Continuous measurement of changing limb segment angle is possible using a recently introduced flexible electrogoniometer but, unfortunately, the magnitude of linear displacements occurring during flexion and extension of the trunk has prevented its use as a measure of spinal mobility and posture. To overcome this limitation, a specially constructed, lightweight, aluminium carriage has been developed to allow free linear travel of the electrogoniometer, while maintaining the angle recorded between the thoracolumbar junction and the sacrum. The reliability and practicability of this new approach for recording changes in lumbar posture have been investigated. Following bench tests and operator training, measurements of maximal flexion and extension of the lumbar spine were repeated on 12 subjects. The extent of sagittal mobility was compared with measures from a fluid-filled inclinometer and a flexicurve. Analysis of the data showed that the use of the aluminium carriage did not markedly influence the accuracy of the electrogoniometer, and the technique as modified was found to be comparable with standard techniques for estimating the extent of lumbar sagittal flexibility. A field based study, involving four garage mechanics, was undertaken to examine the usability and acceptability of the electrogoniometer technique for continuous measurements. Lumbar curvature was monitored continuously for a two hour period during the course of the mechanics' working day. Dedicated software provided angle-time plots from which it was established that extreme flexed or extended postures were maintained only for relatively short periods; projection to an eight hour shift would indicate that a total of approximately 40 min is spent in extension beyond 10° and flexion beyond 50°. The technique has the potential to provide quantitative information on lumbar posture necessary for non-invasive, biomechanical investigations of spinal loading. It will be particularly useful for both sporting and occupational ergonomics field-studies.