Microdifferentiation in historical Massachusetts: a comparison of migration matrix and isonymy analyses

Historical studies of population structure allow assessment of the effects of demographic change on genetic variation among and within populations. This paper investigates the degree of microdifferentiation among four towns in north-central Massachusetts during the 18th and 19th centuries, a time of rapid population growth. Migration matrices were derived from 4,223 marriage records from 1785 through 1849 and then used to estimate genetic kinship using the methods of Imaizumi et al. (Genetics 66:569-582, 1970) and Harpending and Jenkins (In: Genetic Distance. New York: Plenum Press, 1974). Kinship matrices based on isonymy were also derived using the surnames of 4,039 marriages according to Morton's (In: Genetic Structure of Populations. Honolulu: University Press of Hawaii, 1973) methods. In addition, an isonymy distance matrix was constructed based on 4,659 marriage records from 1741 through 1849 to compare surname frequencies across space and time simultaneously. For both migration and isonymy matrices, the degree of micro-differentiation among the four towns, RST, was computed for each of six time cohorts. Both migration and isonymy show a reduction in among-group variation over time, reflecting population growth and stable, but high, rates of long-distance migration. Estimates of RST from migration and isonymy are very similar. The one notable exception reflects known historical events. Comparison of the kinship estimates for individual towns (rii) also shows close correspondence between migration and isonymy. The differences seem to reflect a tendency for migration matrices to overestimate kinship in earlier generations relative to isonymy. A principal coordinates analysis of the isonymy distance matrix shows the relative isolation of towns from one another in all time periods but with a reduction in differentiation over time.