Incongruent range dynamics between co‐occurring Asian temperate tree species facilitated by life history traits

Postglacial expansion to former range limits varies substantially among species of temperate deciduous forests in eastern Asia. Isolation hypotheses (with or without gene flow) have been proposed to explain this variance, but they ignore detailed population dynamics spanning geological time and neglect the role of life history traits. Using population genetics to uncover these dynamics across their Asian range, we infer processes that formed the disjunct distributions of Ginkgo biloba and the co‐occurring Cercidiphyllum japonicum (published data). Phylogenetic, coalescent, and comparative data suggest that Ginkgo population structure is regional, dichotomous (to west–east refugia), and formed ˜51 kya, resulting from random genetic drift during the last glaciation. This split is far younger than the north–south population structure of Cercidiphyllum (~1.89 Mya). Significant (recent) unidirectional gene flow has not homogenized the two Ginkgo refugia, despite 2Nm > 1. Prior to this split, gene flow was potentially higher, resulting in conflicting support for a priori hypotheses that view isolation as an explanation for the variation in postglacial range limits. Isolation hypotheses (with or without gene flow) are thus not necessarily mutually exclusive due to temporal variation of gene flow and genetic drift. In comparison with Cercidiphyllum, the restricted range of Ginkgo has been facilitated by uncompetitive life history traits associated with seed ecology, highlighting the importance of both demography and lifetime reproductive success when interpreting range shifts. Asian Ginkgo population structure is regional and dichotomous (to west‐east refugia) resulting from random genetic drift during the last glaciation. This structure has persisted despite (recent) unidirectional gene flow and dates from ~39 k years ago mirroring the lag before FST reaches gene flow‐drift equilibrium. The temporal dynamics of gene flow in this system challenge existing isolation hypotheses (with or without gene flow) proposed to explain the variance in post‐glacial expansion to former range limits among species of temperate deciduous forests in eastern Asia. We argue that both demography and life history traits are required to interpret these range shifts.