Forty two patients with haemorrhage from peptic ulcers with visible vessels were enrolledin a randomised study comparing endoscopic haemostasis with adrenallne (1:10000) injections (adrenallne group) and adrenallne injection+ neodymium yttrium-aluminium-garnet (Nd:YAG) laser photocoagulation (adrenallne +laser group). The two groups (21 patients each) were well matched for factors affecting outcome. Surgery was performed for continued haemorrhage uncontrolled by endoscopic treatment or rebleeding after two endoscopic treatments. Haemostasis after one treatment was similar in the two groups: adrenallne 16/21 (76%), adrenallne+laser 18/21 (86%). Haemostasis after two treatments was numerically (OO5<p<0.10) greater in the adrenallne+laser group: 21/21 (100%) v 18/21 (86%). Three patients (14%) in the adrenallne group underwent uneventful emergency surgery. There were no deaths or procedure related complications in either group. Most bleeds from peptic ulcers with visible vessels can be controlled endoscopically without the need for surgery. Both treatments in this study proved highly efficacious in securing haemostasis. Adrenallne injection treatment seems to be the treatment of choice in view of its simplicity, low cost, and availability. Additional Nd:YAG laser treatment may provide a marginal improvement in efficacy, although a much larger trial would be required to prove this.