parcellating interneurons controlling left/right alternation or flexor and extensor activity between T12 and L5, it was necessary for us to monitor a larger field of view. For this purpose, we used low magnification to detect, in the same focal plane, the phase preferences of the maximum number of spinal cells located not only near the midline, but also more laterally. This experimental procedure allowed for the first time to estimate the full neuronal activity along the entire lumbar SC. Several types of loading techniques have been previously used to label the spinal cells. Retrograde loading of Ca2+ green dextran amines has been used to label motoneurons and detect the spatiotemporal organization of their recruitment during locomotor-like activity. Electroporation has also been used to load neurons from L5 to S2 with Ca2+ sensitive dyes in the isolated SC of the neonatal mouse in vitro, Toltrazuril whereas pressure ejection of the dye between T13 and L2 segments has been performed to detect the activity of interneurons during locomotion. In the present study, all cells at or near the surface of the cord were loaded with membrane-permeable Ca2+ indicators to detect the Ca2+ transients from heterogeneous cells at the preparation��s surface. This loading method together with automated detection of cellular activity was shown to be an efficient procedure to measure cellular activity at the system level in the respiratory network. This method has the great advantage of easily resolving the activity of individual neurons with epifluorescence microscopy. Because fluo-4 AM is known to label both neuronal and glial populations, it cannot be ruled out that some of the signals may be of glial origin. Nevertheless, in neurons, Ca2+ influx accompanying action potentials is rapid and due to influx Etofibrate through high voltageactivated channels, while Ca2+ influx in glia is typically slower. Spontaneous astrocytic Ca2+ oscillations in situ drive NMDAR-mediated neuronal excitation, thus based on the steep rise associated with Ca2+ signals recorded here, it is likely that they are of neuronal origin. In all experiments of this study, care was taken to rule out the possibility that the difference of the concentration of neurons in L1 was due to an experimental bias: first, we verified that the cut was uniform at all the spinal cord levels by performing transverse sections from T13 to L5. Second, in contrast to local dye application methods used by others, all networks from T13 to L5 were uniformly labeled by bath-application of the membrane permeant indicator. Thus, the observation of locomotor-related activity concentrated at L1 likely indicates functional specialization of networks in this region. Although optical recordings revealed a higher density of locomotion-modulated interneurons at L1, lesions to this region did not eliminate fictive locomotion. L1 lesions typically elicited a longer lasting disruption of locomotor output, slowed locomotor period, and decreased VR burst amplitude, but in all cases, alternating, coordinated locomotor output returned. These results support two hypotheses: either networks mediating flexor-extensor coordination were outside the lesion sites, or multiple mechanisms give rise to flexor-extensor coordination such that a lesion to one network led to compensatory activation of the other. These findings are consistent with a wide range of studies that reveal the robustness of motor patterns to focal lesion in classic ����labeled line���� systems, and to both coarse and focal lesions to networks in ventrolateral medulla that generate respiratory rhythm as well as a classic study characterizing the robustness of memory to lesion.