Alcohol intake is known to impair memory in animals and humans. However, five studies reported that "medium" (0.05 ml/kg body weight) to "high" (1.0 ml/kg body weight) doses of alcohol improved memory when drunk immediately after initial learning of verbal or visual material. It was proposed that alcohol brought about this retroactive facilitation either through enhanced consolidation of memory traces or by protecting against retroactive interference. The present double blind study compared the performances of an alcohol and a placebo group on a kinesthetic memory task before alcohol or placebo intake and at retest 1 h after consumption. A second experiment was identical to the first except that all subjects carried out two trials on a T maze in the hour between testing and retesting. The alcohol group in the first experiment performed significantly better than the placebo group at retest (p < .05) but this was not the case in the second experiment. Alcohol therefore enhanced performance on the kinesthetic memory task in the first experiment but may not have protected against the moderate interference from the T maze in the second. The low levels of alcohol could have had a stimulant effect on trace consolidation, perhaps via raised blood glucose levels.