120 
PROFESSOR W. RAMSAY AND DR. S. YOUNG 
it may be surmised that in the act of combination of ammonia with hydrogen 
chloride, the molecule of hydrogen chloride is not split but combines as a whole. 
Should this conjecture not be accepted as correct it will be necessary to seek for an 
explanation of the phenomena observed, by some relations yet to be discovered. 
Should it be accepted, the classification of those substances in which an evident 
molecular rupture takes place, suggested in § 77, may still be adopted ; and it is 
probable that an idea of the amount of dissociation may be gathered from the relation 
of the two curves. 
§ 83. Some points still remain to be discussed. 
Vapour-'pressures of substances undergoing partial dissociation .—A substance 
capable of exerting vapour-pressure, and at the same time capable of dissociating, 
exerts a pressure which is the sum of the vapour-pressures of the undecomposed 
substance, and of the bodies resulting from its dissociation. If these two pressures 
could be determined separately by experiment we should have two separate curves, 
the resultant of which would be that representing the pressure determined by the 
barometer-tube method. 
The curve representing the true vapour-pressure of the dissociating body would, to 
begin with, show increase of pressure with rise of temperature. But at a certain 
temperature pressure must decrease with rise of temperature, for if the temperature 
be sufficiently raised dissociation is complete, and there is no true vapour-pressure. 
The curve must, therefore, have the general form A, fig. 6 or fig. 7. 
At the same time the products of dissociation exert a pressure which increases more 
and more rapidly with rise of temperature (as in B, figs. 6, 7, 8). Now the combina¬ 
tion of two such curves might yield a curve of double flexure, as shown in C, fig. 6 ; 
