There are various methods of measuring proprioception at the knee. Beard et al (1993) have described a delay in reflex hamstring contraction in anterior cruciate deficient knees. We have repeated their experiment and were unable to detect any significant difference in reflex hamstring contraction between the injured and uninjured legs. We discuss possible neurophysiological and biomechanical causes for the conflicting results and conclude that this method may not be a valid measure of proprioception.