A model has been suggested by Raijinder Pal (J. Colloid Interface Sci. 231 (2000) 168) that describes experimental data on the viscosity of concentrated emulsions reasonably well. To deduce the final equation the author assumed that droplets are covered with a layer of surfactant molecules. This means that the effective volume of a single emulsion droplet is increased by a factor K > 1. It has been found that K should vary between 1.166 and 2.070 to fit the experimental data. If the drop radii are around I pm (for an estimate) then the drops would be covered with a layer which thickness should range from 550 A (which is 10 times of the size of SDS micelles) to 3570 A. No doubt adsorption of surfactant molecules results in an increase of the effective drop radius but not by that much. We present an alternative theoretical model for description of the viscosity-volume fraction of droplets behavior of emulsions. The model is based on the assumption that clusters of drops (doublets, triplets, and so on) are formed under the influence of colloidal, hydrodynamic interactions and/or applied shear. According to the proposed model clustering determines the volume fraction dependency of the effective viscosity of emulsions. Two limiting cases are considered: a developed flocculation and a low-flocculated emulsion. In the first case the final equation is close to that deduced by Pal; however, the physical meaning of the equation is substantially different. Comparison with available experimental data shows good agreement with the equations deduced in both the cases of a developed flocculation and a low-flocculated emulsion. (C) 2003 Elsevier Science (USA). All rights reserved.