Between April and August 1972, a prison education program was conducted in British Columbia Penitentiary, a maximum-security institution, and Matsqui Penitentiary, a medium-security prison. Program volunteers were required to have completed the eighth grade and to demonstrate the capacity for understanding university-level material. The selected volunteers were randomly assigned to experimental and control groups at each institution (18 in each group at the maximum-security institution and 15 in each group at Matsqui). The 5-month program consisted of instruction in English and history at the first-year university level. One teacher was assigned full-time to each institution and was in the classroom with the students all day. Extensive homework assignments kept the students busy up to 12 hours each day. Inmates making satisfactory progress were allowed to take additional first-year courses in sociology and psychology, using self-instructional packages supplemented by visiting lecturers. Remedial instruction was provided for the inmates on an individual basis. The majority of inmates completed the two credit courses which formed the core of the program, and a number completed the introductory-level courses in sociology and psychology. An examination of the recidivism rates of the experimental and control groups found no significant difference between the two groups in either institution. As a recent study shows, extending the length of the program and providing for release shortly after completion of the program could make a difference in the program's impact on recidivism. Tabular data and 12 references are provided.