The effect of chronic iron intake on diarrhoeal disease was evaluated in children in a community of low socio-economic stratum in Santiago, Chile. Children were incorporated into each of two consecutive cohorts; each cohort was divided into two groups, one receiving iron-enriched milk (12 mg/l) (monthly average = 70 children) and the other a control milk (1 mg/l) (monthly average = 83 children), and each cohort was followed up for 6 months. The incidence of diarrhoea was higher among the iron-supplemented children (30.4 vs 25.5 episodes/100 children/month, P < 0.025). This was mainly due to results obtained in infants 3-8 months of age during the summer months. Supplemented infants had more bowel movements on day 1 (P < 0.03) and liquid or semi-liquid stools were passed for more than 15 days more frequently (P < 0.05). While no differences were detected in aetiology, Shigella-associated episodes were less common among iron-supplemented infants (P < 0.008). Asymptomatic shedding of enteropathogens significantly increased in infants 12-18 months of age receiving iron-supplemented milk. In areas with inadequate environmental sanitation, chronic iron supplementation may have negative effects on diarrhoeal morbidity, despite improving iron nutritional status.