The aim of the present study was to identify the specific afferent projections to the rostral and caudal nucleus raphe magnus, the gigantocellular reticular nucleus pars alpha and the rostral nucleus raphe pallidus. For this purpose, small iontophoretic injections of the sensitive retrograde tracer choleratoxin (subunit b) were made in each of these structures. In agreement with previous retrograde studies, after all injection sites, a substantial to large number of labeled neurons were observed in the dorsal hypothalamic area and dorsolateral and ventrolateral parts of the periaqueductal gray, and a small to moderate number were found in the lateral preoptic area, bed nucleus of the stria terminalis, paraventricular hypothalamic nucleus, central nucleus of the amygdala, lateral hypothalamic area, parafascicular area, parabrachial nuclei, subcoeruleus area and parvocellular reticular nucleus. In addition, depending on the nucleus injected, we observed a variable number of retrogradely labeled cells in other regions. After injections in the rostral nucleus raphe magnus, a large number of labeled cells were seen in the prelimbic, infralimbic, medial and lateral precentral cortices and the dorsal part of the periaqueductal gray. In contrast, after injections in the other nuclei, fewer cells were localized in these structures. Following raphe pallidus injections, a substantial to large number of labeled cells were observed in the medial preoptic area, median preoptic nucleus, ventromedial part of the periaqueductal gray, Kölliker-Fuse and lateral paragigantocellular reticular nuclei. Following injections in the other areas, a small to moderate number of cells appeared. After gigantocellular reticular pars alpha injections, a very large and substantial number of labeled neurons were found in the deep mesencephalic reticular formation and oral pontine reticular nucleus, respectively. After the other injections, fewer cells were seen. Following rostral raphe magnus or raphe pallidus injections, a substantial number of labeled cells were observed in the insular and perirhinal cortices. Following caudal raphe magnus or gigantocellular reticular pars alpha injections, fewer cells were found. After raphe magnus or gigantocellular reticular pars alpha injections, a moderate to substantial number of cells were localized in the fields of Forel, lateral habenular nucleus and ventral caudal pontine reticular nucleus. Following raphe pallidus injections, only a small number of cells were seen. Our data indicate that the rostral and caudal parts of the nucleus raphe magnus, the gigantocellular reticular nucleus pars alpha and the nucleus raphe pallidus receive afferents of comparable strength from a large number of structures. In addition, a number of other afferents give rise to stronger inputs to one or two of the four nuclei studied. Such differential inputs might be directed to populations of neurons with different physiological roles previously recorded specifically in these nuclei.