This study reports on the development and evaluation of a predictive model for the accumulation of polychlorinated biphenyls (PCBs) in phytoplankton which incorporates the kinetics of the process. The model includes a surface sorption term, uptake and loss rate coefficients, and a biomass dilution term. Laboratory data collected from the accumulation of 40 representative PCB congeners in four representative algal species were used to parameterize the model, and the performance of the model was evaluated on a set of independent laboratory data and a set of field data collected from Green Bay, Lake Michigan. Model predictions were compared to predictions from an equilibrium model. Under low growth conditions, the predictions of the kinetics model and the equilibrium model were similar. However, for data collected during periods of intense growth, equilibrium predictions deviated significantly from the observed values and from the kinetics model predictions by as much as 3 orders of magnitude. Concentrations calculated on a lipid basis by both models significantly underestimated the observed accumulation and questioned the hypothesis that PCBs accumulate in the lipid portion of phytoplankton. However, on a relative basis, the kinetics model reproduced the observed accumulation significantly better than the equilibrium approach.