Chronic low back pain is, in the developed world at least, a costly problem. Costly to the individual in terms of the personal misery that accompanies it and costly to society in terms of working days lost, Sickness and Invalidity Benefits and healthcare provision. It has been estimated that during any 1-month period, 66% of patients who will ever have low back pain are symptomatic (Papageorgiou et al 1995). In spite of its prevalence, however, up to 85% of back pain patients cannot be given an accurate diagnosis (Moffett and Richardson 1995). Addressing this diagnostic problem relies, to some extent, on improving our understanding of the mechanics of the spine and how disorders might reveal themselves during spinal motion. Part 1 of this thesis considers the various methods of measuring spinal movements especially those concerned with dynamic imaging. The basis of some commonly used kinematic indices is also discussed before reviewing key studies, both in vitro and in vivo, of intervertebral motion. In addition, Part 1 includes a review of the concept of lumbar segmental instability. Part 2 considers the possible applications of digitised fluoroscopy including comparison to other spinal measures and a suggested role in improving selection of spinal surgery. Finally, Part 3 looks at recent developments in the evolution of digitised videofluoroscopy and discusses results from a study of lumbar spinal motion in a group of asymptomatic volunteers under a new passive motion protocol.