Investigated the utility of subtests from the Wechsler-Bellevue Intelligence Scale and the Halstead-Reitan Neuropsychological Test Battery to differentiate between alcoholics and nonalcoholics (N = 76). Analyses of variance indicated that the alcoholics were more impaired than nonalcoholics. It was found that the Wechsler-Bellevue Performance subtests were more discriminative than were Verbal subtests, with performance on the Halstead-Reitan variables being intermediate. The nine measures that differentiated most significantly between groups were subjected to a stepwise multivariate discriminant analysis. The resulting function correctly classified Ss with an overall accuracy of 74.7%. It was found that the Block Design subtest was the best single discriminator. The findings were discussed in relationship to previous findings and with respect to general issues of clinical neuropsychological assessment.