The pharmacokinetics of PCBs are complicated by numerous factors, not the least of which is the existence of up to 209 different chlorinated biphenyls. Whereas all PCB congeners are highly lipophilic and most are readily absorbed and rapidly distributed to all tissues, PCBs are cleared from tissues at very different rates, and the same congeners may be cleared at different rates by different species. With the exception of special situations in which PCBs may be passively eliminated in lipid sinks, e.g. milk or eggs, clearance is minimal prior to metabolism to more polar compounds. Rates of PBC metabolism vary greatly with species and with the degree and positions of chlorination. Mammals metabolize these compounds most rapidly, but even among mammalian species rates of metabolism vary greatly. In all species studied, the more readily metabolized chlorinated biphenyls have adjacent unsubstituted carbon atoms in the 3-4 positions. Congeners that do not have adjacent unsubstituted carbon atoms may be metabolized very slowly and are therefore cleared very slowly. These PCBs not readily cleared concentrate in adipose tissue. A physiologic pharmacokinetic model best illustrates how the concentrations of PCBs in all tissues approach equilibrium with the blood and with one another. Thus, the model illustrates how a depot of PCBs in any tissue, e.g. adipose tissue, will result in exposure of all tissues in proportion to the respective tissue/blood ratios and the body burden. The disposition of a number of PCBs in the rate has been accurately described by a physiologic model, and the model has been extrapolated to predict the disposition of these same PCBs in the mouse (58). Therefore, the physiologic pharmacokinetic model is believed to offer the best opportunity to extrapolate data obtained with laboratory animals to predict the disposition of PCBs in other species, including man. Most of the parameters of a model of PCB disposition in man are available or could be estimated. The major limitation to the construction of such a model is the absence of accurate estimates of metabolic clearance of individual PCBs by man. Accurate estimates of metabolic clearance depend on development of suitable in vitro methods to accurately predict clearance in vivo.