1. In the present study possible relationships between cardiovascular and respiratory effects and plasma concentrations were investigated after administration of midazolam and diazepam. Eight healthy volunteers were given three injections at 20 min intervals of equipotent sedative doses of midazolam (0.05 mg kg-1) and diazepam (0.15 mg kg-1) in a randomized double-blind cross-over design. Blood pressure, blood-gases and respiration measured nonivasively, were monitored throughout the experimental session of 160 min, and frequent blood samples were collected during the session. 2. Correlations between the blood pressure reduction, the increase of PaCO2 in blood, and plasma concentrations were found for both drugs. A maximal reduction of blood pressure and PaCO2 was produced after sedative doses of midazolam and diazepam. 3. A possible acute tolerance development towards the blood pressure reduction was found after the repeated administration of diazepam but not after the midazolam administration. 4. The plasma concentrations producing half the maximal effects after administration of midazolam was 50-60 ng ml-1, indicating that the influence on blood pressure and PaCO2 after drug administration is evoked at lower plasma concentrations than sedation. 5. No correlation between the respiratory effects and plasma concentrations was found for either drug.