The aims of this review are (a) to critically examine the epidemiologic evidence for a possible association between smoking and the sudden infant death syndrome (SIDS), (b) to review the pathology and postulated physiological mechanism(s) by which smoking might be causally related to SIDS, and (c) to provide recommendations for SIDS prevention in relation to tobacco smoking. Over 60 studies have examined the relation between maternal smoking during pregnancy and risk of SIDS. With regard to prone-sleep-position intervention programs, the pooled relative risk associated with maternal smoking was RR = 2.86 (95% CI = 2.77, 2.95) before and RR = 3.93 (95% CI = 3.78, 4.08) after. Epidemiologically, to distinguish the effect of active maternal smoking during pregnancy from involuntary tobacco smoking by the infants of smoking mothers is difficult. Clear evidence for environmental tobacco smoke exposure can be obtained by examining the risk of SIDS from paternal smoking when the mother is a non-smoker. Seven such studies have been carried out. The pooled unadjusted RR was 1.49 (95% CI = 1.25, 1.77). Consideration of the pathological and physiological effects of tobacco suggests that the predominant effect from maternal smoking comes from the in utero exposure of the fetus to tobacco smoke. Assuming a causal association between smoking and SIDS, about one-third of SIDS deaths might have been prevented if all fetuses had not been exposed to maternal smoking in utero.