The Ministry of Health in Sri Lanka commenced a vitamin A supplementation programme of school children with a megadose of 105 μmol (100 000 IU) vitamin A in school years 1, 4 and 7 (approximately 5-, 9- and 12-year-olds, respectively) in 2001. We evaluated the vitamin A supplementation programme of school children in a rural area of Sri Lanka. A cross-sectional study was conducted among children supplemented with an oral megadose of vitamin A (105 μmol; n 452) and children not supplemented (controls; n 294) in Grades 1–5. Children were clinically examined and a sample of blood was taken for serum vitamin A concentration estimation by HPLC. Socio-demographic information was obtained from children or mothers. Supplemented children had a higher proportion of males and stunted children, were younger and lived under poorer conditions as compared to controls. There was no difference in the prevalences of eye signs and symptoms of vitamin A deficiency in the two groups. Supplemented children had higher serum vitamin A concentrations than controls (1·4 (sd 0·49) μmol/l v. 1·2 (sd 0·52) μmol/l). The serum vitamin A concentrations were 1·6 (sd 0·45), 1·4 (sd 0·50), 1·3 (sd 0·44) and 1·1 (sd 0·43) μmol/l in children supplemented within 1, 1–6, 7–12 and 13–18 months of supplementation, respectively. Vitamin A concentrations were significantly greater than controls if supplementation was carried out within 6 months after adjustment. The oral megadose of 105 μmol vitamin A maintained serum vitamin A concentrations for 6 months in school children.