549who, on the basis of observations on pigs, postulated that selenium and/or vitamin E deficiency may cause sudden death in human infants. In addition, there is a widespread notion tihat breast-feeding may protect against this syndrome,2 which would be reasonable in the light of Money's hypothesis and the high concentration of vitamin E in human milk (7)(8) IU/1.).3 However, we have previously shown that the blood selenium levels of infants dying suddenly and inexplicably are essentiallv identical to those of normal controls4; their plasma vitamin E levels appeared to be slightly lower than normal, but our initial studies were performed on only a small number of samples and thus did not permit a strong conclusion regarding this point.We now have obtained additional data on the plasma vitamin E levels of infants dying suddenly and controls. The observed vitamin E levels of neonatal plasmna (0-28 ± 0.11 mng/100 ml), while lower than those noted by Nitowsky et al.,7 were similar to more recently published values;8 the neonates had lower plasma vitamin E levels than either the infants dying suddenly (0-49 ± 045 mng/100 ml) or control infants (0-79 ± 026 mg/100 ml; P=0.01). The infants who died suddenly (age at death 2-33 ± 2-85 months), though younger than the controls (4-28 ± 1-97 months), did not have significantly lower vitamin E levels than the controls (P= 005). Accordingly it seems unlikely that infants dying suddenly are seriously deficient in vitamin E during any period of their life.To determine the effect of breast-feeding on the incidence of the sudden infant death syndrome we have obtained dietary histories of 46 infants dying suddenly and 38 control infants in San Diego County. The two groups were similar in respect of parental social class, age, occupation, education, race, and income, date and hosoital of birth, sex, race, birth order, and birth weight (P>0 05). In these samples similar proportions of infants dying suddenly (39%) and control infants (27°', ) were breast-fed during early infancy (P = 0.22), comparable with the results of other studies.9 10 While tihe average age at death of the 27 infants dying suddenlv who had been wholly formula-fed was 3.65 ± 2779 months, the 19 totally or partially breastfed infants died at 1-78 + 0.70 months (P=002); thus breast-feeding does not appear to protect against sudden death.All the infants who died suddenly appeared to have received an adequate amount of dietary vitamin E, since the infant formulas fed to them contained 4-3-7-9 IU of vitamin E./.3 In addition, some had received extradietary vitamin supplements containing 5 IU of vitamin E/ml; interestinaly, however, fewer infants dying suddenly than controls had received vitamin supplements (P =002). Though the significance of the latter observation still remains to be explained, no other differences between the dietary histories of the two grou,ps were found. In agreement with earlier studies," we also noted a slightly higher incidence of maternal smoking during pregnancy in the sudden infant death gro...