The kinetics of a measure of pharmcologic effect (prolongation of the QT interval) of procainamide, as well as the kinetics of the plasma concentration, urine excretion, and saliva concentration of the drug were investigated in 14 trials in 4 subjects. A single 500-mg dose was given by rapid intravenous infusion, and frequent subsequent determinations of the above variables were made. A 2-compartment pharmacokinetic model with a third compartment for the saliva was used to fit the plasma, urine, and saliva data simultaneously. Analysis of the data reveals that the kinetics of the drug concentrations in saliva and of the pharmacologic effect are indistinguishable. They both must be considered to be different from those of the drug concentrations in plasma. Thus, in normal individuals under the conditions of this study, saliva concentrations more precisely indicate the time-course of drug at a cardiac site of action, although they do not parallel plasma drug concentrations until 6 hr or more after a rapid intravenous infusion. The following average pharmacokinetic parameters for plasma were found: terminal half-life, 2.9 hr; total clearance, 828 ml/min; renal clearance, 334 ml/min; and steady-state volume of distribution, 180 L. Average distribution pseudoequilbrium half-time (t1/2 alpha) was 5.2 min from an initial volume of distribution of 36.6 L.