Objective: To gain information about the micronutrient status of urban, middle-income, breast-feeding mothers in relation to zinc and selected fat-soluble vitamins in plasma and breast milk and to assess possible interaction between the measured micronutrients. Design: Cross-sectional study Subjects: 91 mothers and their infants living in middle-income areas of Jakarta, Indonesia Results: None of the measured anthropometric data of the mothers (e.g. BMI: 22.0 AE 3.1 kgam 2 ) and their infants (birth weight: 3.2AE 0.5 kg) gave any indication of undernutrition. The mean concentrations in blood were 124 AE 18 gal for hemoglobin, 385 AE 111 mgal for retinol, 34AE 23 mgal for a-carotene, 104AE 72 mgal for bcarotene, 7.7 AE 3.3 mgal for a-tocopherol, 0.57 AE 0.23 mgal for g-tocopherol, 855AE 242 mgal for zinc, and the median concentration of lycopene was 29 mgal. The median breast milk concentrations were 420 mgal for retinol, 7.8 mgal for b-carotene, and 2.7 mgal for zinc. With increased duration of lactation, vitamin A and zinc concentrations signi®cantly decreased in breast milk whereas plasma zinc concentration increased. Plasma a-and b-carotene were positively correlated (P`0.0001) with each other and with plasma lycopene. Breast milk bcarotene was positively correlated with breast milk retinol and with plasma b-carotene (P`0.0001). There was no correlation between zinc and vitamin A in either breast milk or plasma. Forty per cent of the mothers were anemic, 29.1% had a low plasma zinc concentration, and 23.7% had a moderately low plasma vitamin A concentration. Breast milk from 70% of the women had a low concentration of vitamin A and that from 66% had a low concentration of zinc. Conclusions: Multi-micronutrient intervention should be considered to provide a suf®cient supply of zinc and vitamin A for growth of exclusively breast-fed infants.