An oral loading dose was combined with intravenous infusion of 6% alcohol at rates adjusted on-line to close the gap between measurements of breath alcohol concentration (B,AC) and a target of 50 mg%. The goal was to minimize the deviation from a prescribed course of B,AC over time. In a pilot study of 10 young men, subjects underwent three experimental sessions: twice at 50 mg% and once in a 0 mg% control condition. The pilot study assessed the performance of the B,AC clamp, its potential utility in studies of acute tolerance to alcohol, and the retest reliability of directly measuring the alcohol elimination rate (AER) calculated from the steady-state infusion rate. Reduced variance was demonstrated in 4 of 5 experimental parameters, compared with results of an earlier approach using a split-dose oral administration procedure. Subjects' perceptions about alcohol effects were measured in one B,AC clamping session, using Schuckit's Subjective High Assessment Scale: 3 of 15 Schuckit's items demonstrated statistically significant indices of acute tolerance to alcohol. Within-subject AERs calculated in the steady-state had a coefficient of variation of 6.5%. Details of the B,AC clamping procedure are provided. The pilot study demonstrated the ability to prescribe experimental parameters of the brain's exposure to alcohol while preserving experimental flexibility in studies of acute tolerance to alcohol and AER.