A biophysically based kinetic model of chemical absorption via human skin was developed and applied to the penetration kinetics of 12 chemicals: aspirin, benzoic acid, benzyl nicotinate, caffeine, chloramphenicol, colchicine, dinitrochlorobenzene, diethyltoluamide, malathion, methyl nicotinate, nitrobenzene, and salicylic acid. The pharmacokinetic model is linear and includes four first-order rate constants: (1) k1 describes penetrant diffusion through the stratum corneum; (2) k2 relates to further transport across the viable epidermal tissue to the cutaneous blood vessels; (3) k3 is a parameter which delays the partitioning of penetrant at the stratum corneum-viable tissue interface and, in conjunction with k2, reflects the penetrant's relative affinity for the stratum corneum over the viable tissue; and (4) k4 characterizes the elimination rate of chemical from blood to urine. Previously determined diffusion coefficients and molecular weight corrections were used to estimate k1 and k2; k4 values employed were those measured experimentally. Urinary excretion rate data following topical administration were simulated and k3 was estimated for each penetrant by optimizing the fit of the model to the data points. Ratios of k3/k2 should be related to the partition coefficients for the chemicals between stratum corneum and viable tissue and it was shown that these ratios agreed reasonably well with the corresponding octanol-water partition coefficients. This approach may have potential for predicting the general percutaneous absorption kinetics of chemicals based on recognized cutaneous biology and penetrant molecular weight and lipophilicity.