University of Missouri—Columbia Boys learn to read more slowly than girls in American schools, and for some years American educators have been concerned about this fact and the problems it poses. Several explanations have been advanced for sex differences in reading achievement, including hypotheses based on physical maturation, female teacher bias, teacher discrimination, feminization of reading, differential response to pupil behaviors, and sex-relevant teaching styles. Each of these hypotheses is conceptualized here, and evidence for and against each hypothesis is reviewed. Current evidence is found sufficient to reject only two of the hypotheses, and it is suggested that more than one of the remaining hypotheses may be needed to explain sex differences in reading achievement. Implications of the hypotheses for classroom teaching are explored. Sex is a strong predictor of human conduct, and many differences have been documented between the attitudes, behaviors, and achievements of males and females (Block, 1976; Deaux, 1976; Frieze, Parsons, Johnson, Ruble, & Zellman, 1978; Katz, Bower master, Jacobson, & Kessell, 1977; Maccoby & Jacklin, 1974). Although controversy has appeared about the size, causes, and implications of these sex differences, there is no disagreement about the contention that male and female sex roles are different in many ways. Given these differences, it is not surprising that people also expect males and females to act differently. Indeed, expectations about sex roles and the evaluations of male and female performance are sometimes more disparate than the observed behaviors and achievements of the two sexes (Goldberg, 1968; Pheterson, Kiesler, & Goldberg, 1971). Such expectations are not mere curiosities. Among other things, they lead adults to socialize young boys and girls in quite different ways, thus perpetuating sex roles in the new generation. So strong is this influence that differences have been detected in the play behavior of 1-year-old boys and girls (Goldberg & Lewis, 1969), and substantial differences appear in Requests for reprints should be sent to Barbara J. Bank, Center for Research in Social Behavior, 111 East Stewart Road, Columbia, Missouri 65211.