Infection and undernutrition in young children are thought to act synergistically. However, studies of the relationship between low height-for age (stunting) and morbidity in young children have had inconsistent findings and there are few adequate data on the effects of nutritional supplementation on morbidity. 129 stunted and 21 non-stunted children aged between 9 and 24 months, from poor Kingston neighbourhoods, identified from a house-to-house survey, were studied. The stunted children were randomly assigned to supplementation or no supplementation. Every week for 24 months the mothers were asked about the occurrence of any symptoms of illness. Supplementation had no consistent effect on the incidence or duration of symptoms. The stunted children had significantly more attacks of diarrhoea, fever, anorexia and apathy than the non-stunted children. The differences remained after controlling for social background and previous attacks of diarrhoea. There was also some indication of more severe illness in the stunted than in the non-stunted children.