This paper suggests that major explanations of union membership have been flawed by the free rider paradox. It outlines a recent British theory which claims to overcome this dilemma and draws from the model hypotheses which are tested by an analysis of a longitudinal survey of London school-leavers. The findings seem to offer support for the model in its categorization of core and remainder members, and further investigation is encouraged. © 1991, Taylor & Francis Group, LLC. All rights reserved.