Maximum elbow flexor and extensor muscle strength was measured in 9 patients with Parkinson's disease on and off antiparkinsonian medication. In addition, the rate of force generation, the rate of actively returning force to resting levels, and passive release of force "relaxation" were measured in submaximal contractions. The measures of strength and contraction time were correlated with changes in clinical status as measured by the Unified Parkinson's Disease Rating Scale. When patients were off medication, their reduction in strength was significantly greater in extension than flexion. The reduction in flexion strength did not reach statistical significance. The extensor weakness was primarily due to decreased tonic activation of the extensor muscles and not to muscle coactivation. Muscle relaxation time was much more prolonged than was force generation time or active force return time. The increase in relaxation time and the decrease in extensor strength both correlated with changes in clinical status. Finally, changes in extensor torque correlated with the time to actively return force, suggesting that reduced strength is related to a reduced ability to generate rapid contractions in some patients with Parkinson's disease. These results suggest that there is an asymmetric distribution of muscle weakness in Parkinson's disease and that selected measures of muscle strength and muscle relaxation correlate with changes in clinical status.