Let's assume the research is sound. (Inferring from mci's comment, probably not.) But let's suppose we saw this phenomenon in a specific country's population,

Let's assume the research is sound. (Inferring from mci's comment, probably not.) But let's suppose we saw this phenomenon in a specific country's population, measured without any sampling bias. It could be from: - selection pressure for some other trait that happens to involve some genetic material that affects the forearm arteries. - the recent removal of strong selection pressure for having two arteries, on a trait that generally tends to drift towards three arteries in many directions. - a dominant gene isolated to one province, which then got diluted across the population when people traveled more. - a trait found more often in the rural population, which over the past century had more children.