(901 ) 
Though these equations hardly require a proof, I may point out 
that they can also be directly derived from: 
dp\ _ dp dv de dp 
de bin dv NOR) bin OD Sy 
and 
d? d d?, dv d?) 
EEE eg, See 
ra ay da bin da dv \ daz /bin dx’ 
As 
dp dp 
dv 7 da 5 
and 
dp dp (dv dp (dv d*p 
2 —— en 
A e 5 E Joa i: da dv (G)+ ze | 
If == ==, (3) TEK au must be =. 0; or 
de ) bin ; dv ) dx bin da z 
dv dv dp dp dv d’v 
ENGEN, 
mep 0. tl gen =(a) d the Jatt tit 
— — |= : as the latter 7 
ie Saas en ee an er quantity 
2 
1b) 
is zero in the point C also ( :) == ()) 
dx? ) bin 
In the point for the equilibrium liquid-vapour, in which the two 
d pn 
phases have equal concentration, (2) is indeed = 0, but (32 3 is 
US bin 27) bin 
dv dv : 7 
negative. For this point =| — |; but taking into account that 
de bin \de p 
PY | omeen mont ()> it: 
5 1S negative, we fin Or 18 pom da? ‘a de? k 
But at the same time we may now also draw attention to the 
following important property. The point in which for liquid and 
vapour the concentration is the same, so where (2 ee 0, or p 
wn 
maximum, lies at smaller value of wv than that of the point C. This 
property holds certainly if we may assume it to be of general 
validity that if a p-line touches a binodal curve of a plait, the curva- 
ture of this p-line is always such that the plait lies on its concave 
side. For a plaitpoint this is certainly the case. But the question 
is whether we may also accept it as valid for that point of a binodal 
el , ea dv 
line in which the value of p is maximum. Then (5) and a fortiori 
wv 
p 
