A beneficial effect of periodic vitamin A supplementation on childhood mortality has been demonstrated, but the effect on morbidity is less clear. We investigated the effect of vitamin A supplementation on diarrhoea and acute lower-respiratory-tract infections (ALRI) in children from northeastern Brazil in a randomised, double-blind, placebo-controlled community trial. 1240 children aged 6-48 months were assigned vitamin A or placebo every 4 months for 1 year. They were followed up at home three times a week, and data about the occurrence and severity of diarrhoea and ALRI were collected. Any child with cough and respiratory rate above 40 breaths per min was visited by a paediatrician. The overall incidence of diarrhoea episodes was significantly lower in the vitamin-A-supplemented group than in the placebo group (18.42 vs 19.58 x 10(-3) child-days; rate ratio 0.94 [95% Cl 0.90-0.98]). The benefit of supplementation was greater as regards severe episodes of diarrhoea; the incidence was 20% lower in the vitamin A group than in the placebo group (rate ratio 0.80 [0.65-0.98]). With the standard definition of diarrhoea (> or = 3 liquid or semi-liquid stools in 24 h) the effect of vitamin A on mean daily prevalence did not reach significance, but as the definition of diarrhoea was made more stringent (increasing number of stools per day), a significant benefit became apparent, reaching for diarrhoea with 6 or more liquid or semi-liquid stools in 24 h a 23% lower prevalence. We found no effect of vitamin A supplementation on the incidence of ALRI. The reduction in severity of diarrhoea may be the most important factor in the lowering of mortality by vitamin A supplementation.