By the 1960s it became recognized that certain organic compounds that had been used in large quantities and dispersed into the environment could persist and accumulate in biota to concentrations that cause adverse impacts. Examples included DDT, toxaphene, and polychlorinated biphenyls (PCBs). This understanding led to efforts to identify substances with high bioaccumulation potential. Early methods relied on laboratory bioconcentration tests that involved uptake and depuration phases in which aquatic organisms (often fish) were exposed to a constant concentration of the substance dissolved in water followed by exposure to clean water. Organisms were periodically sampled over the course of the test and analyzed to determine tissue concentrations of the substance investigated. These data were fitted to a simple model to estimate the uptake and elimination rates and the bioconcentration factor (BCF). The BCF represented the concentration ratio between the aquatic organism and water at steady state. Subsequent efforts led to the development of models to predict BCF from substance properties, most notably the correlations between BCF and log octanol-water partition coefficient (log K OW ) [1,2].While standardized laboratory bioconcentration tests provided a logical, empirical approach to identify substance-specific bioaccumulation concerns, further progress was gained through the introduction of mechanistic mass balance models that systematically described the various uptake and loss processes. One early example involved the prediction of PCB bioaccumulation in a Lake Michigan food chain [3]. This modeling framework incorporated a number of important concepts. First, 4 trophic levels were simulated (phytoplankton, zooplankton, forage fish, piscivorous fish) using a set of coupled, first-order kinetic equations under steady-state conditions. Second, 2 principal routes were considered for each trophic level beyond phytoplankton: uptake from water and ingestion of prey from the previous trophic level. Thus, unlike previous food-chain models that required prey concentrations as inputs, this model computed the prey concentrations associated with each trophic level based on the dissolved water concentration. Third, the rate of chemical uptake from water and food and role of growth dilution was linked to the bioenergetics of the organism using allometric scaling equations for each trophic level. Lastly, chemicalspecific toxicokinetic parameters (e.g., gill elimination rate, assimilation efficiency of contaminant from water and food) needed for model calibration were estimated using results from earlier lab studies.An important insight gained from model predictions and confirmed by field data compiled from Lake Michigan was that laboratory derived BCFs could significantly underestimate the extent to which PCBs were accumulated in field biota. This observation had significant management implications for derivation of water quality limits and emission reductions needed to achieve tissue concentrations that met regulatory guid...