The pendulum test, where the leg is dropped and the knee allowed to swing passively in the manner of a pendulum, has been recognized as a reliable and potentially valuable measure of hypertonia. In this study the traditional form of the test, where a goniometer is attached to the leg, has been replaced by a procedure involving computerized video motion analysis. Using 77 control subjects, a battery of 10 amplitude and time-based measures were obtained for purposes of investigating reliable and valid indicators of the damped, unsustained, oscillatory motion that characterizes this test. The results from the control subjects were congruous with those obtained by other workers using the goniometric version of the test. The measures of the response considered in this study showed evidence of increasing resistance to pendular passive joint motion with advancing age. Retest reliability of pendulum test responses in 14 subjects examined on average 26.8 days apart, produced Intraclass Correlation (2,1) values for the entire pendulum test response of an average of 0.84. It is concluded that the video-based pendulum test is a simple reliable source of measures with considerable potential for the clinical and physiological investigation of neurological and nonneurological features of normal and abnormal passive joint motion, and as a standard against which the effects of therapeutic intervention, such as medication, may be evaluated.