Hypothetical in vitro biotransformation rate and affinity values for fish were extrapolated to a set of in vivo whole-body metabolism rate constants. A one-compartment model was then used to investigate potential effects of metabolism on chemical bioaccumulation as a function of octanol/water partitioning (Kow). In a second model-based effort, in vitro data were incorporated into a physiologically based toxicokinetic (PBTK) model for fish. The two models predict similar effects on bioaccumulation when calculated in vivo intrinsic clearance values (CL(IN VIVO,INT) are less than 50% of estimated liver blood flow (Q(LIVER). When CL(IN VIVO,INT) approaches Q(LIVER), the PBTK model predicts a greater effect on bioaccumulation than the one-compartment model. This result is attributed to the structure of the PBTK model, which provides for first-pass clearance of chemicals taken up from food. Uncertainties inherent to in vitro-in vivo extrapolations of hepatic metabolism data include the effects of protein binding, inaccurate estimation of in vivo metabolism by in vitro assays, and failure to account for metabolism in other tissues. Model-based predictions of bioaccumulation within a natural setting also must account for possible metabolism at multiple trophic levels. The models described in this study can be used to perform in vitro-in vivo metabolism comparisons with fish, estimate in vitro biotransformation parameters on the basis of measured chemical residues in field-collected animals, and calculate the level of in vitro metabolic activity required to limit bioaccumulation of all compounds to a specified value.