1. A quantitative and longitudinal analysis of locomotion was made after bilateral lesions of the dorsolateral funiculi (DLF) and/or the dorsal columns (DC) in the lower thoracic cord (T12 or T13) in five adult cats. All cats were chronically implanted several weeks before the spinal cord lesion to permit the recording of electromyographic (EMG) activity from selected flexor and extensor muscles of the fore- and hindlimbs of each side. This allowed each cat to act as its own control when comparing the pattern and amplitude of EMG activity before and after the lesion. All experiments were also videotaped to allow an analysis of the kinematic changes before and after the lesions. Kinematic data were only analyzed for the side of the cat facing the camera; for all cats this was the left side. 2. After recovery periods of 2-5 mo, wheat-germ-agglutinated horseradish peroxidase (WGA-HRP) was injected caudal to the lesion site (normally at L2). The extent of the lesion was verified both histologically and by evaluating the number of HRP-labeled neurons in different supraspinal structures. These analyses showed that the cortico- and rubrospinal tracts (CST and RST, respectively) were completely interrupted, bilaterally, in two of five of the cats; in one of these cats the DCs were also interrupted. In the other three cats there was more variable damage, and the CST and RST were only completely interrupted on the right side of one of these animals. The DCs were completely sectioned in two of these cats. 3. During the 1st wk subsequent to the lesion, most cats had difficulty in supporting their weight and in walking. However, within 10 days all were able to walk, unaided, for extended periods on the treadmill at speeds of at least 0.35 m/s. In the two cats with the complete, bilateral DLF lesions, the animals dragged both their left and right hindpaws along the treadmill belt during the swing phase of the step cycle (paw drag) throughout the testing period of 3-5 mo. In the other three cats, paw drag in either hindlimb was only seen in the 1st 2-3 wk after the lesions, with the exception of the cat with the complete lesion of the DLF on the right side, which showed sustained paw drag in the right hindlimb throughout the testing period. 4. Significant increases in step cycle and swing duration following the lesion were observed only in the two cats with the largest lesions. In all five cats, statistical comparisons of the slopes from a linear regression analysis showed that the relationship between swing and step cycle duration was unchanged by the lesions. 5. Joint angles in the left hindlimb of the two cats with the largest lesions were generally smaller (more flexed) than in the prelesion controls. This was particularly true for the knee and ankle joints. The other three cats showed changed joint angle values for the hip, knee, and ankle only in the 1st 2-3 wk after the lesions. All of the cats, except the one with the least damage to the left DLF, exhibited increased joint excursions at the metatarsophalangeal (MTP) joint of the left limb throughout the recovery period. 6. There were sustained changes in the coupling between the hip and the knee of the left limb, together with smaller changes in the coupling of the knee and ankle in the two cats with the complete lesions of the DLF. In contrast to the prelesion controls, flexion at the ankle occurred before swing onset in these same two cats. The changes in the coordination between the hip and the knee were associated with changes in the temporal coupling between the hip flexor, sartorius (Srt), and the knee flexor, semitendinosus (St). Despite some improvement, the coupling between these two muscles never fully recovered to the prelesion values. Similar, but smaller, changes in the delay between Srt and St were also seen in two of the other three cats. (ABSTRACT TRUNCATED)