Revealing the secrets of African annonaceae : systematics, evolution and biogeography of the syncarpous genera Isolona and Monodora

The goal of this PhD project was to study the evolution, systematics and biogeography of two African genera from the pan-tropical Annonaceae family: Isolona and Monodora. Both genera are unique within the family in that the female reproductive parts (or carpels) are fused into a single unit. All other Annonaceae have freely arranged carpels. We investigated the phylogenetic relationships of Isolona and Monodora at the intra-familial and intra-generic levels. In Chapter 2, we explore the influence of priors when using the novel Bayesian based posterior mapping to study the evolution of morphological characters. Up to now, it was unclear if these priors had any influence on the results. Using a family level molecular phylogeny of the Annonaceae, we study the evolution of two morphological characters under different prior values. We show that different prior values will return different results. Thus, inadequate prior values can lead to erroneous conclusions over the evolution of the studied morphological characters. We also indicate a practical way to choose the prior values when using the posterior mapping approach to study morphological character evolution. In Chapter 3, using the posterior mapping approach, we study the evolutionary origins of syncarpy in Annonaceae. The closest relatives of Isolona and Monodora are elucidated. We generate a well resolved phylogeny which included for the first time the majority of African Annonaceae genera. We also study additional morphological and palynological characters relevant to Annonaceae classification in general. Our phylogenetic analyses recover a fully resolved clade comprising twelve endemic African genera, including Isolona and Monodora, which was nested within the so-called long-branch clade. This is the largest and most species-rich clade of African genera identified to date within Annonaceae. Our results indicate that syncarpy arose by fusion of a moderate number of carpels. The alternative hypothesis that syncarpy arose by multiplication of an initial single carpel receives no support. In Chapter 4 we use African Annonaceae as a model family to study the biogeographical aspects of the evolutionary origins of African rain forests. It is generally thought that the large West-Central rain forest blocks was continuous during the Eocene with the now fragmented and smaller forests of East Africa, explaining the strong floristic affinities between both areas. Using dated molecular phylogenies we provide evidence of the recurring break-up and reconnection of this pan-African rain forest during the Oligocene and Miocene. The reconnections allowed for biotic exchange while the break-ups induced speciation enhancing the levels of endemicity, thus providing an explanation for present-day patterns in the distribution and diversity of plants in African rain forests. In Chapter 5, we perform a detailed analysis of pollen morphology within a strongly supported monophyletic group of five African genera, including Isolona and Monodora. We specifically assess if pollen characters are useful for classification purposes within Isolona and Monodora using a species-level molecular phylogeny. The results show a wide pollen morphological diversity. The pollen types defined within Isolona and Monodora provide little taxonomic information for major clades within both genera. However, pollen variation proves useful as a support of phylogenetic relatedness between groups of closely related species. Finally in Chapter 6, a monographic revision of both Isolona and Monodora is presented. Isolona consists of 20 species with five endemic to Madagascar and one newly described species. Monodora has a total of 14 species, three of which were described during this PhD project from Tanzania. Detailed descriptions as well as keys are provided. The conservation status of each species is assessed following the IUCN recommendations. Just under half of the total number of species from both genera is assigned to some level of threat (12 species or 60% in Isolona and four species or 28% in Monodora).