Several hypotheses suggest that alcohol is often used to “escape”—to forget unpleasant feelings resulting from personal problems or social stresses. To test alcohol's effects on memory for feelings, 1.1 ml/kg of alcohol (A) or placebo (P) was given in two sessions, 48 hours apart, to 32 subjects divided into four equal drug groups (P-P, P-A, A-P, A-A). The subjects filled out the Profile on Mood States (POMS) after drug ingestion in session 1 while experiencing and learning these emotions. Their memories were tested when they tried to replicate these POMS ratings after drug ingestion in session 2. Alcohol given before memory testing decreased accuracy on each of the six POMS scales, especially Fatigue. However, alcohol did not impair memory for previously learned verbal and pictorial stimuli, and it produced no measurable changes in current feelings. These findings indicate that the impairment of memory for feelings may be a newly discovered specific pharmacological effect of alcohol.