In a spinal turtle, unilateral stimulation in the rostral scratch receptive field elicited rhythmic fictive rostral scratching in ipsilateral hindlimb motor neurons; contralateral hip motor activity was also rhythmic and out-of-phase with ipsilateral hip motor activity. When left and right rostral scratch receptive fields were stimulated simultaneously, bilateral rhythmic fictive rostral scratching was produced; left hindlimb scratching was out-of-phase with right hindlimb scratching. Thus, spinal circuits coordinate interlimb phase during bilateral fictive scratching. We examined the contributions of contralateral spinal circuitry to the normal pattern of right hindlimb fictive rostral scratching by removing the left halves of the D7 segment and the hindlimb enlargement (D8-S2 segments). After left- hemicord removal, stimulation in the right rostral scratch receptive field usually elicited a variation of rostral scratching with rhythmic right hip flexor activity and no right hip extensor activity; thus, right hip flexor rhythm generation does not require left hindlimb enlargement circuitry. Normal right hindlimb rostral scratching with rhythmic alternation between hip flexor and extensor activities was rarely observed; thus, contralateral spinal circuitry contributes to the production of normal ipsilateral fictive rostral scratching. After left-hemicord removal, stimulation in the left rostral scratch receptive field elicited rhythmic right hip extensor activity; thus, contralateral spinal circuitry can generate a hip extensor rhythm during ipsilateral rostral scratch receptive field stimulation. Our observations and those of Berkowitz and Stein (1994a,b) support the concept that an ipsilateral hindlimb's normal rostral scratch motor pattern is generated by a modular central pattern generator that is bilaterally distributed in the spinal cord.

Spinal cord neuronal circuits generate motor neuron activity patterns responsible for rhythmic hindlimb behaviors such as scratching and swimming. Kinematic analyses of limb movements generated by this motor neuron output reveal important characteristics of these behaviors. Intralimb kinematics of the turtle hindlimb were characterized during five distinct rhythmic forms of behavior: three forms of scratching and two forms of swimming. In each movement cycle for each form, the angles of the hip and knee joints were measured as well as the timing of a behavioral event, e.g., rub onset in scratching or powerstroke onset in swimming. There were distinct differences between the kinematics of different forms of the same behavior, e.g., rostral scratch versus pocket scratch. In contrast, there were striking similarities between forms of different behaviors, e.g., rostral scratch versus forward swimming. For each form of behavior there was a characteristic angular position of the hip at the onset of each behavioral event (rub or powerstroke). The phase of the onset of knee extension within the hip position cycle occurred while the hip was flexing in the rostral scratch and forward swim and while the hip was extending in the pocket scratch, caudal scratch, and back-paddling form of swimming. The phase of the onset of the behavioral event was not statistically different between rostral scratch and forward swim; nor was it different between pocket scratch and caudal scratch. These observations of similarities at the movement level support the suggestion that further similarities, such as shared spinal circuitry, may be present at the neural circuitry level as well.

Hindlimb interlimb coordination was examined in turtles during symmetrical "same-form" behaviors in which both hindlimbs utilized the same movement strategy ("form") and during asymmetric "mixed-form" behaviors in which the form exhibited by one hindlimb differed from that of its contralateral partner. In spinal turtles, three forms of scratching were examined: rostral, pocket, and caudal. Bilateral symmetrical same-form scratching was studied for each of the forms. Asymmetric mixed-form scratching (rostral scratching of a hindlimb and pocket scratching of the other hindlimb) was also examined. In intact turtles, two forms of swimming were examined: forward swimming and back-paddling. The symmetrical behavior of bilateral forward same-form swimming and the asymmetric behavior of turning mixed-form swimming (forward swimming of 1 hindlimb and back-paddling of the other hindlimb) were studied. For all behaviors examined, most episodes displayed absolute or 1:1 coordination; in this type of coordination, during each movement cycle that began and ended with the onset of ipsilateral hip flexion, there was a single onset of contralateral hip flexion. For most of these episodes there was out-of-phase coordination between hip movements; the onset of contralateral hip flexion occurred near the onset of ipsilateral hip extension midway through the ipsilateral movement cycle. Bilateral caudal/caudal same-form scratching displayed out-of-phase 1:1 coordination during some episodes and in-phase 1:1 coordination during other episodes. During in-phase coordination, the onset of contralateral hip flexion occurred near the onset of ipsilateral hip flexion close to the start of the ipsilateral movement cycle. In a few cases of bilateral same-form scratching there were episodes of relative or 2:1 coordination; in this type of coordination, during each movement cycle of the slowly moving limb that began and ended with ipsilateral hip flexion, there were two distinct occurrences of the onset of contralateral hip flexion. The observation that out-of-phase movements of the hip occurred during symmetrical as well as asymmetric behaviors is consistent with the hypothesis that timing signals related to hip movement play a major role in interlimb phase control. The neural mechanisms responsible for interlimb phase control are not well understood in vertebrates. The present demonstration of bilateral scratching in spinal turtles suggests that this preparation may be suitable for additional experiments to examine mechanisms of vertebrate interlimb phase control.

The purpose of this study was to investigate the effect of back injury and load on the ability to replicate a target standing posture. Subjects included 16 back-injured males and 16 age-matched controls. Subjects were asked to reproduce a target standing posture and to repeat this task while holding a weight. Trunk inclination was measured using an electrogoniometer. Analysis of variance indicated that back-injured subjects were less accurate in reproducing the target posture than were control subjects. Back-injured subjects typically overestimated how much they had moved and therefore underestimated the target position. Both groups demonstrated a small but statistically significant difference between the load and no-load condition, with greater degrees of trunk inclination under the load condition. Analysis of covariance indicated that neither self-reported pain level nor maximum voluntary range of motion contributed to significant differences between the two groups. These data indicate that, following injury to the back, individuals may develop an inaccurate perception of body position. This study has implications for training the injured worker.

No abstract

No abstract