To improve female students' attitudes toward mathematics and to increase the enrollment of females in upper level mathematics courses, students enrolled in grade 10, 11, and 12 at A. Y. Jackson Secondary School were placed in sex-segregated math classes. This study focused on four main areas: attitudes, achievement, enrollment, and perceptions of the program. The longitudinal analysis of the attitudes for students provided very little evidence that segregated classes had a beneficial effect on female students' attitudes. The pattern of changes in final mathematics grades differed for the sex-segregated and the comparison groups of students. The possible reasons are discussed. The degree of enjoyment and challenge in math and perceptions of the usefulness of math declined over time to a significantly greater degree for A. Y. Jackson students than for students at the comparison school. Although a sizeable proportion of all A. Y. Jackson students felt that the program had no effect on their attitudes toward, or achievement in mathematics, males were more likely than females to believe that segregated classes had a negative effect on attitudes and achievement; females were more likely than males to believe that the program had a positive or no influence on their attitudes and achievement. Appendices include student and parent questionnaires, interview materials, statistical tables, and responses to the questionnaires. (YP)