Slopes of the dose-response relationships for all available neuromuscular blocking drugs appear to be essentially parallel and to approximate a log-dose/logit value of 4.75. We tested the possibility of estimating both 50% effective dose (ED(50)) and 95% effective dose (ED(95)) values from a single dose-response data point when that slope is postulated. We compared the ED(50) and ED(95) values of rocuronium and succinylcholine calculated by using traditional log-dose/logit regression analysis with the same values obtained by averaging individual estimates of potency as determined by using the Hill equation. After the induction of anesthesia (propofol/alfentanil), tracheal intubation was accomplished without the administration of neuromuscular blocking drugs. Anesthesia was maintained with nitrous oxide and propofol. The evoked electromyographic response to 0.10-Hz single stimuli was continuously recorded. After baseline stabilization, a single IV bolus of succinylcholine (0.08-0.26 mg/kg, n = 50) or rocuronium (0. 13-0.30 mg/kg, n = 40) was administered and the peak effect noted. By using log-dose/logit regression analysis, we calculated ED(50) and ED(95) values for rocuronium of 0.17 and 0.33 mg/kg and 0.14 and 0.27 mg/kg for succinylcholine. When potency was calculated from the Hill equation, the resultant ED(50) and ED(95) values did not differ by more than +/-4% from those obtained by using regression analysis. Averaging of single-dose estimates of neuromuscular potency provides a useful adjunct and reasonable alternative to conventional regression analysis.