Humans walk and run over a wide range of speeds with remarkable efficiency. For steady locomotion, moving at different speeds requires the muscle-tendon units of the leg to modulate the amount of mechanical power the limb absorbs and outputs in each step. How individual muscles adapt their behavior to modulate limb power output has been examined using computer simulation and animal models, but has not been studied in vivo in humans. In this study, we used a combination of ultrasound imaging and motion analysis to examine how medial gastrocnemius (MG) muscletendon unit behavior is adjusted to meet the varying mechanical demands of different locomotor speeds during walking and running in humans. The results highlighted key differences in MG fascicle-shortening velocity with both locomotor speed and gait. Fascicle-shortening velocity at the time of peak muscle force production increased with walking speed, impairing the ability of the muscle to produce high peak forces. Switching to a running gait at 2.0 m·s −1 caused fascicle shortening at the time of peak force production to shift to much slower velocities. This velocity shift facilitated a large increase in peak muscle force and an increase in MG power output. MG fascicle velocity may be a key factor that limits the speeds humans choose to walk at, and may explain the transition from walking to running. This finding is consistent with previous modeling studies.muscle mechanics | biomechanics | preferred transition speed A nkle plantar-flexor muscles are a vital source of mechanical power for human locomotion (1, 2). During walking, plantar-flexor muscles provide body weight support, contribute to propulsion, and accelerate the limb into swing (3). In running, the ankle acts in a spring-like manner, absorbing energy in plantarflexor muscle-tendon units during early stance and providing energy to accelerate the body in late stance (1). The mechanical work required to produce whole-body movement during walking and running varies between gaits and across speeds (4, 5). Thus, the plantar-flexors may need to adjust their mechanical work output with gait and speed to meet the changing demands on their contribution to total mechanical work.A recent experimental study in humans used an inverse-dynamics approach to examine how the mechanical power outputs of muscles acting at the hip, knee, and ankle joint were modulated for walking and running at a range of speeds (6). It was found that positive power output at the ankle, in conjunction with the knee and hip, increased with walking speed. Also, Hansen et al. (7) showed that at walking speeds above those preferred, the net positive work done at the ankle increased. When switching from walking to running gait, the relative contribution of ankle positive power output to total positive power output also increased (6). It was inferred from these data that plantar-flexor muscle mechanics were adjusted to accommodate faster walking speeds and then again with the switch to running gait. If this is truly the case, then it may hav...