Skip to main content

The cradle of flowering plants: biogeographical investigations in the most paleogeographically complex region on Earth, South-East Asia

Final Report Summary - CRADLE (The cradle of flowering plants: biogeographical investigations in the most paleogeographically complex region on Earth, South-East Asia)


Since Darwin suggested that the cradle of flowering plants was found in an isolated southern continent now vanished, the spatial aspect of their origin has been neglected. We used several data sets to address this question, focusing on families Pandanaceae and Sapindaceae, the large clade Magnoliids, and genus Psychotria (Rubiaceae). Our analyses showed that in the case of Sapindaceae, South-East (SE) Asia acted both as a cradle and as trigger for the diversification of several groups of plants, especially during the abrupt climate change at the Eocene-Oligocene boundary. For Pandanaceae, we produced the first large-scale phylogenetic analysis of the family, which resulted in the description of a new genus largely found in SE Asia. This showed that Pandanaceae dispersed at least three times to SE Asia where is diversify widely (one of these dispersal is the newly described genus). To investigate the effect of dispersal modes on diversification, we used the genus Psychotria and focusing on species occurring on the remote island of New Caledonia. We showed that the colonisation success in this group is linked to fruit morphology and niche pre-emption. The work on Magnoliids is ongoing and will allow the evaluation of biogeographical scenarios in the basal lineages of flowering plants.

Work performed and results

Objective 1: Was proto-SE Asia, formed by terranes that rifted from southern Gondwana, the birthplace of flowering plants?

Work: This work required field- and laboratory-based work as well as bioinformatic tools. Sven Buerki has been involved in three collecting trips in New Caledonia (February 2011), Fiji (February 2011) and Borneo (March 2012) with several collaborators from the Missouri Botanical Garden (USA) and the IRD, Noumea (New Caledonia).

Results and conclusions: Plant material and deoxyribonucleic acid (DNA) extracts were obtained from the field and living collections and stored in the Kew DNA bank. We have sequenced the plastid regions matK and rbcL for these samples and built an aligned matrix. Sequences available on GenBank were also incorporated in this dataset. We are currently collaborating with Dr Hervé Sauquet (France) and Dr Chrissen E. C. Gemmill (New Zealand) to improve our dataset and define an approach to date the phylogenetic tree based on fossil records.

Impact: This is ongoing work that I will pursue with the host researcher in the coming year. Two manuscripts are in preparation and will be submitted to high impact factors journals.

Objectives 2, 3: Has SE Asia acted as a centre of diversification for many flowering plant families through time? Has SE Asia been a source area of dispersal into other regions?

Work: This work required field- and laboratory-based work as well as bioinformatics tools. During each of the field mission plant material of the following families were collected and DNAs were stored in the Kew DNA bank: Sapindaceae, Pandanaceae, Psychotria (Rubiaceae) and Myrtaceae. Sven Buerki has led the research on Sapindaceae and sought taxonomic assistance for the other families to Dr Martin Callmander (Missouri Botanical Garden, Pandanaceae), Ms Laure Barrabe (IRD Noumea, Psychotria) and Dr Eve Lucas (Kew, Myrtaceae). Collaborations were also developed with Dr Nadir Alvarez (University of Lausanne) and Dr Tanja Stadler (ETH Zurich) for biogeographic and diversification analyses.

Two Master students from Imperial College (Tejal Raval and Lauren Chappell) and one sandwich student (Thomas Booth) from Manchester University were also involved in the project. We have slightly refined the objective 2 by focusing on assessing the effect of the abrupt climate change that occurred at the Eocene-Oligocene boundary on the distribution of tropical forests on Earth. The same period has also seen the emergence of SE Asia, caused by the collision of the Eurasian and Australian plates. How the combination of these climatic and geomorphological factors affect the spatiotemporal history of angiosperms is little known. We have investigated this topic by using family Sapindaceae with a worldwide distribution as a case study and applied this approach to the other families. In order to investigate this hypothesis we have implemented an approach to perform biogeographical analyses that take into account phylogenetic and dating uncertainty and also incorporate paleogeographic evidence. This approach was first applied on a dataset of the fly genus Chiastocheta in collaboration with Dr Alvarez and members of his team and published in Molecular Phylogenetics and Evolution (63: 466-474, 2012). We also used two datasets of monocot plants produced by the host researcher's group to test the approach applied for the diversification analyses used in this project. Two publications were published, in PLOS ONE (7(6): e39377, 2012) and in Annals of Botany (2013).

Results and conclusions: Based on cutting-edge methods in biogeography (constrained using paleogeographic evidence), divergence time estimation and diversification, we recorded an increase in the number of dispersal events at the Eocene-Oligocene boundary, which intensified during the Miocene. This pattern is associated with an increase in the appearance of new genera in Sapindaceae. We explain these results in light of the importance of SE Asia as a tropical bridge allowing multiple contacts between areas and additional speciation across Laurasian and Gondwanan continents. This study was submitted to Annals of Botany and was based on a dataset previously published by Sven Buerki and expanded to include more representatives from New Caledonia and SE Asia. This latter data came from a study published in Taxon (61(1): 109-119, 2012) that allowed the definition of the final sampling strategy for this project.

To further investigate the effect of SE Asia on the spread and diversification of Pandanaceae, the first familial phylogeny was published in Taxon (61(5): 1010-1020, 2012) followed by the description of a new genus resulting from this study, Benstonea Callm. & Buerki, a group mostly restricted to SE Asia (Candollea 67(2): 323-345, 2012). The phylogenetic tree was inferred from four plastid regions and > 200 species and showed that SE Asia was colonised at least three times (2 clades within Pandanus, and 1 clade within Benstonea). Thomas Booth, a sandwich student currently working under the supervision of Dr Buerki, is currently working on the phylogeny and biogeography of this genus in SE Asia with the objective of applying the methods develop by Dr Buerki for Sapindaceae to infer the spatio-temporal history of Pandanaceae. Fossil evidence suggests that the family originated in Laurasia (more likely in Eurasia) sometimes during the Cretaceous and subsequently spread into the tropics (it is currently restricted to the paleotropics) as a consequence of climate change. Several publications will result from this work and be submitted to high impact factor journals. As part of a collaboration with Ms Laure Barrabe, we have been able to see that several lineages of Psychotria colonised SE Asia and the Pacific islands very recently (Neogene) and utilised an east-west road of dispersion from Vietnam to the Pacific islands. One paper was published in Taxon (61(6): 1251-1258, 2012) and another submitted to Molecular Phylogenetics and Evolution. The dataset of Psychotria was also used to investigate objective 4.

In addition, a review on the endemic genera of angiosperms in Madagascar provide further evidence for SE Asia acting as a source area for other regions; we found in this analysis that the sister lineage of > 9 % of the genera endemic to Madagascar are present in SE Asia (which is the second most important area after Africa to have contributed most to the establishment of the Madagascan endemic generic flora). This invited review was published in Botanical Journal of the Linnean Society (171: 304-329, 2013).

Impact: A solid parametric framework was implemented to investigate the spatio-temporal history of plants that could easily be applied to other organisms. This project stressed the importance of the combined effect of geomorphological (e.g. the emergence of most islands in SE Asia ca. 30 million years ago) and climatic (e.g. the dramatic Eocene-Oligocene climate change that shifted the tropical belt and reduced sea levels) factors in shaping species distribution in angiosperms; this is especially the case in SE Asia, as the region hosts the highest diversity on earth and also the highest rate of species loss due to human activities. The fact that SE Asia acted as a tropical bridge during the abrupt climate change at the Eocene-Oligocene boundary (allowing multiple contacts between southern and northern hemispheres) provides valuable insights in the assessment of the effects of climate change and human activity on the flora of the region.

Objective 4: How different dispersal modes will affect the ability of taxa to colonise other geographic regions and the resulting biogeographic patterns?

Work, results, conclusions, impact

Work: To investigate this question we have worked in collaboration with Ms Laure Barrabe and used the Psychotria dataset mentioned above. We focused on Indo-Pacific lineages, but especially looked at the success of dispersal in New Caledonia and the subsequent diversification processes on this island. New Caledonia is a remote island and is colonisation demonstrates strong dispersal adaptations. For this project the same Bayesian dating framework was used as in objectives 2 and 3 and we applied a range of diversification methods (especially focusing at comparing diversification rates between sister clades). Species distribution models were also inferred using MaxEnt for some of the lineages. Finally, a matrix was built recording characters important for insuring dispersals (e.g. fruit type, pollination and dispersers).

Results and conclusions: New Caledonia was colonised four times by Psychotria lineages and its allied genera (belonging to the tribes Psychotrieae and Palicoureeae), long after its re-emergence from the sea during the Neogene. One clade of Psychotrieae occurring all over the Indo-Pacific region - considered to be one of the largest plant diversifications in this region - is absent from New Caledonia. Since the lack of colonisation of New Caledonia by the group seems unrelated to dispersal limitations, we hypothesised that this lineage was not able to invade this island due to niche pre-emption. Although the arrivals of the four New Caledonian lineages were simultaneous during the Neogene, they had different evolutionary histories, revealed by unevenness in species richness and net diversification rates and this was strongly linked with fruit morphology. The genus Geophila has not diversified. The genus Margaritopsis had a moderate diversification, as on other Pacific islands. The Psychotria clade NC1 is a relictual lineage, which probably underwent extinction, and would have only been preserved in shrubby vegetation, on ultramafic soils, and whose expansion and diversification were probably limited by a restricted dispersal mode. The Psychotria clade NC2 is the largest, youngest and most rapid diversification of an endemic monophyletic plant group on the archipelago and could be the result of a 'non-adaptive radiation', originating from Australian rainforests.

Impact: This study constitutes a benchmark for the understanding of plant radiations in New Caledonia (no similar studies were performed trying to investigate this aspect of plant evolution on the islands) and will provide useful guidelines to protect this unique flora from the effect of climate change (especially in regard to invasive species, which are becoming an important problem in this region). The novel approach presented in this study could also be applied over a wide range of organisms and regions and we are currently implementing a similar approach with SE Asian species of Benstonea. A manuscript has been submitted to Molecular Phylogenetics and Evolution and additional studies will also result from this work.