Throughout the twentieth century there has been academic debate about the relationship between early life experience and adult health. This thesis examines the origins of that debate, its manifestation in different scientific fields of inquiry, and its recent re-emergence in epidemiological research. It shows how the changing nature of the debate was inextricably related to changes in the notion of adult health, the development of methods of empirical investigation, and the consequent availability of scientific evidence. The thesis therefore spans a number of disciplines and draws on the relevant knowledge from each. During the first forty years of this century the debate was policy-led. Adult health was assumed to depend on child health but empirical investigations of the link were limited. Research concentrated on early life factors, such as the behaviour of the mother, which were thought to influence morbidity and mortality in infancy and childhood. The focus of the academic debate was whether these factors were environmental or genetic in origin. This debate had important policy implications. Public acceptance of the significance of the early environment provided the rationale for the emerging infant and child health services. More recent interest in the influence of early life on adult health sprang from two different hypotheses about the basis of adult health, both of which focus on adult chronic disease. The first hypothesis puts forward the view that adult lifestyle is the main source of risk for chronic disease, and considers early life factors only to the extent that they are associated with the development of healthy and unhealthy lifestyles. The interdisciplinary, American dominated debate associated with this hypothesis is policy-led, and lacks a common conceptual model for understanding the risk processes that may be involved in the lifetime development of health related behaviour. In contrast the second hypothesis gives a causal and dominant role to environmental factors during critical periods of growth in utero and infancy which affect particular body systems, with long term consequences for adult chronic disease. This epidemiological research is science-led and dominated by one British investigator (Professor DJP Barker) and his research team who have developed the concept of 'environmental programming'. The academic debate associated with this hypothesis concerns the interpretation that is given to research findings which show associations between early life and adult chronic disease. Whereas for Barker they are evidence of a biological process occurring at the beginning of life, for others they reflect continuity of social deprivation throughout life. It is the relative influence of the intrauterine, childhood and adult environment which is in dispute. The thesis addresses these questions in respect of cardiovascular and respiratory disease by drawing on data from the Medical Research Council's National Survey of Health and Development, a unique prospective study of the health and development of over 5000 men and women followed up since their birth in March 1946. Evidence is presented which suggests that the effect of early life factors on adult blood pressure, lung function and overall health status is irrespective of later socioeconomic experiences, thus providing support for the environmental programming model. A second model, based on Rutter's concept of 'chains of risk' is put forward to explain the lifetime development of health related behaviour, and is used to account for the links between adult physical activity and early social, educational and individual characteristics which are found in data from the National Survey. In the recent debate there has been little engagement with the policy process, although Barker's new theories have attracted considerable public attention. The thesis draws on its historical reviews of epidemiological research and child health services to consider what effects the evidence presented may have on health and social policies for children in the 1990s.