Moderating effects of meal composition on psychomotor performance impairment and feelings after alcohol were examined in a between-subjects design. Fifty-one male volunteers fasted or received either a high carbohydrate (85% energy) or a high protein (94% energy) meal. Alcohol was administered at a dose to achieve a blood alcohol level (BAL) of 60 mg/100 ml, as a placebo. Subjects performed a dual task of primary tracking and secondary reaction time and a five-choice reaction time task. Feelings were also assessed by rating. The high carbohydrate meal reduced BAL at peak and 2 h after drinking, but a high protein meal had no significant effect. Although performance was impaired by alcohol, neither meal significantly reduced impairment and there was no effect of meal type on performance in the placebo condition. However, alcohol increased rated intoxication and the high carbohydrate meal reduced this effect. Subjects who had consumed high protein meals had more negative affect 2 h after alcohol than did subjects who had consumed high carbohydrate meals or fasted. It is concluded that there is only a weak relationship between BAL and performance impairment and food has only limited effects on impairment, although it reduces BAL.