Female hatchling alligators were castrated and implanted with a graft of either skeletal muscle tissue or testicular tissue from male hatchlings. Following surgery, each female also received a sustained-release pellet (Innovative Research of America, Rockville, MD) implanted subcutaneously that delivered one of the following treatments: a control substance (0.1 mg), 17 beta-estradiol (0.01 mg), or testosterone propionate (0.1 mg). Treated and final control females were sacrificed 8 weeks after surgery, and the mullerian and wolffian ducts were removed and examined histologically. The testis graft induced regression of the mullerian ducts in both the testosterone- and control-treated females. In the estradiol-treated females, however, no mullerian-duct regression occurred, indicating that estradiol prevented testis-induced regression. In the females that received a muscle graft, the mullerian ducts of placebo-treated females were morphologically similar to those of the intact final control females, but they were smaller in size. On the other hand, estradiol-treated mullerian ducts exhibited significant hypertrophy, but differentiation of muscle tissue in the stroma was not induced. Some of the testosterone-treated ducts were also slightly stimulated, but none showed any signs of regression. This suggests that testosterone does not induce regression in this species. Finally, there was no effect of either graft treatment or steroid treatment on the wolffian ducts. The evolutionary significance of these results is discussed.