Despite the putative role of opioids in disorders of the developing human brainstem, little is known about the distribution and ontogeny of opioid-specific perikarya, fibers, terminals, and/or receptors in human fetuses and infants. This study provides baseline information about the quantitative distribution of opiate receptors in the human fetal and infant brainstem. Brainstem sections were analyzed from three fetuses, 19-21 weeks gestation, and seven infants, 45-68 postconceptional weeks, in whom the postmortem interval was less than or equal to 12 hours. Opiate receptors were localized by autoradiographic methods with the radiolabelled antagonist 3H-naloxone. Computer-based methods permitted quantitation of 3H-naloxone binding in specific nuclei, as well as three-dimensional reconstructions of binding patterns. High 3H-naloxone binding corresponds primarily to sensory and limbic nuclei, and to nuclei whose functions are known to be influenced by opioids, e.g., trigeminal nucleus (pain), nucleus tractus solitarii and nucleus parabrachialis medialis (cardio-respiration), and locus coeruleus (arousal). The regional distribution of opiate receptors as determined by 3H-naloxone binding is similar in human infants to that reported in human adults and animals and corresponds most closely to that of mu receptors. We found, however, that opiate receptor binding is high in the fetal and infant inferior olive, in comparison to low binding reported in this site in adult humans, primates, and rodents. In addition, opiate receptors are sparse in the fetal and infant substantia nigra, as in reports of the adult human substantia nigra, compared to moderate densities reported in primates and rodents. By midgestation, the regional distribution of 3H-naloxone binding in human fetuses is similar, but not identical, to that in infants. Highest 3H-naloxone binding occurs in the inferior olive in fetuses at midgestation, compared to the interpeduncular nucleus in infants. Tritiated naloxone binding quantitatively decreases in virtually all nuclei sampled over the last trimester, but not to the same degree. The most substantial binding decrease (two- to fourfold) occurs in the inferior olive and may reflect programmed regressive events, e.g., neuronal loss, during its development. Definitive developmental trends in 3H-naloxone binding are not observed in the postnatal period studied. The heterogeneous distribution of opiate binding in individual brainstem nuclei underscores the need for volumetric sampling in quantitative studies.