During the last decade, growth monitoring has been promoted us an important intervention for child survival, but questions have been raised about its electiveness and feasibility in less-developed countries. A growth-monitoring programme was carried out by the Bangladesh Rural Advancement Committee for over four years, covering about 20,000 children under two years of age. The programme was equally accessible to all socioeconomic groups and both sexes. Children were weighed monthly in village centres, and their mothers were given health and nutrition education. A recent evaluation found modest coverage (43 %) of the target children. Accuracy in determining ages of the target children was reasonably good, with more than 90% within 30 days of actual age. Eighty-seven per cent of the Salter round scales used gave accurate results, compared with only 17% of the Salter cylinder scales. Local volunteers, mostly women, participated in growth-monitoring sessions by weighing, recording, and demonstrating how to prepare supplementary diets. Growth monitoring was associated with increased use of selected child-survival interventions such as immunization. The nutrition status of participating children was not significantly better than that of a comparable group of children who did not participate (p =.051).