Activity in fragments left from sagittal sections.A, Before sagittal sectioning (left), 20 μm 5-HT and 7.5 μm NMDA induced a regular bursting pattern in an intact Th12–S1 preparation. After sectioning, the same drug concentration failed to induce rhythmic activity in the small fragment (top two traces, right; indicated by the number 2 in D, right), whereas it was preserved in the large fragment (bottom traces, right) of the cord, although with a longer period. B, Micrograph of a transverse histological section of the preparation in A showing the localization of the lesion (arrows). Scale bar, 200 μm. C, Ventral root responses to stimulation of the ventral funiculus on the rostral end of the small fragment. Scale bars: 10 msec, 100 μV. D, Summary of activity in all small fragments left from sagittal sections. Period (left) and modulation amplitude (right) before the lesion (open circles) are shown with the corresponding measurements on the same side after the lesion (closed circles). In all cases, the same concentrations of 5-HT and NMDA were used before and after the lesions. The level of section is indicated as a fraction of the distance from the midline (M, below abscissa) to the lateral border of the gray matter (L). The schematic on the right shows the position of the medial border of the lateral fragment in the transverse plane. E, Summary of activity in all large fragments left from sagittal sections. Graphs as in D. The schematic on the right shows the position of the lateral border of the large fragment in the transverse plane. Note that the level of the cut surfaces differed in the smaller and larger fragments even when they were derived from the same preparation, as was generally the case (see text). Rhythmic activity was present in all large fragments, but it was only observed after the most medial section in the small fragments (indicated by the number 1 in D, right). The period increase as a function of the section level in the large fragment was fitted with a power function (curve inE, left; data points omitted for clarity).