Mesencephalic tissue containing newly generated dopamine neurons was collected from brains of embryonic African green monkeys at 44 and 49 days of gestation and stereotaxically implanted into multiple sites in the caudate nucleus of adult monkeys previously treated with the dopamine protoxin, 1-methyl-4-phenyl-1,2,3,6-tetrathydropyridine. Ultrasonography was utilized to assess the developmental stage prior to hysterotomy. Brains were removed for combined histochemical and biochemical analyses at 3 1/2 months after grafting to determine the extent of graft survival and growth. The dopamine content of the target nucleus was assessed from microdissected "punches" placed in proximity to grafts identified in unfixed brain slices prior to fixation. Tissue dopamine levels adjacent to the grafts were elevated markedly, reaching 25-50% of control levels at some sites in the caudate nucleus. Morphometric analysis of graft size and dopamine cell numbers was performed with computer-enhanced, video-based imaging. Exceptionally large grafts that far exceeded their initial size at the time of implantation were seen at each placement site. The dopamine cell count was as high as 3500 in a single graft from E44 tissue, but only as high as 550 from the E49 donor. Up to 15,000 tyrosine hydroxylase-positive neurons were stained in the host monkey that received E44 tissue; only 1/10 as many were seen in each of the recipients of E49 day samples. The earlier donor grafts occupied as much as 15% of the caudate nucleus as seen in a single coronal section; summation of all sections that contained grafts at each placement from the E44 donor revealed average areas occupied by the grafts ranging from 3 to 8% of the caudate nucleus. In comparison, grafts produced from an E49 donor averaged between 2.4 and 5.4% of the area of the target. Qualitatively, grafts from each gestational stage showed well-developed dopamine neurons with morphological characteristics equivalent to those of all three ventral mesencephalic dopamine cell groups. The attainment of large, well-differentiated grafts with thousands of dopaminergic neurons from early gestation tissue suggests that optimal cell survival in primates is dependent on the degree of postgerminal development of the dopamine neuron. Neurite extension may be critical in this regard as well as other, at present, undefined factors. Maximal graft development and cell survival may be a critical element in the ability of neural grafts to reverse a neurological disability and to maintain improvement in the event of continued degeneration of host dopamine neurons.