Background: In the 1980s and 1990s Magnetic Resonance Imaging (MRI) was gaining popularity as the diagnostic imaging technique of choice for the investigation of disorders affecting the lumbar spine. The sensitivity of MRI to detect previously undetectable pathologies (e.g. tears in the annulus fi brosus) was well recognised but led the authors of this paper to question if MRI was suffi ciently specifi c. Method: Sixty-seven individuals who had never had lowback pain, sciatica, or neurogenic claudication underwent MRI scans of their lumbar spine. The scans were reviewed by three neuro-radiologists who were blinded to the reports of the other radiologists and to the presence or absence of clinical symptoms in the subjects. Results: Approximately one-third of the subjects were found to have an abnormality on their MRI scan. In the group under the age of 60, 20 % were found to have herniated discs and one had spinal stenosis. In subjects >60 years old, 57 % had 'abnormal' scans. The abnormalities that were detected included herniated discs in 36 % of subjects, and 21 % had spinal stenosis. Minor abnormalities such as degeneration or bulging of a disc were found in 35 % of the subjects aged between 20 and 39 years, and in all but one of the over 60 age group. Conclusion: The authors conclude that the high incidence of bulging and degenerate discs seen on MRI scans in asymptomatic subjects indicates that they should be viewed as normal or as part of the aging process. They recommend that before operative treatment is contemplated abnormalities on magnetic resonance images must be strictly correlated with age and any clinical signs and symptoms.