with increasing parity (Funk et al., 1990). Debski et al. (1989) reported that the milk of vegetarians in California contained high concentrations of this trace element.
The enzyme glutathione peroxidase contains selenium. Its activity in milk correlates positively with the activity of this enzyme in maternal plasma and with both the linoleic acid and selenium contents of the milk. The enzyme's presence in milk may protect milk lipids from oxidative damage (Ellis et al., 1990). This suggests that the types of fatty acids consumed by the mother and the adequacy of her energy intake may influence the form and quantity of selenium secreted.
Several investigators have reported mean fluoride levels of 7 to 11 µg/liter in human milk (Ekstrand et al., 1984a; Esala et al., 1982; Spak et al., 1983); the American Academy of Pediatrics (Committee on Nutrition, 1985) suggests that 16 µg/liter be used as a normative value. Reports indicate that there is relatively little effect of maternal fluoride intake on the fluoride concentration of milk. Ekstrand and coworkers (1984b) observed that when a large fluoride dose (11.25 mg) was administered to a mother, only 0.2% of the dose was transferred through her milk to the infant. Spak et al. (1983) reported no significant difference in fluoride concentration of human milk when the fluoride content of the mother's drinking water was increased fivefold (from 0.2 ppm to 1.0 ppm). Although Esala et al. (1982) reported 50% higher levels in the milk of mothers whose drinking water contained 1.7 ppm of fluoride compared with mothers in areas with drinking water containing 0.2 ppm fluoride, the total amount of fluoride delivered through the milk of mothers in both groups was small. Singer and Armstrong (1960) suggested that plasma fluoride concentrations (and thus milk fluoride concentrations) do not increase unless water fluoride content exceeds 1.4 ppm.
Iodine is unique among the trace elements because the mammary gland avidly accumulates it. Its level in human milk correlates directly with maternal intake; major sources of iodine in the United States are bread, dairy products, iodized salt, and seafood (NRC, 1989). The mean iodine value for the milk of U.S. women in the 1980s was 178 µg/liter. In one study, iodine values in human milk were reported to be as high as 731 µg/liter (Gushurst et al., 1984). Such milk would provide more than 500 µg of iodine to the nursing infant per day. This level of intake is approximately 10 times greater than the RDA for infants (NRC, 1989). In contrast, the concentration of iodine in human milk is 20 µg/liter in northwestern Zaire, where the iodine supply to the lactating