Molecular phylogeny and population genetics of the Grevillea Thelemanniana group (Proteaceae)

Species are the basis for biological studies and ultimately the most important unit for conservation strategies. The definition and delimitation of species has been debated heavily in taxonomy, and no all encompassing definition has been defined yet. The recent expansion of molecular technology, allowing for larger sample sizes and multi-locus datasets, has shifted the field of systematics from morphological or single locus taxonomy towards integrative taxonomy applying multiple data sources. New concepts have been created which utilize population genetic theory to analyse multi locus data and discover patterns of evolutionary lineages to aid in discovering species limits and boundaries. By using molecular data from different regions of the genome, researchers can better understand evolutionary patterns. Additionally, new software employing Bayesian methods and molecular dating has increased our understanding of evolutionary patterns and speciation. Understanding the different patterns which make up species will ultimately lead to better conservation decision making.
In this study the Grevillea Thelemanniana group of Western Australia was investigated using chloroplast sequences from two regions and 12 microsatellites. Microsatellites were specifically designed for Grevillea thelemanniana ssp. thelemanniana and cross amplified across the group. The Thelemanniana group is made up of 16 species, four subspecies and two phrase named species that, based on morphology, likely have affiliations to the group. The resulting gene tree did not support all the original taxonomically defined species and species relationships and did not support the Thelemanniana group as monophyletic in relation to species from other taxonomically defined groups. However, two significantly different lineages were recovered from the chloroplast analyses, which corresponded to a northern and southern distribution. Using molecular methods and dating technology a significant lineage divergence within the Thelemanniana group was evident, dating back to the Miocene. The patterns of evolution within lineages in part support evolutionary hypothesis of the Southwest floristic region of Australia, but also illustrated alternate patterns.
Incongruence between resulting gene tree and species tree and also incongruence between sequence and microsatellite data, has often been attributed to incomplete lineage sorting (deep coalescence), historical introgression and/or hybridization.
Although systematical studies often discuss these issues, few have enough data to distinctly resolve them. In the second part of this thesis I aimed to resolve a subclade within the Thelemanniana group to address potential issues of incomplete lineage sorting, hybridization and introgression. I collected greater sample sizes from multiple populations for four taxon, applying chloroplast sequences and microsatellites, to address different levels of divergence between species and populations and delimit species boundaries. Results indicated molecular support for most species as unique and little evidence for recent introgression or hybridization.
Finally, I used a landscape genetic approach using microsatellites and geographic data to discover the spatial genetic structure of Grevillea thelemanniana ssp. Cooljarloo. Landscape genetics aims to provide information on how landscape features interact with evolutionary processes such as gene flow, migration and drift. Grevillea thelemanniana ssp. Cooljarloo is restricted to a 20 Km2 region and is threatened by an active and expanding mine site within the region. The population sizes were large and continuous; however there appears to be habitat selection for creek lines and floodplains. Through riparian models and spatial Bayesian clustering analyses, regional clustering of creek line subpopulations and strong divergence between creek lines was revealed. Also, a surprising amount of genetic admixture from multiple sources was witnessed in flood plains which could be explained by regular flooding and pollinator movements.
The results from this study are important in explaining the evolutionary past of Grevillea, the third largest genus of flowering plants in Australia, and also adding to the greatly understudied southwest of Western Australia biodiversity hotspot. This work is one of the first to investigate the molecular phylogeny of Grevillea, and the first to do so on a Western Australian group. The results are important in informing conservation decisions as the majority of species within the Thelemanniana group are rare or threatened.