Feeding of breast milk in the first weeks of life appears to have a strong protective effect against necrotising enterocolitis. Nevertheless breast milk also seems to be positively linked to the development of jaundice and to late haemorrhagic disease in infants who have not received vitamin K supplements. There is no consistent evidence that other childhood conditions such as insulin dependent diabetes or cancer are less prevalent among children who have been breast fed. Among adult conditions suggested to be less prevalent in the breast fed, only single reports of significant findings for multiple sclerosis and breast cancer exist and convincing corroboration is not available. There are a number of studies that indicate a relationship between breast feeding and later cholesterol levels--and one that has considered the mortality of ischaemic heart disease among adult males. There is some suggestion that breast feeding (during the first year of life) is the optimal protection against future raised lipid levels and mortality from coronary heart disease, but the evidence is far from conclusive. The major health advantage of breast feeding that has been clearly demonstrated remains in the protection of the infant from certain infections in early life. If there are other long-term health advantages they have yet to be fully elucidated and confirmed.