316 Proceedings of Royal Society of Edinburgh. [sess. 
coincides at R with the coefficient of absorption in the liquid, 
which, as we saw, is on the increase in the critical region. Obvi- 
ously then the latter does not approach infinity. 
The same result would have been arrived at if we had considered 
the gas absorbed in unit volume instead of in unit mass of the 
liquid, hut we could not in that case have used the diagram which 
is drawn for mixtures of constant composition. 
It might be tried to use the coefficient for a mixture in the 
homogeneous condition — e.y., above the critical point, i.e., to the 
right of the critical curve ; at moderate pressures the approximate 
formula 
m _d 
V d\ 
still holds, but d x is not now a constant as it was for saturated 
vapour but is proportional to the partial pressure of the 
vapour ; by changing the amount of the solvent we may under 
these circumstances give the coefficient any value we like. In 
this case it would be better to consider the gas dissolved in unit 
volume. The formula then becomes 
which gives an approximately constant value for the coefficient at 
a given temperature. But in any case no special advantage 
attaches to the use of the term in this case, and it seems more 
appropriate to reserve it for conditions of equilibrium between a 
vapour and a liquid. 
