ABSTRACT
In human populations, the correct historical interpretation of a genetic structure is often hampered by an almost inherent inability to differentiate between ancient and more recent influences upon extant gene pools. One method to trace recent population movements is the analysis of surnames, which, at least in Central Europe, can be thought of as traits 'linked' to the Y chromosome. Illegitimacy, extramarital birth and changes of surnames may have substantially obscured this linkage. In order to assess the actual extent of correlation between surnames and Y-chromosomal haplotypes in Central Europe, we typed Y-chromosomal short tandem repeat markers in 419 German males from Halle. These individuals were subdivided into three groups according to the origin of their respective surname, namely German (G), Slavic (S) or 'Mixed' (M). The distribution of the haplotypes was compared by Analysis of Molecular Variance. While the M group was indistinguishable from group G (PhiST=-0.0008, P>0.5), a highly significant difference (PhiST=0.0277, P<0.001) was observed between the S group and the combined G+M group. This surprisingly strong differentiation is comparable to that of European populations of much larger geographic and linguistic difference. In view of the major migration from Slavic countries into Germany in the 19th century, it appears likely that the observed concurrence of Slavic surnames and Y chromosomes is of a recent rather than an early origin. Our results suggest that surnames may provide a simple means to stratify, and thereby to render more efficient, Y-chromosomal analyses of Central Europeans that target more ancient events.