It is generally accepted that biliary secretion is rate-limiting for the plasma-to-bile transport of a saturating load of bilirubin. Previous studies from this laboratory have suggested a relationship between the hepatic bilirubin uridine diphosphate glucuronosyltransferase activity and the apparent maximal rate of bilirubin secretion (excretory transport maximum). The present study was undertaken to further investigate this relationship in rats with a wide range of transferase levels and to analyze the effects of transferase activity on the formation of bilirubin monoglucuronides and diglucuronides in vivo and in vitro. Animals with moderately decreased enzyme activity still showed a near-normal excretory transport maximum but decreased fractional amounts of conjugated bilirubins in plasma and liver and a decreased ratio of bilirubins diconjugates to monoconjugates in bile, liver, and plasma. A more pronounced decrease in the level of transferase activity yielded lower excretory transport maximum values as well. Enhanced transferase activities were obtained by pretreatment with a series of enzyme-inducing agents, usually at dosages not influencing bile flow. An enhanced transferase activity produced significant increases in bilirubin excretory transport maximum, decreased concentrations of both unconjugated and conjugated bilirubins in plasma, and an increased ratio of diconjugates to monoconjugates in bile and in liver. In all animals, the ratio of disconjugates to monoconjugates observed in bile was correlated with the ratio found after incubation of liver homogenates in vitro at low concentrations of bilirubin substrate. Overall, the relationship between bilirubin excretory transport maximum and uridine diphosphate glucuronosyltransferase activity can best be described by a hyperbolic curve. At "saturation" of the excretory transport maximum values, no increase of conjugates in liver or in plasma could be observed, suggesting that not the biliary secretion, per se, but rather the conjugation rate in vivo is rate-limiting. The latter seems lower than the in vitro assayed enzyme activity. The present investigations point to an equilibration of bilirubin pigments between plasma, liver, and bile. The in vivo conjugation rate can be estimated from the in vitro assay (but not at high enzyme activities) and determines the maximal biliary secretion rate, the ratio of diconjugates to monoconjugates in the three compartments studied, and the relative amount of conjugates in plasma. Our studies suggest that the conjugation rate in vivo rather than the biliary secretion step is the major determinant of the maximal bilirubin secretion rate.