Gastric bypass (GBP; 227 cases) was compared with gastroplasty (GP; 189 cases) with a follow-up of six months to four years. After two years, mean weight loss of patients with GBP was 43.4 kg (34.2% of preoperative weight and 63.5% of excess weight loss), with a 5.7% failure rate. Corresponding results for those receiving GP were 27 kg (22.1% and 44.7 %), with a failure rate of 42.5%. The revision rate for GP (11.1%) was twice that for GBP (4.3%). There were two GBP and five GP perforations. There were two deaths among the 416 patients (0.48%), both in the GBP group. Gastric bypass was found to be more effective with fewer revisions than gastroplasty, but the mortality was higher. Roux-en-Y GBP is our current choice for surgical treatment of morbid obesity.