Clinical leakage of the anastomosis follows low anterior resection for rectal carcinoma in 5- 10 per cent of patients despite standard stapling techniques. A modification of this method has obviated the need for a distal purse string. A flexible transverse stapling instrument (Roticulator 55) is applied across the rectum below the tumour, and a double-staggered row of staples is inserted as a substitute for the distal purse string. End-to-end stapled anastomosis is then performed with peranal insertion of a Premium CEEA stapling instrument. In 111 patients the indications for operation were colorectal carcinoma (96 patients), diverticulosis (ten), megarectum (four) and ulcerative proctocolitis (one). Three patients had clinical evidence of anastomotic leakage; all survived. The incidence of radiological leakage on Gastrografin enema 10-12 days after operation was 9 per cent. The perioperative mortality rate was 2 per cent; all deaths were from cardiovascular causes. Local recurrence of tumour occurred in eight patients (7 per cent) after a mean follow-up of 40 months. In conclusion, double-stapled end-to-end anastomosis has made low anterior resection for rectal carcinoma a safe procedure with a low mortality rate, an acceptable local recurrence rate and minimal (clinical) anastomotic leakage.Clinical leakage of the anastomosis follows low anterior resection for rectal carcinoma in 5- 10 per cent of patients despite standard stapling techniques. A modification of this method has obviated the need for a distal purse string. A flexible transverse stapling instrument (Roticulator 55) is applied across the rectum below the tumour, and a double-staggered row of staples is inserted as a substitute for the distal purse string. End-to-end stapled anastomosis is then performed with peranal insertion of a Premium CEEA stapling instrument. In 111 patients the indications for operation were colorectal carcinoma (96 patients), diverticulosis (ten), megarectum (four) and ulcerative proctocolitis (one). Three patients had clinical evidence of anastomotic leakage; all survived. The incidence of radiological leakage on Gastrografin enema 10-12 days after operation was 9 per cent. The perioperative mortality rate was 2 per cent; all deaths were from cardiovascular causes. Local recurrence of tumour occurred in eight patients (7 per cent) after a mean follow-up of 40 months. In conclusion, double-stapled end-to-end anastomosis has made low anterior resection for rectal carcinoma a safe procedure with a low mortality rate, an acceptable local recurrence rate and minimal (clinical) anastomotic leakage.