The isothermal compressibilities KT for cyclohexane + benzene, cyclohexane + toluene and benzene + toluene systems at 25, 35, 45 and 60°C have been used to test the Prigogine-Flory theory using Van der Waals and Lennard-Jones energy potentials. Flory's energy parameter X 12 was calculated for these systems at the four temperatures. From X 12 for the equimolar mixture, the following excess functions were calculated: (∂VE/∂p)T which is related to K T E , the heat of mixing H E , and the excess volume V E . The theory and any of the two potentials give (∂VE/∂p)T which fit the experimental data, but H E and V E , calculated using the same X 12 parameter, depart appreciably from the experimental data even though they agree in sign and have the essential features of the excess functions. The departure is apparent in both magnitude (in particular for the cyclohexane + benzene, and cyclohexane + toluene systems) and in the temperature dependence. The conclusion is that the X 12 parameter does not predict the thermodynamic properties of these systems and the Lennard-Jones potential, involving a more complicated expression, does not contribute any improvement over the Van der Waals potential.