1. To determine whether the motor cortex is involved in the modification of the hindlimb trajectory during voluntary adjustments of the locomotor cycle, we recorded the discharge patterns of 72 identified pyramidal tract neurons (PTNs) within the hindlimb region of pericruciate area 4 during a task in which cats stepped over obstacles attached to a moving treadmill belt. Data were also recorded from representative flexor and extensor muscles of the fore- and hindlimbs contralateral to the recording site. 2. To step over the obstacles, the cats increased flexion sequentially at the knee, ankle, and then the hip to bring the leg above and over the obstacle. This flexion movement was followed by a strong extension of the whole limb that repositioned the foot on the treadmill belt. These changes in limb trajectory were associated with large changes in the level of the activity of many flexor and extensor muscles of the hindlimb, and especially of the knee flexor, semitendinosus. On the basis of the time of onset of the knee and ankle extensor muscles in those steps when the limb was the first to be brought over the obstacle, the swing phase of the modified step cycle was subdivided into two parts, Phase I and Phase II, which correspond respectively to the flexion of the limb (F) and the initial extension (E1). 3. The temporal sequence of the movement was the same whether the hindlimb was the first (lead) or second (trail) to step over the obstacle, although the relative time between flexion at the three joints was changed in the two conditions. 4. Seventy-two PTNs were recorded from the posterior bank of the cruciate sulcus during the voluntary gait modifications. Sixty-three (63/72) of these PTNs had receptive fields that were confined to the contralateral hindlimb, or were recorded from penetrations in which such cells were found. Nine (9/72) PTNs had receptive fields on both the contralateral fore- and hindlimbs. Microstimulation applied through the recording electrode evoked, in all cases, brief twitch responses only in contralateral hindlimb musculature. 5. Forty-two (42/63) of those PTNs with receptive fields confined to the hindlimb showed a significant increase in their discharge frequency when the limb contralateral to the recording site was the first to step over the obstacle (lead limb). Twenty-nine PTNs (29/63) discharged maximally during the swing phase (18 in Phase I and 11 in Phase II), including two PTNS that also increased their discharge frequency during stance.(ABSTRACT TRUNCATED AT 400 WORDS)