This pilot study investigated the feasibility, toxicity and effect of continuous, hyperfractionated, accelerated radiotherapy (CHART) in 19 patients with adenocarcinoma of the rectum who were treated at Mount Vernon Hospital between April 1992 and April 1994. A total dose of 54 Gy was given in 36 fractions over 12 consecutive days; three fractions of 1.5 Gy were employed each day with an interfraction interval of 6 hours. Of these 19 patients, 13 had local pelvic recurrence from rectal carcinoma, four had a primary rectal carcinoma (which was unsuitable for abdominoperineal resection because of patient frailty), and two had large, fixed, unresectable tumours and were given radiotherapy preoperatively to a lower dose of 42 Gy in 28 fractions over 10 days to facilitate surgery. However, none of the 19 patients subsequently underwent surgical resection. Treatment compliance with radiotherapy was 95%. Acute bowel toxicity was assessed by means of a daily bowel chart for 6 weeks and then at 8, 12 and 26 weeks after the completion of radiotherapy. Patients were subsequently scored using the World Health Organization (WHO) grading system. No toxic deaths were observed in this study. There were five grade 3 and one grade 4 toxicities (WHO scale) observed in six patients. There has been minimal acute bladder toxicity in two patients and skin reactions have been surprisingly minimal in all but three patients. Small bowel obstruction and late bowel damage, possibly attributable to radiotherapy, occurred in three patients. Complete pain relief was achieved in recurrent rectal cancer in 100% of patients, at a median of 15 days since commencing radiotherapy (range 7-63 days), and has been maintained for mean and median durations of 19 and 18 months respectively (range 1.5-45 months). Five patients have had a recurrence of pain at 3, 4, 8, 9 and 18 months respectively. All four patients with primary rectal adenocarcinoma achieved a complete response, which has been sustained for 4 months (until death), 54 and 57 months, although one patient relapsed locally at 8 months. In summary, CHART appears to be a feasible treatment for pelvic tumours with regard to toxicity. Patients with pain from a locoregional recurrence achieved rapid complete pain relief, which was sustained for many months.