Injections of HRP in the nucleus raphe magnus and adjoining medial reticular formation in the cat resulted in many labeled neurons in the lateral part of the bed nucleus of the stria terminalis (BNST) but not in the medial part of this nucleus. HRP injections in the nucleus raphe pallidus and in the C2-segment of the spinal cord did not result in labeled neurons in the BNST. Injections of 3H-leucine in the BNST resulted in many labeled fibers in the brain stem. Labeled fiber bundles descended by way of the medial forebrain bundle and the central tegmental field to the lateral tegmental field of pons and medulla. Dense BNST projections could be observed to the substantia nigra pars compacta, the ventral tegmental area, the nucleus of the posterior commissure, the PAG (except its dorsolateral part), the cuneiform nucleus, the nucleus raphe dorsalis, the locus coeruleus, the nucleus subcoeruleus, the medial and lateral parabrachial nuclei, the lateral tegmental field of caudal pons and medulla and the nucleus raphe magnus and adjoining medial reticular formation. Furthermore many labeled fibers were present in the solitary nucleus, and in especially the peripheral parts of the dorsal vagal nucleus. Finally some fibers could be traced in the marginal layer of the rostral part of the caudal spinal trigeminal nucleus. These projections appear to be virtually identical to the ones derived from the medial part of the central nucleus of the amygdala (Hopkins and Holstege 1978). The possibility that the BNST and the medial and central amygdaloid nuclei must be considered as one anatomical entity is discussed.