Little is known about the mechanisms that underlie motor recovery after incomplete spinal cord injury (iSCI) in humans. This study assessed changes in corticospinal tract (CST) function by measuring motor-evoked potentials (MEPs) and ankle motor control at 1, 3, and 6 months after acute iSCI. In 12 iSCI patients and matched controls, MEPs (evoked at 20% of maximal voluntary contraction [MVC]) were combined with a comprehensive ankle motor assessment protocol that measured ankle dorsiflexor strength (MVC, manual muscle testing, maximal movement velocity [MMV]), dexterity (the ability to accurately time ankle dorsiflexion movements) and gait (speed, walking aids). In the first 6 months after iSCI, all measures of muscle strength, gait and the MEP amplitudes significantly increased. The level of background electromyography (EMG) at 20% MVC remained stable, although absolute MVC increased. The MEP latencies were significantly delayed and remained unchanged during the first 6 months after iSCI. In addition, dexterity was preserved throughout rehabilitation. The percentage increase in MEP amplitude was significantly related only to the percentage improvement in MMV. The finding of unchanged CST conductivity, as assessed by MEP latencies in acute iSCI patients recovering motor function, is in accordance with previous studies in human SCI on this issue. The increased MEP facilitation at stable background EMG might indicate improved synchronization of the descending volley and/or responsiveness of motoneurons to supra-spinal input. The absence of a relationship between MEP amplitudes and recovery of ambulation and muscle strength implies that plastic changes in spinal neural circuits and preserved motor units might have contributed to the functional improvement.