Twenty-nine adult rhesus monkeys underwent complete laceration of both tibial nerves at mid-thigh level and repair by different methods to study the relative efficacy of autogenous interfascicular nerve grafts. Sixteen animals in Group I had an interfascicular graft repair using short sural nerve autografts on one limb and fascicular repair without grafts on the other limb. Thirteen animals in Group II, after having a 1-cm segment of tibial nerve resected, had an interfascicular graft repair without tension in one limb and an epineurial repair under moderate tension in the other limb. Evoked nerve and muscle action potentials and muscle strenght in response to repetitive and tetanic stimulation were recorded as baseline values prelaceration and then on re-exploration at 4, 6, 9, or 12 months. All nerves were examined by light and electron microscopy. Electrophysiological data, particularly muscle strength response, showed non-graft repairs to be superior at 4 and 6 months of regeneration. However, by 9 and 12 months the graft repairs had caught up and were equal to the non-graft repairs. Histologically, it was observed that many axons missed the graft segments and were present in extrafascicular connective tissues. Nonetheless, enough axons regenerated to the distal nerve to explain the success of these relatively short grafts. From the results of these experiments, it is concluded that use of autogenous interfascicular grafts offers no advantage over end-to-end non-graft repair. When and end-to-end repair cannot be achieved, use of short interfascicular nerve grafts is feasible and will work.