The effect of repetitive conditioning lesions was tested on regeneration of the rat sciatic nerve. The nerve was conditioned by crush lesions one, two or three times with an interval of 2 or 4 days between each successive lesion. Axonal elongation was measured 3 days after a final test crush lesion. Two conditioning lesions stimulated axonal elongation more than one, while a third conditioning lesion had no further effect on axonal outgrowth. However, if the number of conditioning lesions were varied within a constant conditioning interval, outgrowth after the test lesion was the same. This suggests that the conditioning interval and not the number of conditioning lesions determined the outgrowth after a test lesion. When the conditioning lesion(s) and the test lesion were made at the same place, outgrowth was longer than if the lesions were spatially separated. Incorporation of [3H]thymidine in the regenerated nerve segment showed that proliferation of non-neuronal cells was initiated by each lesion. By counting the number of cell nuclei this proliferation was shown to correspond to an increase of cells in the regenerating nerve. It is therefore possible that the greater number of non-neuronal cells in the distal nerve segment accounts for the enhanced conditioning lesion effect in nerves where the conditioning and test lesions are made at the same place.