Summary There is little evidence to support the general assumption that dietary carotenoids can improve vitamin A status. We investigated in Bogor District, West Java, Indonesia, the effect of an additional daily portion of dark-green leafy vegetables on vitamin A and iron status in women with low haemoglobin concentrations (<130 g/L) who were breastfeeding a child of 3-17 months. Every day for 12 weeks one group (n=57) received stirfried vegetables, a second (n=62) received a wafer enriched with β-carotene, iron, vitamin C, and folic acid, and a third (n=56) received a non-enriched wafer to control for additional energy intake. The vegetable supplement and the enriched wafer contained 3·5 mg β-carotene, 5·2 mg and 4·8 mg iron, and 7·8 g and 4·4 g fat, respectively. Assignment to vegetable or wafer groups was by village. Wafers were distributed double-masked. In the enriched-wafer group there were increases in serum retinol (mean increase 0·32 95% Cl 0·23-0·40 μmol/L), breastmilk retinol (0·59 0·35-0·84 μmol/L), and serum β-carotene (0·73 0·59-0·88 μmol/L). These changes differed significantly from those in the other two groups, in which the only significant changes were small increases in breastmilk retinol in the control-wafer group (0·16 0·02-0·30 μmol/L) and in serum β-carotene in the vegetable group (0·03 0-0·06 μmol/L). Changes in iron status were similar in all three groups. An additional daily portion of dark-green leafy vegetables did not improve vitamin A status, whereas a similar amount of β-carotene from a simpler matrix produced a strong improvement. These results suggest that the approach to combating vitamin A deficiency by increases in the consumption of provitamin A carotenoids from vegetables should be re-examined.