SUMMARY A randomised controlled study of photocoagulation compared with no treatment in central retinal vein occlusion is reported. Forty-eight patients were allocated into treated and control groups by a random procedure 3 months after their first visual symptoms. For the analysis they were subdivided into (1) ischaemic type, and (2) hyperpermeability-response macular-oedema type of central retinal vein occlusion.In neither group did treatment confer benefit as far as visual acuity was concerned. However, iris, disc, and retinal neovascularisation improved after treatment in the ischaemic group. In addition none of the treated patients progressed to neovascular glaucoma. Macular-oedema improved in treated patients with the hyperpermeability response, but the visual field was affected and atrophic changes at the macula precluded visual improvement.It is concluded that photocoagulation should be used only to prevent complications in the ischaemic type of central retinal vein occlusion. It does not appear to be of value in the hyperpermeability group.Photocoagulation has been shown to be effective in the treatment of both macular oedema and neovascularisation in diabetic retinopathy. It is therefore reasonable to assume that similar results can be achieved in retinal vein occlusion, where macular oedema and new vessel formation are common complications. There are several reports indicating 'good results' (Zweng et al., 1974;Theodossiadis et al., 1974;Freyler and Nichorlis, 1974;Wetzig and Thatcher, 1974;Oosterhuis and Sedney, 1975;Sedney, 1976) 3 months after initial visual symptoms. At the 3-month examination patients whose visual acuity was 6/24 or worse entered the study provided they did not have neovascular glaucoma or other eye disease affecting vision and precluding treatment, and provided they agreed to randomisation.Randomisation was by the use of pre-prepared sealed envelopes which indicated whether an eye was for treatment by photocoagulation or remained untreated, i.e., control. No other therapy for vein occlusion was given to any of the patients, but glaucoma and medical conditions were treated as required. There were 24 treated and 24 control patients. The mean age of the patients was 62 years (range 44-79 years) in the treated and 61 6 years (range 36-79 years) in the untreated group.The patients were followed up for at least 1 year after entering the trial whenever possible. Five patients were not followed up for the required time. Two died, 4 and 10 months after entering the trial, 1 left the country, and 2 defaulted from follow-up.During the course of the study the initial scheme was slightly modified so that eyes with considerable capillary closure were entered into the trial as soon as the first fluorescein angiogram became available, i.e., before 3 months expired.Prior to analysis patients were subdivided into two subgroups according to the main cause of visual loss as revealed by fluorescein angiography. The 741