Case study evidence from developing countries suggests that availability of educated female labour is an important determinant of success in exporting manufactures. This paper, based on a general model in which the pattern of trade is determined by supplies of human and natural resources, uses crosscountry regression analysis as an alternative way to test the hypothesis of a linkage between female education and manufactured exports. The results give only weak support to the hypothesis, but are afflicted by collinearity in the cross-country data. Further research should thus focus on evidence at a level intermediate between case studies and cross-country regressions, and in particular on country studies.