The genetic structure of the British populations and their surnames

Until large-scale ancient DNA studies of archaeological samples become practical, genetic studies of current human populations, with a few caveats, are a productive way to investigate past movements of peoples (Røyrvik 2010). Recent advances in techniques for typing and analysing thousands of genetic markers in large numbers of individuals offer exciting opportunities to investigate population structure at a fine geographic level. For example, investigating population structure between regions of the British Isles can shed light on the impact people from different parts of Europe had on Britain. These methods are a natural progression from earlier studies on a small number of genetic markers into the underlying genetic structure of Europe and the processes that are the likely causes. One of the earliest such informative studies was based on data from the human leukocyte antigen (HLA)region, a gene rich region, many of which are associated with the immune response. Menozzi et al. (1978) used a principle components analysis (PCA) approach to demonstrate that there was a genetic gradient from the Middle East to north-west Europe. This gradient suggested that there was a gradual migration of people, leading to a corresponding dilution of difference between the origin and extremity. This is likely to have started in the Middle East about 10,000 years ago, moving north-west at about 1 km a year, in parallel with agriculture, and reaching the UK about 5,000 years ago. When investigating smaller geographic area, a good example of hidden heterogeneity is ‘Little England beyond Wales’ (Watkin 1956; 1960). This is the region of south Pembrokeshire that is linguistically and toponymically different from the rest of the region, most likely due to settlement of Vikings in the 9th century AD(Loyn 1976) and/or Flemish farmers, settled by Henry I in the early 12th century (MacKintosh 1866). Watkin (1956; 1960) observed that there was a higher frequency of blood group ABOtype A in that region than amongst the Welsh in general. He thought that this was likely to be caused by Viking settlement because the Norwegians had a similar frequency of A. Contradicting this, a PCA analysis of Y chromosome data, that included samples from Haverfordwest, groups Haverfordwest with the rest of Wales and appears to rule out a Norse contribution, although this analysis is only of the male lineage (Capelli et al. 2003).