Carefully executed surgery for rectal cancer has reduced the incidence of local recurrence after restorative resection. Three recent large prospective series have confirmed the perception of a higher positive circumferential resection margin (CRM) rate after abdominoperineal resection. Tumour spread is different for low tumours and the surgical technique of abdominoperineal resection, perhaps better known as anorectal excision, may vary between surgeons. There is a need to redefine the place of anorectal excision and the contribution that can be made by pre-operative chemoradiation and/or extended surgery to reduce local recurrence and increase survival. Defined surgery, validated by histopathological assessment, as applied to TME surgery, would determine whether the perceived higher rates are due to the differences in routes of tumour spread or to surgically related variables. Unnecessary R1 or R2 resections and operative perforation can be minimized by an understanding of the surgical anatomy, the pattern of spread and difference in operative technique between anorectal excision and a low restorative operation. Surgical technique to maximize R0 resection should be based on a detailed understanding of the pelvic fascia and the levator ani and the use of pre-operative imaging to define lines of excision. With the adoption of even lower restorative resection (intersphincteric) there is a need to reassess the method of anorectal excision. This may be achieved by histopathological assessment of CRM positivity and MDT audit to improve results. Clinical trials are essential.