Muscle atrophy is clearly related to a loss of muscle torque, but the reduction in muscle size cannot entirely account for the decrease in muscle torque. Reduced neural input to muscle has been proposed to account for much of the remaining torque deficits after disuse or immobilization. The purpose of this investigation was to assess the relative contributions of voluntary muscle activation failure and muscle atrophy to loss of plantarflexor muscle torque after immobilization. Nine subjects (ages 19-23) years with unilateral ankle malleolar fractures were treated by open reduction-internal fixation and 7 weeks of cast immobilization. Subjects participated in 10 weeks of rehabilitation that focused on both strength and endurance of the plantarflexors. Magnetic resonance imaging, isometric plantarflexor muscle torque and activation (interpolated twitch technique) measurements were performed at 0, 5, and 10 weeks of rehabilitation. Following immobilization, voluntary muscle activation (56.8 ± 16.3%), maximal cross-sectional area (CSA) (35.3 ± 7.6cm2), and peak torque (26.2 ± 12.7N-m) were all significantly decreased (p < 0.0056) compared to the uninvolved limb (98.0 ± 2.3%, 48.0 ± 6.8 cm2, and 105.2 ± 27.0 N-m, respectively). During 10 weeks of rehabilitation, muscle activation alone accounted for 56.1% of the variance in torque (p < 0.01) and muscle CSA alone accounted for 35.5% of the variance in torque (p < 0.01). Together, CSA and muscle activation accounted for 61.5% of the variance in torque (p < 0.01). The greatest gains in muscle activation were made during the first 5 weeks of rehabilitation. Both increases in voluntary muscle activation and muscle hypertrophy contributed to the recovery in muscle strength following immobilization, with large gains in activation during the first 5 weeks of rehabilitation. In contrast, muscle CSA showed fairly comparable gains throughout both the early and later phase of rehabilitation.