The relative contribution of drift and selection to phenotypic divergence: A test case using the horseshoe bats Rhinolophus simulator and Rhinolophus swinnyi

Ecol Evol. 2017 May 9;7(12):4299-4311. doi: 10.1002/ece3.2966. eCollection 2017 Jun.

Abstract

Natural selection and drift can act on populations individually, simultaneously or in tandem and our understanding of phenotypic divergence depends on our ability to recognize the contribution of each. According to the quantitative theory of evolution, if an organism has diversified through neutral evolutionary processes (mutation and drift), variation of phenotypic characteristics between different geographic localities (B) should be directly proportional to the variation within localities (W), that is, B W. Significant deviations from this null model imply that non-neutral forces such as natural selection are acting on a phenotype. We investigated the relative contributions of drift and selection to intraspecific diversity using southern African horseshoe bats as a test case. We characterized phenotypic diversity across the distributional range of Rhinolophus simulator (n = 101) and Rhinolophus swinnyi (n = 125) using several traits associated with flight and echolocation. Our results suggest that geographic variation in both species was predominantly caused by disruptive natural selection (B was not directly proportional to W). Evidence for correlated selection (co-selection) among traits further confirmed that our results were not compatible with drift. Selection rather than drift is likely the predominant evolutionary process shaping intraspecific variation in traits that strongly impact fitness.

Keywords: Lande's model; adaptation; diversification; micro‐evolutionary forces; natural selection; neutral evolution; speciation; vicariance.