SUMMARY A new objective clinical method to measure lateral thoraco-lumbar flexion is described. The technique, requiring only a centimetre tape and no special equipment, involves measuring the approximation of skin marks on the lateral trunk. A radiological control study of 43 individuals and assessment of intra- and inter-observer error revealed the method to be reproducible and reasonably accurate. A set of normal values for lateral flexion based on the application of the method to 237 normal subjects is reported. The measurements, which followed a Gaussian (normal) distribution, are tabulated according to mean, S.D., and range for each decade in both sexes. Particular attention is drawn to the decrease of mobility with advancing age, the wide scatter of measurements at each decade, and a sex difference in mobility—females having greater lateral movement than males. It is concluded that the simplicity and reliability of the method lends itself not only to routine clinical practice, but also to epidemiological work. Its value as a diagnostic tool, particularly in differentiating early ankylosing spondylitis from lumbar disc prolapse, is emphasized. Serial use of the technique may also prove helpful in the therapeutic and prognostic assessment of patients with spinal disease.