321 
in a. If in 6G gets the same composition as #, the curve / touches 
the limit-curve Hrp—F-+G, (consequently the sublimation-curve of F’) 
in 6. If it is possible for Z and G to get the same composition in 
a point c, the curve Z then touches the limit-curve Hg = L + G 
inc. 
The known conditions, under which pressure or temperature are 
stationary along curve / (maximum or minimum) follow immediately 
from (7). 
When one of the phases e.g. /, is a gas, then in general V, is 
very large in comparison with V, V,.... We now take on curve Ha 
point a, where 2, = 0. Consequently curve / touches in this point 
a the limitline of the region F’. In this point a is: 
Aa sae ote gs Vea Ol. er 2 (L8) 
and in general different from zero. 
In the vicinity of point a however 4, is no more zero and for 
ZV) the value from (8), not that from (13) remains true. V, being 
large in comparison with V, V,..., (AV) in (8) may be zero for a 
small value of 4, already. When 2(4V )—O then, according to (7) 
dT =O, consequently the temperature is maximum or minimum. 
Consequently we find: “when of the equilibrium 
E=F,+ F, d-.... + Fi 
“one of the phases e.g. F, is a gas the maximum (or the minimum) 
“of the temperature of curve / is situated in general in the proximity 
“of the point of contact of this curve with the limit-curve 
Pps ae et 
“of the region (F').” 
When we apply this to the equilibrium 4 = F+ L + G, discussed 
above, it is apparent, than this curve / has its point of maximum 
temperature in the vicinity of its point of contact with the melting 
line of F. We may also express this in the following way : the point 
of maximum-temperature of the equilibrium # + L + G differs only 
slightly from the melting-point of / under its own vapour pressure. 
When we consider the ternary curve H= F, 4+ F,+ L-+ G, it 
follows that this curve must have its point of maximum tempera- 
ture in the proximity of its point of contact with the limit-curve 
F, 4 F,+ L, the common melting curve (or inversion-curve) of 
F, + F,. We may also say: the point of maximum temperature of 
the equilibrium F,+ F, + LG differs only slightly from the 
common melting-point (or point of inversion) under its own vapour- 
pressure of #, + F,. 
When we imagine the solutions of this equilibrium Z to be repre- 
