Heterogeneity of long-distance migration in studies of genetic structure.

One of the assumptions of migration matrix methods of population structure is that long-distance migrants are all sampled from a genetically homogeneous 'outside world'. This assumption has not often been tested. This paper examines migration and surname data from four towns in historical Massachusetts in order to examine this assumption and potential genetic effects of heterogeneous long-distance migration. Analysis of migration data shows that the rate of long-distance migration is significantly different for the four towns. The distributions of source populations for long-distance migrants into each town are significantly different. Surname analysis shows that in spite of the violation of the assumption of long-distance migrant homogeneity, there is little effect on the degree and pattern of within-group and among-group variation for these towns. This lack of effect seems related to genetic homogeneity of the long distance migrants.