OBJECTIVE Drinking motives have frequently been linked to both the quantity of alcohol consumption and the likelihood of negative consequences. For motivational models of drinking to be useful, however, drinking motives must have predictive power independent of other variables typically associated with alcohol-related problems. Thus, this study evaluated the relationship between drinking motives and alcohol-related problems, after first accounting for high-risk alcohol consumption and gender, METHODS Subjects were 139 male and female (61%) college undergraduates, who completed a battery of self-report questionnaires as part of a course requirement. RESULTS Using hierarchical regression techniques, we found that the combination of maximum daily quantity and negative reinforcement reasons for drinking accounted for 61% of the variance on a measure of lifetime alcohol-related problems. Contrary to earlier research, neither gender nor positive reinforcement motives were significant predictors in the complete model. Follow-up analyses, conducted separately for negative and positive reinforcement motives, indicated that maximum daily quantity partially mediated the relationships between both types of motives and drinking problems; however, both positive and negative reinforcement motives retained unique predictive power. CONCLUSIONS Motives operate both indirectly through heavy drinking and directly to account for drinking problems. Results support the utility of motivational models of alcohol use in understanding alcohol-related problems in college students.