Two fundamental aspects of the dynamic behavior of motor units of the cat medial gastrocnemius (MG) muscle were measured. Force-velocity (FV) relationships were measured with the use of constant velocity shortening and lengthening movements. Effects of mechanical history were assessed via comparisons of forces immediately after or during slow movements with standard isometric forces. Isometric force-length (FL) relations were also measured, and the effect of different stimulation rates on both FV and FL data was assessed. 2. Prior or concurrent movement greatly potentiated motor-unit force, but this movement potentiation was highly dependent on the amplitude of the unit's force. The smallest twitch forces of type S units (< 10 mN) were potentiated more than threefold, but no potentiation occurred for unit forces > 200 mN. It was tentatively concluded that movement potentiation may play little role in normal movements because it does not occur at forces > 1% of maximal isometric force of the MG. 3. During shortening, the normalized FV relations of type S units were relatively steeper than those of type FR or FF units. For lengthening, there was no evident relation between FV steepness and motor-unit type. 4. Stimulation rate affected both the FV and FL relationships of the motor units. The peak of the FL relationship (Lo) clearly shifted to shorter muscle lengths as stimulation rate was increased. The steepness of the FV relationship for shortening was decreased by increasing stimulation rate, but this effect was modest. 5. The shift in motor-unit Lo and the differences in motor-unit FV relationships were hypothesized to play significant roles during normal motor behavior. Realistic computer simulations of FL and FV functions for a population of motor units undergoing normal steady-state recruitment and rate modulation supported these hypotheses. As the level of simulated neural drive increased, the population Lo shifted to considerably shorter lengths, and the normalized FV function became much less steep. The significance of these results for models of muscle are discussed.
