The rate of sciatic nerve regeneration and the effect of ganglioside treatment thereon were studied in the streptozotocin-diabetic rat. Two experimental protocols were used. In the first, sciatic nerves were crushed at 3 weeks of diabetes and treatment with purified bovine brain gangliosides (10 mg/kg/day ip) was begun the day before crush. In the second, nerves were crushed at 5 weeks of diabetes and treatment was started 7 days before crush. Regeneration was evaluated in both cases with the pinch-reflex test at different time points after crush. In either situation untreated diabetic rats showed a decreased rate of regeneration when compared to untreated nondiabetic rats, with a more severe reduction in 5-week diabetic rats. Ganglioside treatment improved regeneration in the second protocol; untreated diabetic rats had regenerated 10.6 ±0.9 mm (mean ± SEM) at 7 days postcrush, while nerves from ganglioside-treated and control rats had regenerated 16.4 ± 1.1 and 20.3 ± 1.0 mm, respectively ( n = 9–11, P < 0.001 for untreated vs ganglioside-treated diabetics with two-way ANOVA). Ganglioside treatment did not alter the rate of regeneration in nondiabetic rats.