The relationships among peak blood concentration (Cmax), the time (tD) during which blood concentrations are maintained above the minimum effective concentration (Cmin), and the duration of a constant-rate intravenous infusion (T) of a drug exhibiting biexponential pharmacokinetics were simulated by digital computer using Newton iterative procedures. These simulations showed that, in contrast with our previous findings for drugs with monoexponential pharmacokinetics, the relationships are more complex due to the larger number of variables. Therefore, investigation of these relationships for biexponential drugs should be done on a drug by drug basis. Accordingly, meperidine and sulfamethoxazole were chosen as examples of drugs which exhibit biexponential kinetics and were used to determine what errors were involved in using the simpler guidelines for drugs which exhibit monoexponential kinetics as an approximation. These simulations showed the following . (a) The effect of T on Cmax may be adequately estimated by using the guidelines for monoexponential kinetics with the elimination half-life (t1/2) taken as t1/2 lambda1, provided that lambda1 is 20 to 30 times lambda2 and that the AUC of the distribution phase (i.e., C1/lambda1) comprises greater than about 20% of the total AUC. (b) As with monoexponential drugs, it is possible to obtain a larger tD by infusing the dose compared to that obtained with a bolus, if the value of C(0)/Cmin less than 2.5. (c) Using th approximation of monoexponential pharmacokinetics to estimate the effect of T on tD will underestimate both tD and the maximum infusion time, which will just attain the Cmin unless the AUC of the distribution phase comprises only a very small proportion of the total AUC. (d) The simulations with meperidine also showed that the nature of the relationship between tD and T varies depending on whether Cmin is maintained in the distribution phase or in both the distribution and elimination phases.