Molecular diffusion in multicomponent systems has been the subject of many theoretical studies. There is a problem with many of these approaches in that they may violate a material balance constraint and lead to non-physical predictions. Theories by Zielinski and Hanley, Alsoy and Duda, and their generalization by Price and Romdhane have been tested. They performed adequately at elevated initial concentrations of the balancing component but fail to satisfy the material balance at reduced concentrations. The inclusion of the balancing component self- diffusivity does not eliminate the problem.