Volunteer male subjects received single 1.0 mg oral doses of alprazolam and of clonazepam on two occasions, during coadministration of 40 mg/day fluoxetine or of placebo. When the sequence of trials was placebo first and fluoxetine second, fluoxetine coadministration significantly prolonged alprazolam half-life (20 versus 17 hours) and reduced clearance (48 versus 61 ml/min). No effect of fluoxetine was seen when fluoxetine was given first and placebo second, because norfluoxetine persisted into the placebo phase even though fluoxetine had been discontinued 2 weeks earlier. Fluoxetine had no significant effects on clonazepam elimination half-life or clearance regardless of the sequence of fluoxetine and placebo administration. In the fluoxetine-placebo sequence, fluoxetine significantly increased the rate of clonazepam absorption. Thus fluoxetine appears to impair clearance of alprazolam by way of microsomal oxidation but does not alter clearance of clonazepam by way of nitroreduction. The very slow elimination of norfluoxetine should be considered in the design of clinical or pharmacokinetic studies that involve fluoxetine.