The purpose of this study was to examine the respiratory depression produced by diazepam and by midazolam. Ventilatory and mouth occlusion pressure responses to CO2 were measured in eight healthy volunteers before and after the intravenous administration of 0.3 mg/kg of diazepam and 0.15 mg/kg of midazolam. The mean ventilatory response to CO2 (mean +/- SEM) decreased after administration of diazepam or midazolam from 2.0 +/- 0.2 to 1.3 +/- 0.1 1.min-1/torr or from 2.1 +/- 0.2 to 1.4 +/- 0.1 1.min-1/torr, respectively. In the same volunteers, the mouth occlusion pressure responses decreased from 0.54 +/- 0.05 to 0.30 +/- 0.04 cm H2O/torr after midazolam and from 0.67 +/- 0.12 to 0.28 +/- 0.07 cm H2O/torr after diazepam. When compared with the control slopes of the ventilatory and mouth occlusion pressure responses, the drug slopes were significantly different. Respiration was similarly depressed after diazepam and after midazolam. That both the ventilatory and mouth occlusion pressure responses to CO2 are equally depressed by intravenous injections of midazolam and of diazepam at equipotent doses suggests a direct depression of the central respiratory drive by these drugs.