Stepwise multiple regressions were used to determine the relative importance of physical attractiveness in the prediction of academic achievement for two samples of college students. In the first study, only ACT scores made an independent contribution to the prediction of males' (N= 55) grade average. For females (N= 65), verbal IQ (as measured by Borgatta and Corsini's Quick Word Test) and father's education yielded the only independent effects. In the second, follow-up study, males' (N= 50) achievement was significantly positively associated with intelligence and high school grades and negatively associated with attractiveness. For females (N= 87), independent effects were obtained for intelligence, high school grades, socio-economic status (positively correlated), and need for approval (negatively correlated). Attractiveness was again not associated with females' academic performance. In a final study involving high school students, attractiveness was not associated with achievement for 84 boys but was negatively associated for 83 girls. The results are discussed in terms of potential underlying mechanisms and the limitations of a “what is beautiful is good” stereotype.