The sudden infant death syndrome (SIDS) is the leading cause of postneonatal infant mortality in the United States today, despite a dramatic 38% decrease in incidence due to a national risk reduction campaign advocating the supine sleep position. Our research in SIDS brainstems, beginning in 1985 and involving a single, large dataset, has become increasingly focused upon a specific neurotransmitter (serotonin) and specific territories (ventral medulla and regions of the medullary reticular formation that contain serotonergic neurons). Based on this research, we propose that SIDS, or a subset of SIDS, is due to a developmental abnormality in a medullary network composed of (at least in part) rhombic lip-derived, serotonergic neurons, including in the caudal raphé and arcuate nucleus (putative human homologue of the cat respiratory chemosensitive fields); and this abnormality results in a failure of protective responses to life-threatening stressors (e.g. asphyxia, hypoxia, hypercapnia) during sleep as the infant passes through a critical period in homeostatic control. We call this the medullary serotonergic network deficiency hypothesis. We review the triple-risk model for SIDS, the development of the dataset using tissue autoradiography for analyzing neurotransmitter receptor binding; age-dependent baseline neurochemical findings in the human brainstem during early life; the evidence for serotonergic, rhombic lip, and ventral medullary deficits in at least some SIDS victims; possible mechanisms of sudden infant death related to these deficits; and potential causes of the deficits in the medullary serotonergic network in SIDS victims. We conclude with a summary of future directions in SIDS brainstem research.