The effects of midazolam maleate and diazepam on cerebral metabolism and circulation were examined for each drug in six dogs maintained on N2O 70 per cent and halothane less than 0.1 per cent. Midazolam maleate at 0.2 mg/kg and diazepam at 0.3 mg/kg (the ED 100 per cent for induction of anesthesia for each drug in humans) did not decrease metabolic rate for oxygen (CMRO2) but did decrease cerebral blood flow (CBF) to about 55 per cent of control. Additional drug administrations (2.0, 5.0, and 10.0 mg/kg midazolam maleate, and 3.0 and 7.5 mg/kg diazepam) resulted in dose-related decreases in CMRO2 to a maximum of 55 per cent of control after 10.0 mg/kg midazolam maleate. Concomitant with the initial decreases in CMRO2, there was a change in the EEG as reflected by a decrease in frequency and an increase in amplitude. This EEG change suggests that 2.0 mg/kg midazolam maleate and 3.0 mg/kg diazepam represent a comparable canine anesthetic dose. Each additional dose of midazolam maleate decreased CBF to a greater extent than did the added doses of diazepam. Brain biopsies taken at the end of the midazolam maleate studies revealed a normal cerebral energy state (phosphocreatine, ATP, ADP, and AMP) and normal glucose, lactate, and pyruvate concentrations. In a hypoxic mouse model (FIO2 = 0.05), midazolam maleate provided greater protection from hypoxia (2.8 x control survival time) than diazepam (1.6 x control survival time). By comparison barbiturates in this model provide a survival time which is 4.0 x control.