The present study is a double-blind control trial of buspirone versus placebo in highly anxious alcoholics who recently completed inpatient detoxification for alcoholism. Subjects met DSM-III-R criteria for generalized anxiety syndrome and/or other nonpanic forms of anxiety disorders and alcohol dependence. Male veterans aged 21 to 65 were randomized to 45 to 60 mg/day of buspirone (n = 33) or placebo (n = 34). Anxiety scores, as measured by the Hamilton Anxiety Scale and other anxiety measures, declined significantly for both groups, but there were no differential group differences throughout the 6-month treatment period. Survival analysis measuring time to study drop out, time to first drink, time to 5 consecutive drinking days, and time to first intoxication indicated no significant differences between groups. The number of standard drinks per drinking day for nonabstainers also did not differ between groups. In the present study anxious alcoholics taking buspirone did not receive any benefit over placebo on a number of anxiety and alcohol use measures.