The commonly used thermodynamic theories (mean field theory and the square gradient theory) to predict interfacial tension between polymers have been modified. The results of these theoretical developments have not yet been fully tested and compared to experimental data. In this paper, experimental data for the effects of temperature, molecular weight, and molecular weight dispersity on interfacial tension for polypropylene/polystyrene polymer pairs are compared with the predictions of the new versions of the above theories. To evaluate these theories, it is necessary to know the Flory-Huggins interaction parameter for the polymer pairs studied. The relation correlating the Flory-Huggins interaction parameter to the Hildebrand solubility parameter was not suitable for evaluating the theoretical predictions of interfacial tension. Instead, the Flory Huggins interaction parameter was expressed as the sum of an enthalpic contribution, chi(H), and entropic contribution, chi(S). In the absence of reliable experimental values, a method was developed to evaluate the two contributions, based on interfacial tension data. The procedure provided an interaction parameter that is allowed to depend on molecular weight. When this approach was used, the predictions of only the new version of the square gradient theory were in good agreement with the experimental data for the influence of temperature and molecular weight on interfacial tension. However, the theory predicted the effect of polydispersity on interfacial tension only at high temperatures.