
dp 
rn now always takes place according to fig. 9 in contrast to the 
U 
other case mentioned for which fig.19 holds, in which case either 
fig. 9 or fig. 8 can hold. 
5 and 6% cases. If we think the a-line displaced still more 
to the left, we get for the course of pj and 7), the figures originating 
by turning fig. 22 and fig. 20 180° round an axis «= €. For the 
region of physical signification this course is the same as that of 
the 4 case, and this holds also for the isobars. 
20. If we make «,, increase a little more without making this 
quantity reach the value 5 (a, + a,) the points a, and a, will move 
further apart. If the distance between them has become greater than 
that between #4, and 6,, it is clear that the situation of the third 
and the fourth ease of the preceding $ is no longer possible. Instead 
of fig. 22 we get then fig. 24, which, however, presents the same 
thing as fig. 22 in the region on the right of a,, and so leads to 
the same results. A new feature, however, appears when we think 
the a-line still somewhat more displaced to the left, or in other 
words if we think fig. 24 rotated over 180°. For the minimum of 
‘yy 
. . . . C C . , 
a now lying on the left of that of 4, becomes negative for 
Av 
e£=-+o, and so a maximum of 7, must appear, a circumstance, 
which I think had up to now been always considered as insepa- 
rably bound to a negative value of a, + a, — 2a,,. 
Of course the diagram of isobars will undergo a modification also 
now; it will constitute the transition to the figure holding for a 
