A comparison of inbreeding rates in India, Japan, Europe and China.

This is a report on an application of a new method for estimating inbreeding rates in large, semi-isolated populations over time. This study is intended to further show the versatility of the method. In a previous study, the method was shown to be useful in analysing the effect of details in a single population curve on inbreeding. In this study, the method is applied to the population curves of India, Japan, Europe and China over the past two millennia, to allow inter-comparisons of the inbreeding within those populations. Anthropologists traditionally concentrate on small isolates within a country for the estimation of inbreeding. Those isolates, however, might not be representative of the country as a whole. The method used in this study attempts to estimate inbreeding over large regions over an extended period of time. The method models the genealogical 'paradox' and yields estimates of the average inbreeding in terms of Pearl's coefficient Z, as a function of time. It is first assumed that the whole population of each country is the adult (breeding) population, corresponding to minimum inbreeding. It is found that the more complex the population curve examined, the greater the precision of the Z curves obtained. The effect of incorporating a single known value of inbreeding into the analyses is also investigated. This procedure produces a more realistic situation where the adult (breeding) population is considerably less than the entire population. It is shown, that if it is assumed that the whole population of the country is the adult (breeding) population, then a present day person from one of the four regions/countries studied is descended from between 72% and 97% of the early medieval population of the particular country. On the other hand, if a known value of inbreeding for Britain is incorporated into the analysis, these values become of the order of 1% for the older-settled regions/countries, and about 16% for Japan. However, that value for Japan is reduced to about 1% when a known value of inbreeding for this country is used. Although some uncertainty in these results remains, the versatility of the method demonstrated here will provide more accurate results, as better input-data become available.