The effects of denervation on limb regeneration of postmetamorphic Xenopus laevis in the early to late stages of regeneration were studied. Limbs that were denervated immediately after amputation did not shbw any signs of regeneration. Moreover regenerating limbs denervated 20, 30, 40 and 60 days after amputation showed significant regression of regenerates. After this regression of regenerates, the regenerative capacity was not restored during a long observation period. Denervation reduced both protein synthesis (3H leucine autoradiography) and mitotic activity of regenerating limbs even in later stages of regeneration. These results on the role of the nerve in limb regeneration of Xenopus were compared with' results in Xenopus and urodeles reported previously.The effects of denervation on regeneration have been studied mainly in urodelen amphibia and results have shown that nerves in the limbs are important in the initiation of regeneration (1). Denervation performed concomitantly with amputation of limbs prevented regeneration in adult newts and Ambystoma larvae (1-8). In a'dvanced stages of regeneration, the nerve was not required for morphogenesis and regeneration proceeded in the absence of nerves, although the regeneration process was slower (7, 8).In anuran species, the role of nerves in limb regeneration has not been studied extensively. Denervation of regenerating limbs of larval Runa spp. caused regression, as in urodeles in early stages of regeneration (5-9). The froglets and adults of Xenopus laevis employed in our experiments, are known to possess the capacity to regenerate limbs in heteromorphic forms (10-13). The role of the nervous system in limb regeneration in this frog has infrequently been studied (1 3-1 5), but results have not yet been conclusive. In the present experiments, we examined the effects of denervation at various regenerative stages and found that limb regeneration in this frog showed characteristic nerve dependency.
MATERIALS AND METHODSFroglets of Xenopus laevis of 4.04.5 cm in snout-vent length were selected from the colony at Gunma University for experiments. They were anesthetized with 0.1 % MS 222 solution (Sandoz), and their forelimbs were unilaterally amputated at the zeugopodia. At the base of the stylopodium, all visible branches of the brachial nerves were excised for a length of 2-5 mm. Denervation was performed at the same time as amputation and 20, 30, 40 and 60 days after amputation of limbs. Groups of 12-15 animals were subjected to each denervation treatment. Unilaterally amputated non-denervated limbs were used as controls. The animals were kept in well water at 22*2"C and fed pellet grower for trout (Oriental Kobo Kogyo, Tokyo). The lengths of the regenerated limbs in each experimental series were measured every 10 days. The regenerates excised were fixed in BOUIN'S 463