Abstract
Subspecies are traditionally defined using phenotypic differences associated with different geographical areas. Yet patterns of morphological and genetic variation may not coincide and thereby fail to reflect species’ evolutionary history. The division of the shrub Banksia nivea Labill. into one widespread (B. nivea subsp. nivea) and two geographically localized subspecies (B. nivea subsp. uliginosa (A.S. George) A.R. Mast & K.R. Thiele and B. nivea subsp. Morangup (M. Pieroni 94/2)) in south-west Australia has been based mainly on variation in leaf shape and pistil length, although flowering time and habitat differences are also evident, and subsp. uliginosa occurs on a different substrate. To assess the genetic divergence of B. nivea subspecies, we genotyped representatives from each subspecies for nuclear microsatellite and non-coding chloroplast sequence variation. We used distance and parsimony-based methods to assess genetic relatedness. Patterns were consistent with the existing taxonomy of subsp. nivea and uliginosa but not subsp. Morangup. Phylogenetic analyses revealed evidence for a more recent divergence of subsp. uliginosa associated with expansion from dryer sandy soils into the winter-wet ironstone soils in the southwest of Western Australia, consistent with progressive long-term climatic drying. Nuclear microsatellites showed low to moderate diversity, high population differentiation overall, and genetic structuring of subspecies in different biogeographical areas. We propose this pattern reflects the predicted impact of a patchy distribution, small populations, and restrictions to gene flow driven by both distance and biogeographic differences in subspecies’ habitats.
Highlights
Many recognized species are not genetically uniform and may be highly structured into historically isolated populations that may warrant consideration as intraspecific units [1].Taxonomically recognized subspecies are often based on geographically discontinuous morphological differences [2,3] or ecotypic differences [4]
Haplotype diversity was very low, and most haplotypes were population-specific (86.7%) two haplotypes were found in multiple subsp. uliginosa populations (Figure 1b)
Genetic diversity was generally lower in the localized subsp. uliginosa than in the widespread subsp. nivea, differences associated with range were not significant, contrary to expectations based on meta-analysis across Australian plants [25]
Summary
Recognized subspecies are often based on geographically discontinuous morphological differences [2,3] or ecotypic differences [4]. Natural phenotypic or ecotypic diversity within species over wide distributions may not be consistent with genetic divergence representing the evolutionary processes within species [5,6]. Genetic divergence within species is influenced by gene flow [7] and affected by geographic (e.g., topography, distance) and environmental (habitat, climate, pollen, and seed dispersal) factors [8]. Reciprocal monophyly is not expected for subspecies, some evidence of restricted gene flow between diverging taxa is expected in patterns of neutral genetic variation [6]. Given there are examples of lineage divergence associated with habitat specialization [9,10,11,12], widespread species containing subspecies that occupy different habitats might be expected to show genetic differentiation among habitats
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
