Between March 1987 and February 1990, 1168 patients with resectable rectal cancer were recruited into a nationwide Swedish trial. Patients were randomly allocated to receive either preoperative irradiation (25 Gy infive fractions over 1 week) followed by surgery within 1 week, or surgery alone. In each group 454 patients underwent curative surgery. The groups were well matched for age, sex and surgical procedure. The postoperative mortality rates did not differ: 4 per cent after irradiation versus 3 per cent after surgery alone, but the postoperative mortality rate was significantly higher in patients treated with a two- rather than three- or four-portal technique (15 versus 3 per cent, P < 0·001). More perineal wound infections were found in patients receiving irradiation (20 versus 10 per cent, P < 0·001). No difference in the incidence of anastomotic dehiscence or other postoperative complication was found between the groups. Preoperative radiotherapy with a short-term high-dose technique can be delivered without major acute adverse effects, provided the irradiation technique is optimized.