In the 11-year period from 1978 and 1988, myelography was performed on 624 patients with suspected herniated lumbar discs. Positive findings, defined as an indentation of the contrast-filled dural sac and/or lack of contrast filling of the nerve roots, were found in 508 (81.4%) patients. Patients with suspected spinal stenosis were excluded from the study. The choice of treatment was determined by the preference of the consultant in charge. In all, 146 patients were treated by chemonucleolysis, 152 patients underwent surgical excision of the disc via laminectomy/fenestration, while 210 patients were listed for further treatment (albeit chemonucleolysis or laminectomy) but sustained relief of symptoms whilst waiting for admission or refused further intervention and consequently continued conservative treatments (such as physiotherapy, manipulations under anaesthetic, epidural or local injections). Following a repetitive questionnaire and clinical review, 78% of those patients that had undergone chemonucleolysis were satisfied with their treatment (i.e. symptoms were either 'gone', 'much better' or 'better'). Significantly fewer patients were satisfied after surgical disc excision or conservative treatments (71% and 61%, respectively; P <0.001). Whereas patients with a short duration of symptoms (less than 1 year) achieved the best outcome after chemonucleolysis and laminectomy, patients with a prolonged history (greater than 2 years) benefitted most from conservative methods. Of those patients in whom symptoms were considered to be worse or the same following treatment, 46%, 51% and 42% were still working following chemonucleolysis, surgery and conservative treatment, respectively. The outcome did not depend upon the size of the disc involved. Although this is a retrospective study, it is the first report to compare the outcome following chemonucleolysis, surgery and conservative methods in patients with disc prolapse confirmed by myelography.