Phoon, Sook-Ngoh (2015) Systematics and biogeography of Elaeocarpus (Elaeocarpaceae). PhD thesis, James Cook University.

Preview

PDF (Thesis) Download (9MB) | Preview

Abstract

The genus Elaeocarpus (family Elaeocarpaceae) comprises 350 – 400 species most of which are rainforest trees distributed in palaeo-tropical regions (except mainland Africa). The morphology of Elaeocarpus has been well documented and has been the basis for the intuitive infrageneric classification systems developed to date. The phylogenetics and evolutionary history of Elaeocarpus have received limited attention, however. Phylogenetic studies to date have been based on few markers and species sampling that was very restricted and biased towards Australian taxa. This thesis uses a much-expanded sample size including representatives from various biogeographical regions and a large DNA sequence dataset comprising over 3000 bp from four regions (plastid psbA-trnH intergenic spacer, trnL-trnF region and trnV-ndhC intergenic spacer, and nuclear Xdh) to address the following aims: (1) investigate phylogenetic relationships between and within Elaeocarpus using both non-parametric (maximum parsimony), and parametric model-based (maximum likelihood and Bayesian Inference) methods, (2) trace the transformation of seed morphological characters (embryo shape and endosperm ornamentation), which are considered as superior in the current infrageneric classifications, on the estimated phylogeny to identify morphological synapomorphies for molecular clades, and (3) estimate the divergence times of lineages, infer the origins and explain current distribution patterns of the genus using an uncorrelated lognormal relaxed molecular clock and five fossil calibration points (four in Elaeocarpus and one in Sloanea). Extending from the molecular evidence developed here, this study also aimed to delimit taxon boundaries within the polystachyus group (i.e. the polystachyus clade + E. polystachyus or the Elaeocarpus polystachyus complex) using morphometrics, a statistically testable and repeatable method and then to compare congruency between the results of the morphometrics and the alpha taxonomy. Finally, this study aimed to provide precursory evidence that morphological similarities within the polystachyus group are not correlated with ecological adaptations, and each taxon does maintained its own morphological characteristics but the observation of these characters might have been limited by the alpha taxonomy method.

The results provide strong support for the monophyly of Elaeocarpus and for its sister relationship with Aceratium. Within the Elaeocarpus clade, E. holopetalus is resolved as a distinct lineage that is placed sister to the remainder of the taxa. Apart from E. holopetalus, a total of 13 main lineages or clades are resolved: E. sedentarius, the obovatus, section Elaeocarpus, ganitrus, group VI, monocera, group VII, group XI subgroup B, acronodia, polystachyus, coilopetalum, New Zealand and New Caledonian groups. All of the clades resolved in the present study are broadly congruent with the current infrageneric classifications, except the obovatus and the New Zealand clades, which are part of group V subgroup D.

The parsimony reconstructions of the ancestral states of two selected seed morphological characters, embryo shape and endosperm ornamentation indicate that both are homoplasious at higher taxonomic levels (i.e. genus level and above), although the curved embryo is homologous within Elaeocarpus (excluding E. holopetalus).

The large sample size with many representatives from various biogeographic regions and much-improved resolution of the phylogenetic relationships within Elaeocarpus provided a strong foundation to investigate the spatio-temporal evolution of this genus comprehensively for the first time. Elaeocarpaceae and its sister (Cunoniaceae + Cephalotaceae) diverged in the late Cretaceous, and diversification within the family (the crown age) is estimated to have begun at c. 83 Mya. Within the family, most of the infrafamilial lineages resolved are congruent with the current infrafamilial groupings (Coode 2004): the Sloanea alliance (Vallea, Aristotelia and Sloanea), the Tremandraceous genera (Platytheca, Tetratheca and Tremandra), and the Elaeocarpus alliance (Sericolea, Aceratium and Elaeocarpus). The exception is the Crinodendron alliance; Crinodendron and Peripentadenia formed a clade but Dubouzetia was placed sister to a clade comprising the Tremandraceous genera and the Elaeocarpus alliance.

The results of historical biogeographic analysis using Fitch parsimony and Dispersal-Extinction-Cladogenesis methods, and molecular dating analysis using Bayesian relaxed-clock methods suggest that Elaeocarpus diverged from its sister – Aceratium – in the Eocene in Australia. Early diversification of Elaeocarpus in Australia occurred when the continent was still at high latitudes and largely covered with megathermal rainforests. Following this, migration events occurred to the surrounding regions, i.e. New Guinea, Central Malesia, West Malesia, New Zealand and the Pacific islands, and further northwards into continental Asia and Madagascar probably via West Malesia. Several reversal migrations are also postulated. Radiation of Elaeocarpus within New Guinea and Borneo, the two current centres of species diversity, may have coincided with mountain building in the Miocene. Geological and climatic changes and zoochorous dispersal mechanisms are hypothesised to have played major roles in shaping the present- day palaeo-tropical distribution patterns of Elaeocarpus.

The polystachyus group (i.e. the Elaeocarpus polystachyus complex) comprises six species (E. cupreus, E. clementis, E. integripetalus, E. multinervosus, E. polyanthus and E. polystachyus) and four varieties (E. clementis varieties clementis, borneensis, clemensiae and kostermansii). All are endemic to West Malesia and share a unique combination of morphological character states, including numerous, unawned stamens that are densely arranged in multiple tiers. Most members of this informal infrageneric group form a clade in the phylogenetic analysis, except E. polystachyus; while no DNA samples were available for E. clementis var. kostermansii, E. integripetalus and E. polyanthus. The group is morphologically well defined and phylogenetically broadly supported, but taxon boundaries within it are unclear. The results of the morphometric analysis and the alpha taxonomy are broadly congruent in supporting six species, but the infraspecific taxa appear to be unsupported. Additionally, the hypothesis proposed in this study where morphological similarities within the polystachyus group are not correlated with ecological adaptations appeared to be supported by the morphometric evidence. This suggests that the morphological differences observed are predominantly genetically, rather than environmentally, controlled.

Item ID:	43783
Item Type:	Thesis (PhD)
Keywords:	biogeography; classification; Elaeocarpaceae; Elaeocarpus; evolution; morphology; phylogenetics; phylogeny; plant systematics; taxonomy
Date Deposited:	18 May 2016 01:53
FoR Codes:	06 BIOLOGICAL SCIENCES > 0603 Evolutionary Biology > 060310 Plant Systematics and Taxonomy @ 100%
SEO Codes:	96 ENVIRONMENT > 9608 Flora, Fauna and Biodiversity > 960803 Documentation of Undescribed Flora and Fauna @ 100%
Downloads:	Total: 1054 Last 12 Months: 36
	More Statistics