Transplantation of fetal dopamine neurons to the adult striatum potentially offers a means to reverse the striatal dopamine deficiency that characterizes Parkinson's disease. Many investigations in rodents have supported the hope that neural grafting may be a useful treatment for parkinsonism. However, clinical studies have generally produced more modest improvements in motor abnormalities than observed in lower species. It is possible that the number of fetal dopamine neurons that survive transplantation is insufficient to restore dopaminergic innervation of the large human striatum to a level where striking recovery is obtained. In fact, there has been no quantitative study of graft outgrowth to indicate what portion of the dopamine-depleted striatum might be reinnervated with present techniques. Furthermore, it has been speculated that regeneration of the host dopamine system in response to the implantation surgery may play an important role in the beneficial effects of neural grafting in primates. The present study used nine 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated parkinsonian monkeys to investigate these issues. Sham implantation procedures produced no increase in either dopamine transporter density (measured by quantitative autoradiography) or tissue dopamine concentration (measured by HPLC) in the striatum of MPTP-treated monkeys. In sham-grafted and nonimplanted MPTP-treated monkeys, the striatal dopamine concentration was reduced by 99%, based on analysis of 16 sampled sites in the caudate nucleus and putamen of each monkey. No behavioral recovery was seen in the sham-grafted and nonimplanted MPTP-treated groups. In contrast, transplantation of fetal dopamine neurons to the caudate nucleus or putamen of MPTP-treated monkeys resulted in a significant elevation of dopamine transporter density and dopamine levels in the grafted striatal nucleus. Each grafted MPTP-treated monkey received ventral mesencephalon dopamine neurons from one donor harvested during putative neurogenesis. Donor ventral mesencephalon was divided equally and implanted into six sites either in the caudate nucleus or putamen. One graft site in each monkey was examined by dopamine transporter autoradiography. In sections in which graft fibers were present, a mean of one-third of the volume of the grafted nucleus was occupied by an elevated density of dopamine transporters. This increase in dopamine transporter density was defined to be at least 5-10% of the control density. However, full behavioral recovery was not observed in the grafted MPTP-treated group. These data provide no support for the hypothesis that regeneration of the host dopamine system occurs in response to a sham implantation procedure in severely parkinsonian monkeys. The current study illustrates the power of the applied techniques for delineating the relationship between the level of host dopamine depletion, the extent of graft-induced dopaminergic restoration, and behavioral recovery.