ABSTRACT Relative to their frequency in the population, women are underrepresented in occupations that call for expertise in complex mathematics. Why? The author of this proposal hypothesizes a complex interaction between developmental differences in the self-concept, differences that occur as a consequence of pubertal changes, and school context to explain the documented discontinuities between boys and girls in math ability. This work has the potential to illuminate the reasons for the apparent superiority of boys in mathematics, and to show how the social factors responsible for these differences might be offset. The basic assumption of the work is that math performance is more crucial to the self-concept of young and adolescent boys than it is to girls. The discontinuity in perceived importance of math ability is assumed to accelerate during puberty. Using sophisticated measurements, the research will assess young students as they move from childhood to adolescence, to determine if the hypothesized changes in math self-concept occur as a function of developmental timing. The outcome of this longitudinal study will have important scientific and social implications. If the hypotheses prove tenable, we will have a much deeper understanding of the reasons that underlie the apparent loss of mathematical skill in females, and, perhaps more importantly, the means of preventing this needless decline.