This study investigated the relationship between breast-feeding and early childhood morbidity. Data on morbidity had been recorded in the Continuous Morbidity Registration of the Department of Family Medicine of the University of Nijmegen, the Netherlands, since 1967. Information on early childhood feeding was collected retrospectively by questionnaires mailed to the parents of all the children; the response rate was 94%. Exposure and outcome data of 1,347 children were available for analyses. Duration of breast-feeding was categorized as follows: 0 days (no breast-feeding), 1-14 days, 15-30 days, 31-90 days, 91-180 days, and more than 180 days. Two thirds of the children had been breast fed. Generally there was an inverse relationship between breast-feeding and morbidity. This was most prominent in the first year of life but was also present in the first three years. After adjustment for potential confounding variables, duration of breast-feeding was associated with fewer morbidity episodes and lower rates of several specific illnesses during the first three years of life. The findings suggest some modest health benefits of breast-feeding for children in the Netherlands.