Two Components 
89 
with solutions containing more and less chlorine than the crystals 
themselves. This, as well as the place occupied by ICI /3 in the 
equilibrium, will be clearer if we consider the compositions of the 
different solutions as shown in the concentration-temperature dia 
gram, Fig. 16. One hundred and twenty-seven grams of iodine 
are equivalent to thirty-five and one-lialf grams of chlorine. The 
composition of the vapors is represented as far as possible. The 
letters have the same significance as in the preceding diagram. The 
curves for the vapors have the same letters as the corresponding so 
lutions, but underlined to distinguish them. The melting point 
of IClar is found to be at 27.2 0 , KGE is the solubility curve for IC 1 / 3 , 
this compound melting at 13. g°. The curve represents a state of 
labile equilibrium at all temperatures. There is a labile •nonvariant 
system possible at K, 0.9 0 , and another at L, 12.o°. At the first 
there coexist I 2 , ICI/ 3 , solution and vapor; at the second, IC 1 / 3 , IC 1 3 , 
solution and vapor. The continuation of the curve EF represents 
the solutions in equilibrium with IC 1 3 , and containing more chlorine 
than the crystals. This curve if prolonged terminates at about -102°, 
when solid chlorine begins to separate from the solution. There 
seems to be no question from the experimental data that the two 
branches of the solubility curve for icin', BD and ED meet at an an 
gle, but this is denied by Stortenbeker on the strength of a thermo 
dynamic formula by van der Waals. In view of the tacit assump 
tions which have been so often discovered in thermodynamical for 
mulas, this conclusion can hardly be considered a wise one, especial 
ly as we are not yet in a position to say at what point iodine ceases 
to be solvent and becomes solute, nor what effect this will have on 
the equations of equilibrium. In this particular case it is assumed 
explicitly 1 that the solution consists of a mixture of iodine and chlo 
rine, and that there is no fused monochloride present. This may or 
may not be true. There is no way at present of determining it. In 
Tables XVII-XVIII are the experimental data for this system. The 
pressures are given in millimeters of mercury. x l denotes units of 
chlorine per unit of iodine ; x 2 , units of iodine per upit of chlorine ; 
x 3 , units of chlorine in one hundred units of solution ; y, units of 
chlorine per unit of iodine in the vapor. The values of y are only 
approximative. 
1 Zeit phys. Chem. 10, 194 (1892).