This paper studies the effect of the long-term relatedness between countries, measured by their genetic distance, on educational migrant selection. Analyzing bilateral migrant stocks of the 15 main destination countries and 85 sending countries for the year 2000, we find that migrant selection and genetic distance follow a nonlinear J-shaped pattern: at low levels of genetic distance, increases in genetic distance reduce the positive selection of migration. However, at higher levels of genetic distance, this pattern is reversed and migration becomes more positively selected. We complement this finding by showing that the net benefits of genetic distance are strongly decreasing for low-skilled migrants with increasing genetic distance, while high-skilled migrants are less responsive to genetic distance in general. Results are robust to conditioning on bilateral control variables, including various destination- and sending-country-specific fixed effects and applying an instrumental-variables approach that exploits exogenous variation in genetic distances in the year 1500.