Swimming rhythm in decerebrated, paralyzed stingrays: normal and abnormal coupling.

Rhythmic motoneuronal activity was recorded from decerebrated, paralyzed stingrays and compared with electromyograms recorded from the same animals. Before and after paralysis, a rostral-to-caudal sequence of alternation occurred between dorsal (elevator) and ventral (depressor) efferents. The swimming pattern was thus observed in the absence of phasic afferent input, and this constitutes fictive locomotion. After paralysis, both the intersegmental delay (time between activation at progressively caudal recording sites) and the burst duration remained linearly related to the swim cycle period. In many instances, neither the slope nor the intercept was significantly altered by immobilization. The intercepts all fell near the origin, indicating that the fictive rhythm remains constant phase coupled. Although the swimming rhythm was obtained after paralysis, some differences occurred. These included fewer and shorter spontaneous sequences, a greater range of cycle periods, and longer burst durations. During fictive swimming, the burst duration:cycle period ratio usually increased to 0.53 from 0.39 observed before paralysis. Therefore, the silent periods seen between burst discharges in antagonist efferents during movement were often absent after paralysis. Mechanical stimulation of the tail reduced both cycle periods and burst durations; however, the burst:cycle ratio remained greater than or equal to 0.50. The linear relation between burst duration and cycle period found for spontaneous sequences was not changed by stimulation of the tail. During fictive swimming the inter- and intrasegmental coupling that characterizes stingray swimming becomes labile. Abnormal coupling appears more often during sequences with long swim cycles. Intrasegmental coupling is tighter than intersegmental coupling at any cycle period. Rhythmic activity at one segmental level can be independent of activity at other levels. This suggests that multiple oscillator circuits exist that are not dependent on propriospinal circuits interconnecting different segments. Rhythmicity in elevator and depressor motoneurons is not dependent on reciprocal connections between the circuitry driving the motor nuclei. Therefore, separate oscillators for elevators and depressors appear to be present within one spinal segment.