2018/07/09 by Matthew J. Christmas, Ed Biffin, Andrew J. Lowe
Biochemistry, Genetics and Molecular Biology · Environmental Science · #Biology #Ecology and Vegetation Dynamics Studies #Evolutionary biology #Genetic diversity and population structure #Genetic variation #Genetics #Species Distribution and Climate Change #Subspecies #Zoology
paper · doi:10.1071/bt17046
openalex publication_date 2018/07/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/21
Subspecies are traditionally defined on the basis of geographic discontinuities in phenotypic traits, and their circumscription is useful to distinguish morphologically differentiated populations. However, the robustness of morphology-based subspecies classifications in the genomics era is coming under increasing scrutiny, and phylogenies inferred from molecular data may not match with morphological approaches. The division of the shrub Dodonaea viscosa into seven subspecies within Australia has been based mainly on variation in leaf shape, which is a notably variable phenotypic character in this species. So as to assess the alignment between genetic variation and subspecies assignment, we genotyped 67 D. viscosa plants, including representatives from each of the seven subspecies, for 941 single nucleotide polymorphisms. We used network- and Bayesian-based methods to assess genetic relatedness between sampled individuals. Structure analysis identified two genetic clusters, with a further substructure being identified within one of the clusters. Genetic clusters partially aligned with subspecies classifications, particularly for the three most morphologically distinct subspecies (ssp. mucronata, ssp. viscosa and ssp. burmanniana). Subspecies inhabiting the arid zone (ssp. mucronata and ssp. angustissima) exhibited the most distinct genetic clustering. For subspecies inhabiting more temperate regions of its range (ssp. angustifolia, ssp. cuneata and ssp. spatulata), genetic groups did not correspond well with subspecies classifications, but rather were better explained by the geographic origin of individuals. We suggest that the current subspecific classification of the hopbush does not accurately reflect the evolutionary history of this species, and recommend that phenotypic variation be reassessed in light of the genetic structure we describe here. The roles of environmental change, selection and geographic isolation are discussed in an attempt to explain the contemporary distribution of genetic variation in D. viscosa in Australia.