( 242 ) 
passes gradually into a coexistence vapour-liquid, till B, the critical 
point of the second component, is reached. 
If now in this case splitting up of a plait takes place, we might 
have the case of fig. 4, analogous to fig. 2. Here, too, we might 
distinguish two sub cases 4 a and 46, analogous to 2 a and 26. And 
also a case like fig. 5, would be possible. 
It is not easy to state in a single word in what relation the clas¬ 
sification adopted here stands to that of Buchner and Van Laar, 
because in our opinion by the latter cases have been classed together 
which should be sharply distinguished, and on the other hand what 
belongs together, has been separated. For Buchner’s type II, Van Laar’s 
type III l ) comprises the cases of retreat and splitting up, which 
are entirely different both theoretically and experimentally. Buchner’s 
fig. 10, Van Laar’s figs. 11 and 2 a may therefore be found in our 
figs. 1, 2 a and 26. On the other hand Buchner’s types I and IIJ, 
van Laar’s I and II (first paper) are distinguished only in this, that 
there are one maximum and one minimum of the pressure more in 
the last plaitpoint line, a circumstance, which is not of great impor¬ 
tance physically, as little as the occurrence of a maximum or a 
minimum of the pressure in the ordinary plaitpoint line is generally 
considered to be of importance (in opposition to a temperature 
minimum or maximum). We find these two types, therefore, com¬ 
bined in our fig. 3, if we consider that F may lie considerably below 
A, and bear in mind what has been said above about the assumption 
of open plaits. There would be an essential difference between fig. 6 
and 13 of Buchner, fig. 1 and 7 of van Laar’s first paper only when 
the line KPD (fig. 13 loc. cit.) was imagined to show a temperature 
minimum and maximum in its further course; then fig. 13 resp. 7 
would agree with fig. 5 here. Neither Buchner nor van Laar, however, 
mentions this. 
In van Laar’s later paper type II has been modified and has become 
different from I in so far, that now not one pair of heterogeneous 
double pinpoints occur, but two pains (resp. one pair and an open 
plait, which comes to the same thing). Such a complicated system 
of plaits cannot be arrived at on the simple suppositions of van Laar, 
: remarkable that van Laar, who on p. 669 of his first paper decidedly 
e difference in the systems with positive and with negative dp/dt, does 
i word about it in his later paper. In connection with this the dotted 
?• 11 of the former paper, which refers to positive dpjdt has been left 
e figure of the later one, probably because the special suppositions 
^possible for normal substances. 
required homogeneous double plaitpoints 
