Normal locomotion in vertebrates and anlaropods is a complex behavior, and the neural mechanisms that coordinate their limbs during locomotion are not known with any certainty. Rhythmic movements of the swimmerets of crayfish are known to be coordinated by neural networks in their segmental ganglia, and the analysis of the intersegmental circuit that coordinates movements of neighboring swimmerets holds promise of new insights into neural mechanisms controlling locomotion. This proposal presents new ideas about intersegmental coordination in the swimmeret system. These follow the recent demonstration that an older hypothesis, which postulated an increased excitability of local circuits in segments that normally lead each cycle of movements, was incorrect; these local circuits do not differ in their excitability. Instead, the intersegmental circuit that coordinates them is polarized. Preliminary physiological results show that axons of ascending and descending coordinating interneurons can be isolated separately, so the predictions of the coupled-oscillator model will be tested by selectively interrupting these axons and comparing the system' s subsequent performance with the prediction. The proposal also presents a new cellular model of the coordinating circuit that is based on published descriptions of individual coordinating units. The cellular model was developed by testing the performance of many alternative patterns of connections with components of the local circuits in each ganglion that coordinating interneurons might make. To test this new model, its responses to non-uniform excitation of local circuits will be compared with the swimmeret system' s responses to non-uniform excitation of individual ganglia. To test the cellular model' s predictions, the neurons that are targets of individual coordinating interneurons will be ma pped electrophysiologically. In experimental preparations, firing of each interneuron will be perturbed with injected currents. The system' s response to these perturbations will be measured and compared with simulations of the same perturbations in the model.
