Progesterone pharmacokinetics were analyzed for plasma hormone concentrations ranging from linear to saturated metabolism in lactating Holstein cows with differing daily milk yields. The adequacy of 2-coupled first-order (bi-exponential equation), hyperbolic (Michaelis-Menten equation), and sigmoidal (Hill equation) kinetic models to describe the experimental progesterone pharmacokinetic profiles was examined on a statistical basis. After nonlinear regression and statistical analysis of the data-fitting capability, a simple one-compartment model based on Hill equation proved to be most adequate. This model indicates an enzyme-catalyzed metabolism of progesterone involving cooperative substrate-binding sites, resulting from allosteric effects that yield a sigmoidal saturation rate curve. Kinetic parameters were estimated for 2 groups of lactating Holstein cows with different daily milk yields. We found, for the first time, a remarkable quantitative agreement of the Hill coefficient value with that reported in pharmacokinetic studies involving cytochrome P450, family 3, subfamily A (CYP3A)-mediated reactions in other mammals, humans included. It seems that positive cooperativity makes enzymes much more sensitive to plasma progesterone concentration, and their activities can undergo significant changes in a narrow range of concentration as characteristic of sigmoidal behavior. Therefore, the values of classical pharmacokinetic parameters, such as the elimination constant, half-life, and clearance rate, were found to be highly dependent on the plasma progesterone concentration.