Eighty school children (47 boys and 33 girls) from a rural community ranging in age from 10 to 16 years and of low socio-economic status were surveyed for riboflavin deficiency. The overall prevalence of subclinical riboflavin deficiency was 41 per cent based upon an activity coefficient (AC) of erythrocyte glutathione reductase (EGR) of more than 1.30. Although there was no significant difference in the overall prevalence of riboflavin deficiency between the boys and girls, the EGR-AC values correlated significantly with dietary riboflavin intake (r = -0.48, P less than 0.002) for the boys only. The regression analyses of riboflavin intake on EGR-AC values indicated that a minimum daily intake of 0.70 mg riboflavin or 0.4 mg/1000 kcal corresponded to acceptable EGR-AC values (less than 1.20) for the boys. Dietary data showed that the intake of energy, thiamine and riboflavin was grossly inadequate for the needs of the children. About 15 per cent of the total protein intake and 14.3 per cent of the total riboflavin intake were obtained from animal products with plant sources contributing the bulk of the intake. Milk intake was very low or non-existent. It is suggested that dietary intakes of large quantities of leafy vegetables and the use of fermented products and dried shrimps in the soup or an oral vitamin supplement would improve their riboflavin intake.