Trauma, especially involving accidents and falls is now the leading cause of death in the UK in the first four decades of life. This paper looks at the existing pattern of Accident and Emergency Units and suggests that because it has grown up in a fragmented and poorly co-ordinated fashion and because it is subjected to demands that it was not designed for, it is inadequate in dealing with the most severe injury cases. Following collaborative work with researchers at the North Staffordshire Hospital, the authors propose a system of thirty major trauma centres to cover the whole of the UK. Each would deal with the most severe accident cases from surrounding hospitals. Using data from the census and from health authorities, analysed within a GIS environment, a locational pattern for these centres is suggested and their catchment areas are outlined.