The sons of alcoholics have repeatedly been found to have reduced P300 amplitude. Further, quantitative behavioral genetic and molecular genetic studies indicating a genetic influence on P300 amplitude have fueled speculation that this component may be a biological vulnerability marker for alcoholism. To further explore this possibility, we examined P300 in adolescent twin pairs from an epidemiological sample who were (a) discordant for alcohol abuse/dependence, (b) concordant for alcohol abuse/dependence, or (c) concordant for the absence of alcohol abuse/dependence and other relevant disorders. For discordant pairs, the alcohol abusing/dependent twins' amplitude did not differ from that of non-alcoholic co-twins. Pairs free of psychopathology had greater amplitudes than both alcoholism discordant and concordant pairs. P300 amplitude was more similar in monozygotic than dizygotic discordant pairs, suggesting a genetic influence on P300 amplitude in this group. The findings are consistent with P300 amplitude being a marker of vulnerability to alcohol use disorders.