In a prospective randomized trial, 361 male patients with histologically proven adenocarcinoma of the rectum, judged preoperatively to require abdominoperineal resection (APR), were treated by surgery alone or were given 3,150 rads of preoperative radiotherapy. Surgical resection was done on 320 patients, 262 having "curative" APR. Only moderate symptoms from radiotherapy were noted and postoperative complications and 30-day mortality were similar in both groups. Five-year survival for curative APR was the same in both groups (50% for both treated and control patients). The incidence of positive lymph nodes in the resected specimens was 35% in treated and 41% in controls. In the first preoperative radiotherapy trial conducted by the group, 5-year survival in patients undergoing "curative" APR was 47% in treated versus 34% in control groups. Additionally, the difference in positive lymph nodes in the resected specimens was substantially greater in the first trial (26% in treated versus 44% in controls).