Testing Wright’s Intermediate Population Size Hypothesis – When Genetic Drift is a Good Thing

In his 1931 monograph, Sewall Wright predicted genetic drift would overwhelm selection in small populations, and selection would dominate in large ones, but he also concluded drift could facilitate selection in populations of intermediate size. The idea that drift and selection would act together in populations of intermediate size has been almost completely ignored even as empirical evidence of rapid evolution associated with population bottlenecks has continued to accumulate. I used forward-time simulations with random mating and discrete generations to test the hypothesis that drift can facilitate selection. I find that selection generates biased distributions of Δq , and this bias is greatest for population sizes between 20 and 200, resulting in drift facilitation. Drift facilitation reduces the accumulation of drift load and segregation load for populations of intermediate size. Small populations accumulated higher levels of drift load and large populations maintained high levels of segregation load. Fixation of beneficial mutations is facilitated in intermediate populations when dominance is low and selection is weak. Accumulation of beneficial mutations over time (fixation flux) was higher across small to intermediate size populations and declined rapidly for large populations. Compared to predictions of fixation time for codominant beneficial mutations from diffusion equations, drift facilitation accelerates fixation in populations of intermediate size, but fixation time is slower in large populations when selection is weak. These results suggest drift facilitation in small and intermediate populations promote purging of genetic load and fixation of beneficial mutations, and may result in relatively rapid adaptation compared to large populations. Impact Summary After the recognition of Gregor Mendel’s contributions to our understanding of the inheritance of genetically-determined traits around 1900, there was confusion as to whether the type of variation Mendel studied could account for evolution by natural selection, as described by Charles Darwin. This controversy was resolved when three theoreticians (Ronald Fisher, Sewell Wright, and J.B.S. Haldane) published books that integrated Mendelian genetics with evolution. Their contributions (referred to as the Modern Synthesis), focused on evolutionary processes occurring within and among populations of a species, and established a mathematical foundation for our understanding of evolutionary biology. The mathematical models developed by the three architects of the modern synthesis, and those who followed, predicted that the effects of natural selection would be overwhelmed by random genetic changes (referred to as Genetic Drift) in small populations, and that genetic drift would be minimal, while selection would be most effective in large populations. Even though one of Wright’s major conclusions was that genetic drift and selection would work together (Drift Facilitation) to promote adaptive evolution in intermediate-sized populations, this idea has been almost completely ignored since it was first introduced in 1931. In this study, I use simulations of evolution in natural populations to evaluate the potential for drift facilitation to promote evolution in intermediate-sized populations. My work largely confirms Wright’s predictions; the removal of deleterious mutations and promotion of adaptive evolution are enhanced in population sizes ranging from about 10 to 200. These results indicate that a new paradigm for our understanding of evolution within populations is needed – one that emphasizes the importance of drift facilitation in small and intermediate size populations, and recognizes that periods of reduced population size are opportunities for enhanced levels of adaptive evolution. “In a population of intermediate size … there is continual random shifting of gene frequencies … which leads to a relatively rapid, continuing, irreversible, and largely fortuitous, but not degenerative series of changes, even under static conditions.” [Wright 1931][1], Page 157. ### Competing Interest Statement The authors have declared no competing interest. [1]: #ref-49