Search

Habitat fragmentation can significantly affect mating and pollen dispersal patterns in plant populations, although the differential effects of the various aspects of fragmentation are poorly understood. In this study, we used eight microsatellite loci to investigate the effect of fragmentation on the mating system and pollen dispersal within one large and eight small population remnants of Banksia sphaerocarpa var. caesia, a bird-pollinated shrub in the southern agricultural region of Western Australia. The large population had a much larger neighbourhood size and lower selfing rate, maternal pollen pool differentiation and within-plot mean pollen dispersal distance than the small populations. Outcrossing was consistently high and ranged from 85.7% ± 2.6 to 98.5% ± 0.9, and mating patterns suggested nearest-neighbour pollination. Pollen immigration into small populations ranged from 2.8% ± 1.8 to 16.5% ± 3.2. Using the small populations, we tested for correlations between various fragmentation variables and mating system and pollen dispersal parameters. We found significant negative linear relationships between population isolation and outcrossing rate; population shape and neighbourhood size; and conspecific density and mean pollen dispersal distance. There were significant positive linear relationships between population shape and pollen pool differentiation and between population size and number of different fathers per seed crop. Our results suggest that birds may use a series of fragmented populations as a vegetation corridor while foraging across the landscape and that population connectivity is a critical determinant of pollinator visitation. Our results also suggest that the effect of a linear population shape on the mating system and pollen dispersal is routinely underestimated. [ABSTRACT FROM AUTHOR]

The evolutionary potential of plant species that reproduce via predominantly clonal means and the conditions under which clonality is favoured are not well known. Long-term clonal reproduction is expected to result in a number of readily detectable genetic signals not present in populations that reproduce by sexual means. We use a hierarchical sampling strategy to assess genotype probabilities and confirm that two rare sister taxa of Banksia ionthocarpa have contrasting modes of reproduction. Banksia ionthocarpa subsp. chrysophoenix reproduces clonally. Populations had low levels of genotypic diversity and were comprised of large clonal patches consisting of many ramets that covered hundreds of square metres and showed little intermixing. The taxon was genetically depauperate (mean Na = 1.97, mean P = 0.66, mean He = 0.282), had high levels of genetic differentiation between populations ( θ = 0.481), and populations exhibited excess heterozygosity and linkage disequilibrium (LD) among loci, suggesting historically high levels of clonality. In contrast, the sister taxon B. ionthocarpa subsp. ionthocarpa, which occurs in an area with more than twice the annual rainfall and less extreme minimum and maximum temperatures, showed no evidence of clonality, high levels of genotypic diversity, greater genetic diversity (mean Na = 3.31, mean P = 0.81, mean He = 0.405), lower levels of genetic differentiation between populations (θ = 0.253) and no evidence of excess heterozygosity or LD among loci. We suggest that the development of clonality in subsp. chrysophoenix is associated with its more marginal environment and enhanced by response to recurrent fires. [ABSTRACT FROM AUTHOR]

The integration of phylogenetics, phylogeography and palaeoenvironmental studies is providing major insights into the historical forces that have shaped the Earth's biomes. Yet our present view is biased towards arctic and temperate/tropical forest regions, with very little focus on the extensive arid regions of the planet. The Australian arid zone is one of the largest desert landform systems in the world, with a unique, diverse and relatively well-studied biota. With foci on palaeoenvironmental and molecular data, we here review what is known about the assembly and maintenance of this biome in the context of its physical history, and in comparison with other mesic biomes. Aridification of Australia began in the Mid-Miocene, around 15 million years, but fully arid landforms in central Australia appeared much later, around 1–4 million years. Dated molecular phylogenies of diverse taxa show the deepest divergences of arid-adapted taxa from the Mid-Miocene, consistent with the onset of desiccation. There is evidence of arid-adapted taxa evolving from mesic-adapted ancestors, and also of speciation within the arid zone. There is no evidence for an increase in speciation rate during the Pleistocene, and most arid-zone species lineages date to the Pliocene or earlier. The last 0.8 million years have seen major fluctuations of the arid zone, with large areas covered by mobile sand dunes during glacial maxima. Some large, vagile taxa show patterns of recent expansion and migration throughout the arid zone, in parallel with the ice sheet-imposed range shifts in Northern Hemisphere taxa. Yet other taxa show high lineage diversity and strong phylogeographical structure, indicating persistence in multiple localised refugia over several glacial maxima. Similar to the Northern Hemisphere, Pleistocene range shifts have produced suture zones, creating the opportunity for diversification and speciation through hybridisation, polyploidy and parthenogenesis. This review highlights the opportunities that development of arid conditions provides for rapid and diverse evolutionary radiations, and re-enforces the emerging view that Pleistocene environmental change can have diverse impacts on genetic structure and diversity in different biomes. There is a clear need for more detailed and targeted phylogeographical studies of Australia's arid biota and we suggest a framework and a set of a priori hypotheses by which to proceed. [ABSTRACT FROM AUTHOR]

Pollen dispersal was investigated in six populations of Calothamnus quadrifidus, a bird-pollinated shrub in the fragmented agricultural region of southern Western Australia. Paternity analysis using six microsatellite loci identified a pollen source within populations for 67% of seedlings, and the remainder were assumed to have arisen from pollen sources outside the populations. Outcrossing was variable, ranging from 5% to 82%, and long-distance pollen dispersal was observed in all populations with up to 43% of pollen sourced from outside the populations over distances of up to 5 km. This extensive pollen immigration was positively associated with population size but not isolation. Comparison of two populations of similar size but different density showed greater internal pollination and less selfing in the denser population, suggesting an influence of density on pollinator behaviour. The study revealed extensive long-distance pollen dispersal for C. quadrifidus within this fragmented agricultural landscape and highlighted the interaction between reserve populations and isolated road verge remnants in maintaining genetic connectivity at the landscape scale. [ABSTRACT FROM AUTHOR]

AbstractThe population genetic structure of the Australian plant Lambertia orbifolia was investigated for chloroplast DNA (cpDNA) and rDNA based on restriction fragment length polymorphism. Variation was assessed in 14–20 individuals from six populations with probes covering the majority of the chloroplast genome and the whole rRNA gene unit. For cpDNA, eight mutations were detected which were distributed over five haplotypes. Nucleotide diversity in the species was high and the majority of this diversity was distributed between populations with diversity within populations restricted to a single population. There was significant differentiation between the two regions in the species distribution with the Narrikup region being distinguished by a single haplotype that was characterized by six unique mutations. Variation in rDNA was detected with three gene length variants present in most individuals. However, the Narrikup region was characterized by homogenization of the gene unit to a single length variant in all individuals. The divergence of the Narrikup region suggests that the disjunction in the species distribution has been present for a long time and the two regions represent separate evolutionary lineages. [ABSTRACT FROM AUTHOR]