In this paper we extend the concept of educational attainment to cover the field of education taken in addition to the conventional level of education attained. Our empirical investigation uses register records containing childbearing and educational histories of an entire cohort of women born in Sweden (about a quarter-million individuals). This allows us to operate with a high number of educational field-and-level combinations (some sixty in all). It turns out that the field of education serves as an indicator of a woman's potential reproductive behavior better than the mere level attained. We discover that in each field permanent childlessness increases some with the educational level, but that the field itself is the more important. In general, we find that women educated for jobs in teaching and health care are in a class of their own, with much lower permanent childlessness at each educational level than in any other major grouping. Women educated in arts and humanities or for religious occupations have unusually high fractions permanently childless.Our results cast doubt on the assumption that higher education per se must result in higher childlessness. In our opinion, several factors intrinsic and extrinsic to an educational system (such as its flexibility, its gender structure, and the manner in which education is hooked up to the labor market) may influence the relationship between education and childlessness, and we would not expect a simple, unidirectional relationship.