ing relevant studies in this age group is the inconsistency in control of sun exposure when measuring dietary intake as the intervention and vice versa. Often, relative sun exposure has to be presumed from the country where the study was conducted.

When sun exposure is consistently adequate throughout the year, as it is presumed to be in South Africa, 1- to 8-year-old children of mixed race showed no evidence of vitamin D deficiency (serum 25(OH)D > 77.5 nmol/liter [31 ng/ml]) (Pettifor et al., 1978b). Unfortunately, dietary intake was not evaluated. When vitamin D intakes were less than 2.5 µg (100 IU)/day (mean intakes 0.6 ± 1.8 µg or 25 ± 70 IU/day) in vegetarian children younger than 6 years living in Boston, 4 out of 70 children in this 3-year longitudinal study developed radiologic evidence of vitamin D deficiency (Dwyer et al., 1979).

In the absence of age-specific data, additional observations in slightly older children were considered. In a longitudinal study in Norway, where sun exposure is presumed to vary over the year, an intake of vitamin D of about 2.5 µg (100 IU)/day from fortified margarine maintained normal vitamin D status in children aged 8 to 18 years (Aksnes and Aarskog, 1982). In 6- to 17-year-old Indian children in Turkey, a daily average intake (calculated from a 1-week diet history) of 2.0 ± 0.4 µg (78.6 ± 17.9 IU) compared with 1.1 ± 0.3 µg (45.3 ± 10.8 IU) vitamin D, appeared to support a better serum 25(OH)D status in most children in this observational study (Gultekin et al., 1987). In neither study was sun exposure measured. Taken together, these studies suggest that a dietary vitamin D intake of 1.9 to 2.5 µg (75 to 100 IU)/day may be adequate when skin synthesis of vitamin D is limited by sun exposure or skin pigmentation.

Children aged 2 to 8 years obtain most of their vitamin D from exposure to sunlight (Ala-Houhala et al., 1984; Gultekin et al., 1987; Meulmeester et al., 1990; Oliveri et al., 1993; Pettifor et al., 1978a; Riancho et al., 1989; Taylor and Norman, 1984) and therefore do not normally need to ingest vitamin D. However for children who live in far northern latitudes such as in northern Canada and Alaska, vitamin D supplementation may be necessary.

There are no data on how much vitamin D is required to prevent vitamin D deficiency in children aged 1 through 8 years. Extrapolating from available data in slightly older children (Aksnes and Aarskog, 1982; Gultekin et al., 1987) and from different continents for