Background: The aim was to evaluate the applicability of laparoscopic surgery in the treatment of primary rectal cancer in a training unit. Methods: A cohort analysis was undertaken of consecutive patients undergoing elective surgery for primary rectal cancer over a 7-year interval. Data on patient and operative details, and short-term clinicopathological outcomes were collected prospectively and analysed on an intention-to-treat basis. Results: A total of 306 patients (213 men, 69·6 per cent) of median (i.q.r.) age 67 (58-73) years with a median body mass index of 26·6 (23·9-29·9) kg/m2 underwent surgery. Median tumour height was 8 (6-11) cm from the anal verge, and 46 patients (15·0 per cent) received neoadjuvant radiotherapy. Seven patients (2·3 per cent) were considered unsuitable for laparoscopic surgery and underwent open resection; 299 patients (97·7 per cent) were suitable for laparoscopic surgery, but eight were randomized to open surgery as part of an ongoing trial. Some 291 patients (95·1 per cent) underwent a laparoscopic procedure, with conversion required in 29 (10·0 per cent). Surgery was partially or completely performed by trainees in 72·4 per cent of National Health Service patients (184 of 254), whereas private patients underwent surgery primarily by consultants. Median postoperative length of stay for all patients was 6 days and the positive circumferential resection margin rate was 4·9 per cent (15 of 306). Conclusion: Supervised trainees can perform routine laparoscopic rectal cancer resection.