The origin and evolution of Indomalesian, Australasian and Pacific island biotas: insights from Aglaieae (Meliaceae, Sapindales)

Caroline M pannell

The peopling of Remote Oceanic islands by Austronesian speakers is a fascinating and yet contentious part of human prehistory. Linguistic, archaeological, and genetic studies have shown the complex nature of the process in which different components that helped to shape Lapita culture in Near Oceania each have their own unique history. Important evidence points to Taiwan as an Austronesian ancestral homeland with a more distant origin in South China, whereas alternative models favor South China to North Vietnam or a Southeast Asian origin. We test these propositions by studying phylogeography of paper mulberry, a common East Asian tree species introduced and clonally propagated since prehistoric times across the Pacific for making barkcloth, a practical and symbolic component of Austronesian cultures. Using the hypervariable chloroplast ndhF-rpl32 sequences of 604 samples collected from East Asia, Southeast Asia, and Oceanic islands (including 19 historical herbarium specimens from Near and Remote Oceania), 48 haplotypes are detected and haplotype cp-17 is predominant in both Near and Remote Oceania. Because cp-17 has an unambiguous Taiwanese origin and cp-17–carrying Oceanic paper mulberries are clonally propagated, our data concur with expectations of Taiwan as the Austronesian homeland, providing circumstantial support for the “out of Taiwan” hypothesis. Our data also provide insights into the dispersal of paper mulberry from South China “into North Taiwan,” the “out of South China–Indochina” expansion to New Guinea, and the geographic origins of post-European introductions of paper mulberry into Oceania.

Aim The biogeography of the tropical plant family Monimiaceae has long been thought to reflect the break-up of West and East Gondwana, followed by limited transoceanic dispersal.Location Southern Hemisphere, with fossils in East and West Gondwana.Methods We use phylogenetic analysis of DNA sequences from 67 of the c. 200 species, representing 26 of the 28 genera of Monimiaceae, and a Bayesian relaxed clock model with fossil prior constraints to estimate species relationships and divergence times. Likelihood optimization is used to infer switches between biogeographical regions on the highest likelihood tree.Results Peumus from Chile, Monimia from the Mascarenes and Palmeria from eastern Australia/New Guinea form a clade that is sister to all other Monimiaceae. The next-deepest split is between the Sri Lankan Hortonia and the remaining genera. The African Monimiaceae, Xymalos monospora, then forms the sister clade to a polytomy of five clades: (I) Mollinedia and allies from South America; (II) Tambourissa and allies from Madagascar and the Mascarenes; (III) Hedycarya, Kibariopsis and Leviera from New Zealand, New Caledonia and Australia; (IV) Wilkiea, Kibara, Kairoa; and (V) Steganthera and allies, all from tropical Australasia.Main conclusions Tree topology, fossils, inferred divergence times and ances-tral area reconstruction fit with the break-up of East Gondwana having left a still discernible signature consisting of sister clades in Chile and Australia. There is no support for previous hypotheses that the break-up of West Gondwana (Africa/South America) explains disjunctions in the Monimiaceae. The South American Mollinedia clade is only 28–16 Myr old, and appears to have arrived via trans-Pacific dispersal from Australasia. The clade apparently spread in southern South America prior to the Andean orogeny, fitting with its first-diverging lineage (Hennecartia) having a southern-temperate range. The crown ages of the other major clades (II–V) range from 20 to 29 Ma, implying over-water dispersal between Australia, New Caledonia, New Zealand, and across the Indian Ocean to Madagascar and the Mascarenes. The endemic genus Monimia on the Mascarenes provides an interesting example of an island lineage being much older than the islands on which it presently occurs.

Journal of Biogeography (J. Biogeogr.) (2008) ORIGINAL ARTICLE The origin and evolution of Indomalesian, Australasian and Pacific island biotas: insights from Aglaieae (Meliaceae, Sapindales) Alexandra N. Muellner1* , Caroline M. Pannell2, Annette Coleman3 and Mark W. Chase1 1 Molecular Systematics Section, Jodrell Laboratory, Royal Botanic Gardens Kew, Richmond, Surrey, UK, 2Department of Plant Sciences, University of Oxford, South Parks Road, Oxford, UK, 3Division of Biology and Medicine, Brown University, Providence, RI, USA ABSTRACT Aim The role of long-distance dispersal in the Indomalesian, Australasian and Pacific flora is currently hotly debated. The lack of well-resolved phylogenetic trees for Pacific plants has been a major limitation for biogeographical analysis. Here, we present a well-resolved phylogenetic tree for the tribe Aglaieae in the mahogany family, Meliaceae, and use it to investigate the origin, evolution and dispersal history of biotas in this area. The subfamily Melioideae, including the tribe Aglaieae (Meliaceae, Sapindales), is a plant group with good representation in the region in terms of biomass and species numbers, wide ecological attributes and known animal vectors. The family has a good fossil record (especially from North America and Europe). Genera and species in the tribe Aglaieae therefore provide an excellent model group for addressing this debate. Location Indomalesia, Australasia, Pacific islands. Methods Results from nuclear internal transcribed spacer ribosomal DNA analyses of 82 taxa, based on sequence alignment guided by secondary structure models, were combined with evidence from fossils and distribution data. We used strict and relaxed molecular clock approaches to estimate divergence times within Aglaieae. Putative ancestral areas were investigated through area-based and eventbased biogeographical approaches. Information on dispersal routes and their direction was inferred from the investigation of dispersal asymmetries between areas. Results Our study indicates that the crown group of Aglaieae dates back at least to the Late Eocene, with major divergence events occurring during the Oligocene and Miocene. It also suggests that dispersal routes existed during Miocene–Pliocene times from the area including Peninsular Malaysia, Sumatra and Borneo to Wallacea, India and Indochina, and from the area including New Guinea, New Ireland and New Britain further east to the Pacific islands at the peripheries of the distribution range. The origin of the Fijian species dates back to the Pliocene. *Correspondence: Alexandra Nora Muellner, Grunelius-Moellgaard Laboratory, Department of Botany and Molecular Evolution, Senckenberg Research Institute, Senckenberganlage 25, D-60325 Frankfurt, Germany. E-mail: alexandra.muellner@senckenberg.de Present address: Grunelius-Moellgaard Laboratory, Department of Botany and Molecular Evolution, Senckenberg Research Institute, Senckenberganlage 25, D-60325 Frankfurt, Germany. Main conclusions Dispersal over oceanic water barriers has occurred during geological time and seems to have been a major driving force for divergence events in Aglaieae, with some old Gondwanan land masses (e.g. Australia) colonized only during recent times. Movement from the ancestral area was predominantly towards the east. Extant Fijian species of Aglaia are monophyletic and share morphological features rarely found in species of other areas, suggesting speciation within an endemic clade. Divergence of living taxa from their closest living relatives took place during both the Miocene and the Pliocene, and peaked in the Pliocene. The present-day distribution of many species in the tribe must therefore have arisen as a result ª 2008 The Authors Journal compilation ª 2008 Blackwell Publishing Ltd www.blackwellpublishing.com/jbi doi:10.1111/j.1365-2699.2008.01935.x 1 A. N. Muellner et al. of dispersal rather than vicariance events. Furthermore, colonization from Indomalesia to Australasia and the Pacific has frequently been followed by speciation. Keywords Aglaia, Bayesian relaxed clock, biogeography, dispersal, fossil constraints, internal transcribed spacer, Meliaceae, molecular clock, Sapindales, vicariance. INTRODUCTION The Pacific Ocean covers one-third of the globe, incorporates major sections of the Earth from the poles to the equator, and overlies parts of five major lithospheric plates (Kroenke, 1996). Three-quarters of the world’s islands occur within its boundaries (Keast, 1996a). Included in these are archaic continental remnants such as New Caledonia and New Zealand, continental islands (New Guinea, Borneo), isolated archipelagos (Hawaii, Galapagos) that are unique centres of biological radiation, scores of remote volcanic islands (Societies, Marquesas) and low-lying, biologically depauperate atolls (Tuomotus). These island assemblages differ in age, area, elevation and degree of remoteness, and their biotas vary greatly in taxonomic composition, origin and richness (Keast & Miller, 1996). This multitude of islands provides an excellent test-bed for novel approaches to biogeographical analysis. The role of long-distance dispersal in the evolution of the tropical Asian flora is currently hotly contended (e.g. Yuan et al., 2005; Karanth, 2006; Won & Renner, 2006; Heaney, 2007), but studies employing an appropriate and comprehensive framework to test opposing hypotheses for the Indomalesian, Australasian and Pacific flora are still lacking. The lack of well-resolved phylogenetic trees for these plants, and a shortage of fossils suitable to set constraints for estimation of divergence time, has been a major limitation for biogeographical analysis. Brown et al. (2006) performed a study of the historical biogeography of Rhododendron section Vireya in the Malesian region based on plastid and nuclear ribosomal DNA (nrDNA) data. Since fossils were not, however, available for Vireya, the authors considered that the dating of divergence events was not possible and they presented two opposing hypotheses about the origin and subsequent extension of the group (an old Gondwanan group and vicariant pattern vs. a young group and dispersal pattern). The process of long-distance dispersal of biotas has had strong advocates ever since the earliest days of exploration of island floras. Corroboration of long-distance dispersal was essential to the promulgation of Darwin’s (1859) theory of evolution because it explained the existence of the same species on two widely separated islands or on an oceanic island and a mainland area. The alternative explanation that he considered, but rejected as unlikely, was independent evolution of the same species on two different occasions. Darwin (1859) himself cited 2 examples of potential external seed dispersal on migrating birds: ‘Prof. Newton has sent me the leg of a red-legged partridge (Cacabis rufa) which has been wounded and could not fly, with a ball of hard earth adhering to it, and weighting six and a half ounces. The earth had been kept for three years, but when broken, watered and placed under a bell glass, no less than 82 plants sprung from it.’ In contrast to external transport as described above, which, depending on island group, accounts for c. 8–15% of colonization events, it is estimated that internal transport by birds accounts for c. 40% of colonization events on islands as diverse as the Societies, Marquesas, Rapa, Juan Fernández and Samoa and Tonga (Carlquist, 1996). The distance that seeds are carried internally by birds is dependent upon the length of time that they are retained in the gut before they are regurgitated or excreted. According to Carlquist (1996): ‘since flying time is a function of distance, internal transport of seeds and fruits in birds is moderately strongly distance dependent’. In the case of plants that are poor dispersers (e.g. Sapindaceae), dispersal may be associated with speciation with such regularity that it becomes indistinguishable from a distribution associated with vicariance. On the contrary, in the case of good dispersers (e.g. Polypodiaceae), dispersal may quickly overwrite underlying patterns of vicariance (van Balgooy et al., 1996). For the purposes of this paper, ‘long-distance dispersal’ is defined as movement across a marine gap. Since the vagility of species varies greatly, it is not possible to set precise distances to distinguish between ‘long’ and ‘short’ in this context. Understanding the impact for angiosperms of vicariance and dispersal across Indomalesia, Australasia and the Pacific requires choosing a plant group with good representation in the region in terms of species numbers, wide ecological attributes, known animal vectors and a good fossil record. The subfamily Melioideae, including the tribe Aglaieae (Meliaceae, Sapindales), fulfils all these requirements. The tribe Aglaieae, which includes Aglaia (at least 120 species), Aphanamixis (three species), Lansium (three species), Reinwardtiodendron (seven species) and Sphaerosacme (one species), owes its current circumscription to the work of Pennington & Styles (1975). These five genera are restricted to the Asian tropics and extend into the western Pacific. All except Sphaerosacme, with the single species, Sphaerosacme decandra, restricted to the Himalayas, are represented in Malesia. Aglaieae is one of seven Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae tribes in the subfamily Melioideae, which belongs to the subtropical and tropical angiosperm family Meliaceae (order Sapindales; Pennington & Styles, 1975). All seven tribes are represented in Malesia, two (Guareeae and Trichilieae) are pantropical, and two more are restricted to the Old World (Turraeeae and Melieae). The remaining three (Vavaeeae, Aglaieae and Sandoriceae) are restricted to Indomalesia and the western Pacific (Mabberley et al., 1995). The tribe Aglaieae, which includes Aglaia, the largest genus of Meliaceae and an important component of the moist tropical forests in the Indomalesian region, is thus a good model group for studying the origins of Indomalesian and Pacific island biotas. The total range of Aglaia extends from the tropics of Southeast Asia, including Sri Lanka and India, to Australia (Queensland, Northern Territory, and Western Australia), as far east as the island of Samoa in Polynesia and north to the Mariana (Saipan, Roti and Guam) and the Caroline Islands (Palau and Ponape) in Micronesia (Pannell, 1992). In this study, we performed maximum parsimony, maximum likelihood and Bayesian analyses of DNA sequence data from nuclear ribosomal internal transcribed spacers (ITS) to obtain a phylogenetic framework for the biogeographical analysis of Aglaieae and other members of the subfamily Melioideae. One of our primary aims was to identify groups, either continental or insular, from which insular species might have arisen and to calculate when in the past the separation of ancestral and derived stocks might have occurred. The ITS region was the DNA marker of choice since previous analyses of plastid rbcL, matK and nuclear 26S rDNA did not provide sufficient phylogenetic information for Melioideae (see Muellner et al., 2003, 2006), and screening with other plastid markers (e.g. rps16 intron, atpB-rbcL spacer, trnL intron and trnL-trnF spacer) did not exhibit a sufficient number of informative sites. In addition, single-copy nuclear genes, in contrast to ITS, are, according to Cowan et al. (2006) ‘at present technically demanding to sequence and generally not retrievable from herbarium or other degraded samples because their amplification is highly subject to DNA quality’. This is especially true for tropical plants with large amounts of bioactive and/or polymerase chain reaction (PCR)-inhibitory compounds, such as those belonging to the Meliaceae (A.N. Muellner, personal observation). Based on 82 species, including a representative sampling for all genera, this study of ITS now provides an appropriate phylogenetic framework for inferring the biogeographical history of species in the tribe Aglaieae. Using information on fossil Meliaceae, the extant distribution of diversity and endemism as well as knowledge about the tectonic history of the geographical regions involved, we focused on the following topics: (1) The temporal and geographical origin of Aglaieae: are Aglaieae an ancient or recently evolved group and did they arise where they are most diverse today or elsewhere? The geographical origin of the tribe Aglaieae was investigated using both area-based and eventbased biogeographical approaches. This included analysis of areas of endemism, of present-day distribution patterns of diversity and of putative ancestral areas. A temporal framework for the origin of the group was obtained by means of molecular clock approaches based on multiple fossil calibrations. This enabled us to estimate divergence-times for nodes within the phylogenetic trees. (2) The role of vicariance and long-distance dispersal in the distribution of taxa: is the current distribution of Aglaieae the product of long-distance dispersal or vicariance? First, age estimates for groups within Aglaieae (e.g. Aglaia sect. Aglaia, A. sect. Amoora, Lansium and Reinwardtiodendron), based on our fossil-calibrated phylogenetic trees, provided evidence on whether Aglaieae and its clades are old enough for tectonic events to have influenced major cladogenic events in the group. Second, using the information on ages of clades and reconstruction of the geographical patterns of the phylogenetic tree, we tested whether long-distance dispersal had played a major role in the distribution of the Aglaieae. (3) Dispersal routes and direction: has Aglaieae colonized from east to west, as would be expected for a group with its centre of diversity in the Malesian region and relatively few species on Pacific islands? Having established that dispersal played a major role in the present distribution of Aglaieae, the dispersal routes and their direction were investigated further using an area–biogeographical analysis. This included assessment of dispersal asymmetries between areas, biogeographical patterns through time and the tectonic history of the area in which the tribe occurs. MATERIALS AND METHODS Taxon sampling Meliaceae material was collected during excursions to Thailand, Malaysia, Brunei, Sri Lanka, Australia and from the living collections of the Forestry Research Institute Malaysia (FRIM), Kebun Raya (Bogor Botanic Garden), Indonesia, and the Royal Botanic Gardens, Kew, UK, or taken from herbarium specimens. Herbarium vouchers are deposited at Brunei Forestry Centre, Brunei Darussalam (BRUN), Daubeny Herbarium, University of Oxford (FHO), Herbarium Senckenbergianum, Senckenberg Research Institute (FR), Royal Botanic Gardens, Kew (K), National Botanic Gardens (NBG), University of North Carolina (NCU) and University of Vienna (WU). ITS sequences of 33 taxa of Aglaia and 11 other taxa were available from the first author’s previous work on Meliaceae (Muellner et al., 2005). In addition to 36 species of Aglaia, our ITS matrices include representatives of all tribes of the Melioideae (Turraeeae, Melieae, Vavaeeae, Trichilieae, Aglaieae, Guareeae, and Sandoriceae), plus representatives of the subfamily Swietenioideae. Voucher information and GenBank accession numbers are listed in Appendix S1 in Supplementary Material (classification sensu Pennington & Styles, 1975). Properties of the internal transcribed spacers The internal transcribed spacers (ITS) of nrDNA, defined as the unit containing the ITS1 spacer, 5.8S rRNA gene and ITS2 spacer, are not only useful in assessing relationships at infrageneric but also at higher taxonomic levels in flowering Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 3 A. N. Muellner et al. plants (Hershkovitz & Zimmer, 1996; Soltis & Soltis, 1998). Secondary structure models of RNA transcripts, employed in the taxonomic group under investigation, allow for optimizing alignment of these variable and putatively phylogenetically informative regions even across more distantly related taxa (Hershkovitz & Zimmer, 1996; Gottschling et al., 2001; Coleman, 2003; Muellner et al., 2005). This is due to the fact that the secondary structure of ITS is more conserved than the primary sequence (Mai & Coleman, 1997; Coleman et al., 1998; Coleman, 2007). Isolation of DNA, amplification and sequencing DNA extraction and PCR amplification were carried out following Muellner et al. (2005). PCR products were cleaned using a NucleoSpin Extract II kit (Macherey-Nagel, Dueren, Germany). Sequencing reactions were run on an ABI 3730 capillary sequencer (Applied Biosystems, Inc., Warrington, Cheshire, UK) or a CEQ 8800 Genetic Analysis System (Beckman Coulter, Krefeld, Germany), following the manufacturers’ protocols. Sequence editing and alignment Editing and assembly of the complementary strands were carried out following Muellner et al. (2007). ITS sequences were explored for the presence of several structural motifs: the conserved angiosperm motif GGCRY–(4 to 7n)–GYGYCAAGGAA (Liu & Schardl, 1994); the conserved (C1–C6) and variable (V1–V6) domains determined for plant ITS2 sequences (Hershkovitz & Zimmer, 1996); and the conserved angiosperm motif 5¢-GAATTGCAGAATCC-3¢ within the 5.8S rRNA gene (Jobes & Thien, 1997). Folding predictions of secondary structures of the ITS1 and ITS2 RNA transcripts were made at the M. Zuker web server (http://frontend.bioinfo. rpi.edu/zukerm/home.html) by use of the mfold program, version 3.1 (Mathews et al., 1999; Zuker et al., 1999). Foldings were conducted at 37C. After a first rough alignment with clustal version X (Thompson et al., 1997), corrections were made manually by using secondary structure predictions of ITS1 and ITS2 RNA transcripts as a guide for alignment across genera. Secondary structure predictions were confirmed by hemi-compensatory base changes and full compensatory base changes (see Fig. 1 for examples) that preserved the predicted folding pattern. A total of 805 aligned positions were included in the matrices for phylogenetic analyses of ITS (including ITS1, 5.8S rDNA and ITS2). Gaps were coded as missing data. Aligned matrices are available from A.N.M. (alexandra.muellner@senckenberg.de); new sequences have been deposited in GenBank (http://www.ncbi.nlm.nih.gov/). Phylogenetic analysis Maximum parsimony (MP) analyses of the ITS data set were performed using paup* 4.0b10 (Swofford, 2002). Substitutions at each nucleotide position were treated as independent, 4 unordered, multi-state characters of equal weight (Fitch parsimony; Fitch, 1971). Heuristic searches were performed using 1000 random additions of taxa, tree bisection–reconnection (TBR) branch swapping and the option MulTrees (keeping multiple, shortest trees), but holding only 10 trees per replicate to reduce time spent in swapping for large numbers of trees. After 1000 replicates, we then used the shortest trees found as starting trees for a swapping-to-completion search (but with a tree limit of 10,000). The robustness of clades was estimated using bootstrapping (Felsenstein, 1985) with 1000 replicates, using simple sequence addition, TBR branch swapping and MulTrees, again holding 10 trees per replicate. We consider 75–84% bootstrap values moderate support and 85–100% strong support. Maximum likelihood (ML) analyses were performed with paup* 4.0b10 (Swofford, 2002) and with RAxML, version 2.2.1 (Stamatakis, 2006; http://icwww.epfl.ch/~stamatak/ index-Dateien/Page443.htm), and Bayesian analyses with MrBayes version 3.1.2 (Ronquist & Huelsenbeck, 2003; http://mrbayes.csit.fsu.edu/). The substitution models employed in these analyses were found using Modeltest, version 3.06 (Posada & Crandall, 1998; http://darwin.uvigo. es/software/modeltest.html), which indicated the general time reversible model as best fitting our data with a proportion of invariable sites and a gamma shape parameter alpha to model rate heterogeneity (GTR + I + G). For the Bayesian analyses, model parameters were estimated directly during runs, using four simultaneous chains and 2 million cycles, sampling one tree every 100 generations. Trees that preceded stabilization of the likelihood value were excluded, and the remaining trees were used to calculate posterior probabilities via the construction of a majority rule consensus tree in paup. For the ML searches with RAxML we employed the GTR + G model, using 25 rate categories (instead of four as used in the Bayesian analyses). Divergence time estimation A likelihood-ratio test (LRT) rejected the null hypothesis of rate constancy for ITS, and we therefore employed nonparametric rate smoothing (NPRS; Sanderson, 1997) as implemented in TreeEdit, version 1.0-a4.61 (Rambaut & Charleston, 2000; http://evolve.zoo.ox.ac.uk/software. html?id=TreeEdit) and a relaxed Bayesian clock approach as implemented in the multidivtime program of Thorne & Kishino (2002; http://statgen.ncsu.edu/thorne/). The input topologies for the time estimation were the ITS MP or ML tree obtained with paup. Parameter values in multidivtime were estimated with paml’s baseml, version 3.14 (Yang, 1997; http://abacus.gene.ucl.ac.uk/software/paml.html). The program estbranches (Thorne et al., 1998) was then used to calculate branch lengths and their variance, given the sequence data [82 ITS sequences of a length of 805 nucleotides (nt)], the model parameter values from paml, and the specified rooted topology. Branch lengths from estbranches became the priors for mcmc searches in multidivtime (Thorne Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae (a) (c) (d) (b) Figure 1 Folding predictions of secondary structures of the internal transcribed spacer (ITS)1 and ITS2 RNA transcripts. (a) Secondary structural model for Aglaia elaeagnoidea ITS1. (b) Secondary structural model for Lansium domesticum ITS1. (c) Secondary structural model for Aglaia elaeagnoidea ITS2. (d) Secondary structural model for Lansium domesticum ITS2. For ITS1, the conserved angiosperm motif (Liu & Schardl, 1994) is highlighted; for ITS2, a pair of arrows indicates the pyrimidine–pyrimidine bulge of helix II, and the bracket marks the longest conserved nucleotide sequence on the 5¢ side of helix III. On helix III, an example of a compensatory base change (CBC) and a hemi-CBC are highlighted (CBC = the more basal, hemi-CBC = the more distal pairing). Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 5 A. N. Muellner et al. & Kishino, 2002) that sought to find the most likely model of rate change (with rate change assumed to be log-normally distributed), given the topology, time constraints on nodes (below), and a Brownian motion parameter (m) that controls the magnitude of autocorrelation per million years (Myr) along the descending branches of the tree. Prior gamma distributions on parameters of the relaxed clock model were as follows (following the multidivtime manual for setting the mean and SD; http://statgen.ncsu.edu/thorne/multidivtime. html; Thorne & Kishino, 2002): the mean and SD of the prior distribution for the root age were set to 70 Myr based on fossils (below). The mean and SD of the prior distribution for the ingroup root rate were set to 0.0035 substitutions site)1 Myr)1 by dividing the median of the distances between the ingroup root and the tips by 70 Myr. The prior and SD for m were set to 0.014, following the recommendation of Thorne’s manual (http://statgen.ncsu.edu/thorne/multidivtime.html) that the time between root and tips multiplied by m be about 1. Markov chains in multidivtime were run for 1 million generations, sampling every 100th generation for a total of 10,000 trees, with a burn-in of 10,000 generations before the first sampling of the Markov chain. To check for convergence, we ran two analyses of different chain lengths. We also tested the effect of the root rate by running one analysis with a rate of 0.0035 and one with a rate of 0.0030. Likelihood ratio tests are sensitive to rate variation and may over-reject the clock (Sanderson, 1998). There is currently no method available to determine whether rates are sufficiently autocorrelated to warrant the application of relaxed-clock methods that assume autocorrelation, such as NPRS and multidivtime (Ho et al., 2005; S.S. Renner, University of Munich, personal communication, 2007). We therefore evaluated the estimates obtained from NPRS and the Bayesian relaxed clock against a strict clock model. Branch lengths were calculated in paup under GTR + I + G with the clock assumption enforced, using the same MP and ML topologies as used for the relaxed clock approaches. Constraints and calibrations Absolute time estimates in the Bayesian approach were obtained by simultaneously constraining five nodes (numbered 1–5 below; Table 1); for the NPRS clock and the strict clock we alternatively constrained three nodes (numbers 2–4 below; Appendices S2 & S3): (1) The root node of our data set (i.e. the most recent common ancestor of Melioideae and Swietenioideae) was constrained to maximally 137 Ma, based on the onset of angiosperm radiation (Hughes, 1994; Brenner, 1996). (2) The stem of Cedreleae was constrained to minimally 48.6 Ma (the upper bound of the Early Eocene), based on fruit and seed fossils ascribed to Toona from the London Clay (Reid & Chandler, 1933). These specimens share morphological features of both modern Toona and Cedrela (T. D. Pennington, Royal Botanic Gardens Kew, personal communication, 2005). (3) The clade comprising Guarea and Ruagea was constrained to 6 Table 1 Age estimates (in million years) for key events in the history of Melioideae based on the Bayesian approach (1 million cycles), derived from the maximum parsimony (MP) tree and the maximum likelihood (ML) tree. Node 1. Aglaia archboldiana/A. vitiensis/ A. basiphylla 2. Aglaia samoensis/A. sapindina 3. Aglaia australiensis/A. meridionalis 4. Lansium 5. Sect. Neoaglaia 6. Sect. Amoora 7. Aphanamixis/Sphaerosacme 8. Reinwardtiodendron 9. Sect. Amoora/Reinwardtiodendron/ Lansium 10. Sect. Aglaia 11. Sect. Neoaglaia/Sect. Amoora/ Lansium/Reinwardtiodendron 12. Guarea/Ruagea 13. Aglaia/Reinwardtiodendron/Lansium 14. Aglaieae (crown) 15. Aglaieae (stem) 16. Trichilia 17. Melia/Azadirachta 18. Cedrela/Toona (stem) 19. Guareae/Aglaieae: (i) excl. Vavaea; (ii) incl. Vavaea 20. Melioideae MP tree, mean (SD) 5 (3) 4 5 5 10 12 16 16 30 (3) (3) (3) (5) (5) (6) (6) (8) ML tree, mean (SD) 5 (3) 5 5 6 11 13 17 17 26 (3) (3) (3) (6) (5) (7) (6) (7) 28 (8) 30 (8) 27 (7) 30 (8) 32 (7) 33 (8) 36 (9) 39 (9) 45 (12) 46 (14) NA 56 (12) 31 (7) 33 (8) 36 (9) 39 (9) 40 (11) 45 (13) NA 53 (11) 76 (16) 76 (16) NA, not applicable; SD, standard deviation. minimally 23.03 Ma (the upper bound of the Late Oligocene), based on fossil pollen of Guarea from the Oligocene San Sebastian Formation in northern Puerto Rico (Graham & Jarzen, 1969). (4) The crown group of Melieae was constrained to minimally 20.43 Ma (the upper bound of the Aquitanian in the Early Miocene), based on fossil pollen of Melia (pollen similar to Melia azedarach) from the Early Miocene of Cameroon (Salard-Cheboldaeff, 1978). (5) The clade comprising Chisocheton and the remainder of Guareae and Aglaieae was constrained to minimally 5.3 Ma (the upper bound of the Late Miocene), based on Miocene fossil wood of Chisocheton (Chisochetonoxylon) from the Birbhum district in West Bengal (Ghosh & Roy, 1979). For absolute ages we relied on the geological time-scale of Gradstein et al. (2004). Ancestral area analyses We inferred the ancestral area of species groups within Aglaieae with an area-based biogeographical approach (ancestral area analysis; Bremer, 1992; Table 2) and an event-based approach (dispersal vicariance analysis, diva; http://www.ebc. uu.se/systzoo/research/diva/diva.html; Ronquist, 1996, 1997). diva is particularly appropriate when inferring biogeograph- Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae Table 2 Estimation of the ancestral areas using Bremer’s (1992) ancestral area analysis. Crown group Areas G L G/L AA Aglaieae MALSUBO (D) WALL (E) IND (A) INDOCHI (C) GUI (F) CHITAI (B) SOLVAN (G) AUS (H) PAC (J) FIJ (I) MALSUBO (D) WALL (E) INDOCHI (C) IND (A) GUI (F) CHITAI (B) AUS (H) SOLVAN (G) PAC (J) FIJ (I) MALSUBO (D) WALL (E) IND (A) INDOCHI (C) GUI (F) CHITAI (B) SOLVAN (G) AUS (H) PAC (J) FIJ (I) MALSUBO (D) WALL (E) INDOCHI (C) GUI (F) IND (A) CHITAI (B) AUS (H) SOLVAN (G) FIJ (I) PAC (J) MALSUBO (D) INDOCHI (C) WALL (E) IND (A) GUI (F) AUS (H) CHITAI (B) SOLVAN (G) FIJ (I) PAC (J) MALSUBO (D) WALL (E) INDOCHI (C) GUI (F) CHITAI (B) IND (A) 18 16 14 16 11 7 6 6 2 1 17 15 15 12 10 6 6 5 2 1 10 8 9 9 6 3 4 4 2 1 7 7 6 4 3 3 2 1 0 0 2 3 3 2 2 2 1 1 0 0 4 3 2 1 1 0 7 13 13 16 17 13 16 16 10 10 6 12 14 12 15 11 15 14 9 9 4 7 8 8 11 6 10 11 8 8 2 5 6 4 4 5 4 4 0 0 1 2 2 2 2 2 2 2 0 0 1 3 3 2 3 0 2.57 1.23 1.08 1.00 0.65 0.54 0.38 0.38 0.20 0.10 2.83 1.25 1.07 1.00 0.67 0.55 0.40 0.36 0.22 0.11 2.50 1.14 1.13 1.13 0.55 0.50 0.40 0.36 0.25 0.13 3.50 1.40 1.00 1.00 0.75 0.60 0.50 0.25 – – 2.00 1.50 1.50 1.00 1.00 1.00 0.50 0.50 – – 4.00 1.00 0.67 0.50 0.33 – 1.00 0.48 0.42 0.39 0.25 0.21 0.15 0.15 0.08 0.04 1.00 0.44 0.38 0.35 0.24 0.19 0.14 0.13 0.08 0.04 1.00 0.46 0.45 0.45 0.22 0.20 0.16 0.15 0.10 0.05 1.00 0.40 0.29 0.29 0.21 0.17 0.14 0.07 – – 1.00 0.75 0.75 0.50 0.50 0.50 0.25 0.25 – – 1.00 0.25 0.17 0.13 0.08 – Aglaia & Lansium & Reinwardtiodendron Aglaia sect. Aglaia Aglaia sect. Amoora & Aglaia sect. Neoaglaia & Lansium & Reinwardtiodendron Aglaia sect. Amoora Lansium & Reinwardtiodendron Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 7 A. N. Muellner et al. Table 2 (Continued) Crown group Areas G L G/L AA SOLVAN (G) AUS (H) FIJ (I) PAC (J) 0 0 0 0 0 0 0 0 – – – – – – – – G, number of necessary gains under forward Camin–Sokal parsimony; L, number of necessary losses under reverse Camin–Sokal parsimony; AA, G/L quotients rescaled to a maximum value of 1 by dividing with the largest G/L value. Figure 2 Geographical distribution of Aglaieae and delimitation of areas used for ancestral area analyses (see text for area definitions). The map uses an equal-area, Mollweide projection. ical patterns in complex geological settings where areas have split and fused again (Sanmartı́n et al., 2001). Based on the geological history of the relevant regions and the extant distribution of Aglaieae, we defined 10 areas (Fig. 2) and used the following synonyms and abbreviations: A, ‘IND’; B, ‘CHITAI’; C, ‘INDOCHI’; D, ‘MALSUBO’; E, ‘WALL’; F, ‘GUI’; G, ‘SOLVAN’; H, ‘AUS’; I, ‘FIJ’; and J, ‘PAC’. The areas correspond to the following regions: A, India, Sri Lanka, Bangladesh, Bhutan, Indian Ocean Islands, Burma; B, China and Taiwan; C, Indochina and Thailand excluding the Isthmus of Kra; D, Isthmus of Kra, Peninsular Malaysia, Sumatra, Borneo, Palawan, Java and Bali; E, Wallacea, i.e. Nusa Tenggara, Philippines excluding Palawan, Sulawesi, Maluku excluding Aru and Kai Islands; F, Aru and Kai Islands, New 8 Guinea, New Ireland, New Britain; G, the Solomons and Vanuatu; H, Australia; I, Fiji; J, the Pacific islands at the peripheries of the distribution range (Fig. 2). Since diva requires fully resolved trees, we wanted to estimate the effect of different topologies of the single most parsimonious trees on the results. We therefore randomly selected six most parsimonious trees as independent input trees and ran the diva analyses with each tree separately. Area biogeography and analysis of biogeographical patterns through time We investigated the frequency of dispersal (Table 3) and vicariance events with diva (Ronquist, 1996, 1997). We Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae Table 3 Frequency of dispersal events between single areas, inferred from diva reconstructions based on eight areas. Frequency in time segment (absolute time in millions of years) Dispersal event (from ﬁ to) IND (A) ﬁ INDCHI (B) IND (A) ﬁ MALSUBO (C) INDCHI (B) ﬁ IND (A) MALSUBO (C) ﬁ IND (A) MALSUBO (C) ﬁ INDCHI (B) MALSUBO (C) ﬁ WALL (D) MALSUBO (C) ﬁ GUI (E) MALSUBO (C) ﬁ AUS (F) MALSUBO (C) ﬁ FIJ (G) WALL (D) ﬁ MALSUBO (C) GUI (E) ﬁ FIJ (G) GUI (E) ﬁ PAC (H) FIJ (G) ﬁ GUI (E) < 5.33 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000 0.000 defined four time classes corresponding to Late Eocene and earlier (> 33.9 Ma), Oligocene (33.9–23.04 Ma), Miocene (23.03–5.34 Ma) and Pliocene to present (5.33–0 Ma). We then dated each node in our phylogenetic tree using the age estimates obtained from the Bayesian relaxed clock approach (based on the MP tree) and set diva to sort the inferred events into the time classes according to the age of each node. Since diva’s ‘sumareas’ command, which is used for setting up the program for providing summary statistics for the areas, allows only a maximum of eight areas to be considered for summation, we here redefined the following eight of the ten areas described above: A, ‘IND’; B, ‘INDCHI’; C, ‘MALSUBO’; D, ‘WALL’; E, ‘GUI’; F, ‘AUS’; G, ‘FIJ’; H, ‘PAC’. The areas correspond to the following regions: A, India, Sri Lanka, Bangladesh, Bhutan, Indian Ocean Islands, Burma; B, China, Indochina and Thailand excluding the Isthmus of Kra; C, Isthmus of Kra, Peninsular Malaysia, Sumatra, Borneo, Palawan, Java and Bali; D, Wallacea, i.e. Nusa Tenggara, Philippines (excluding Palawan), Sulawesi, Maluku (excluding Aru and Kai Islands), and Taiwan; E, Aru and Kai Islands, New Guinea, New Ireland, New Britain, the Solomons and Vanuatu; F, Australia; G, Fiji; H, the Pacific islands at the peripheries of the distribution range. RESULTS We checked the ITS sequences for the presence of structural motifs and secondary structure. This enabled us to align the entire ITS region unambiguously across all genera (ingroup and outgroup), without losing any potentially informative characters for subsequent phylogenetic analyses. Secondary structural models for Meliaceae ITS1 and ITS2 sequences are shown in Fig. 1, with two examples of hemi- and full compensatory base changes marked. Also designated on the figure are the hallmark regions of relatively conserved sequences, in ITS1 the Liu & Schardl (1994) motifs and in ITS2 the pyrimidine–pyrimidine bulge near the base of helix II (arrows) and the most extensive example of conserved < 23.03 0.400 0.100 0.400 1.200 1.200 2.222 0.333 0.062 0.333 0.555 0.333 0.000 0.333 < 33.90 0.000 0.000 0.000 0.400 0.200 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 > 33.90 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 All 0.400 0.100 0.400 1.600 1.400 2.222 0.333 0.062 0.333 0.555 0.333 1.000 0.333 sequence, that on the 5¢ side of helix III near the tip (bracket), as described in Coleman (2003). Phylogeny estimation The aligned ITS matrix consisted of 805 characters. For the entire ITS matrix, 517 (64%) positions were variable and 416 (52%) were potentially parsimony informative. The parsimony search produced 320 most parsimonious trees of 2835 steps with a consistency index (CI) of 0.34 and a retention index (RI) of 0.58. Figure 3 shows one of the 320 most parsimonious trees obtained from the MP analysis of 78 ingroup and four outgroup taxa. The Bayesian and ML analyses were all based on the same data set as used for the MP analyses. The Bayesian tree was based on a total of 20,000 trees and a burn-in of 1000 trees (tree not shown). Topologies of trees obtained with parsimony, Bayesian inference and ML (trees not shown) were nearly identical. One exception was the position of Vavaea, which was sister to all Guareae and Aglaieae in the MP and Bayesian trees and sister to Synoum as part of a clade uniting all Guareae and Aglaieae in the ML trees. Divergence time estimation Table 1 and Appendices S2 & S3 summarize the divergence time estimates obtained with the different clock approaches, based on the MP and the ML trees, respectively. Figure 4 shows the Bayesian chronogram derived from the MP tree. Whereas results obtained from the same clock approach but based on either the MP or the ML tree yielded similar age estimates, estimates based on the three different clock approaches (Bayesian, strict clock, NPRS) differed from each other. The results obtained from the strict molecular clock approach (Appendices S2 & S3) are in line with the LRT rejecting the null hypothesis of rate constancy for ITS. Based on the comparison of performance of the different methods (Bayesian, strict clock, NPRS with different fossil calibrations), age estimates derived from the Bayesian approach using either Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 9 A. N. Muellner et al. Figure 3 One of the 320 most parsimonious trees obtained from the maximum parsimony analysis of the internal transcribed spacer (ITS) data set of 82 Meliaceae accessions. Tribes are after Pennington & Styles (1975). Numbers above the branches are estimated branch lengths (DELTRAN optimization), and bootstrap percentages (1000 replicates). Numbers below the branches are Bayesian posterior probabilities (a total of 20,000, burn-in of 1000 trees). Arrows indicate groups not present in the strict consensus tree. 10 Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae (a) (b) Figure 4 Bayesian chronogram based on a maximum parsimony tree obtained from internal transcribed spacer (ITS) sequences of 78 ingroup (Melioideae, including Aglaieae) and four outgroup taxa (latter not shown). Positions with numbers in circles refer to nodes for which ages were estimated (see text and Table 1 for details). Numbers along the bars indicate million years ago (Ma). (a) Fruits and seeds of Toona sulcata, London Clay, UK, Early Eocene (modified after Reid & Chandler, 1933; Chandler, 1964), which were used to constrain the stem of Cedreleae to minimally 48.6 Ma (node 18, not shown). (b) Fossil pollen of Guarea from the Oligocene San Sebastian Formation in northern Puerto Rico (modified after Graham & Jarzen, 1969), which was used to constrain the clade comprising Guarea and Ruagea to minimally 23.03 Ma (node 12). K, Cretaceous; Pa, Palaeocene; E, Eocene; O, Oligocene; M, Miocene; P, Pliocene. Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 11 A. N. Muellner et al. the MP or the ML tree, and based on several fossils, can be viewed as best approximations (Table 1). This confirms the observations of previous studies investigating the performance of different molecular clock approaches (e.g. see reviews in Arbogast et al., 2002; Renner, 2005; Rutschmann, 2006). The Bayesian approach yielded an age estimate for Melioideae (node 20, Table 1) similar to previous estimates based on a smaller set of rbcL sequences (Muellner et al., 2006). Our age estimates (Table 1) suggest that the stem group of Aglaieae dates back to 39 ± 9 Ma and the crown group to 36 ± 9 Ma. The first major divergence resulted in two sister clades, one comprising representatives of Aphanamixis, extending from Sri Lanka and India to Bhutan, tropical China and Indochina, throughout Malesia, to the Solomon Islands (areas IND/ CHITAI/INDOCHI/MALSUBO/WALL/GUI/SOLVAN) and Sphaerosacme, restricted to the Himalayas (IND), of 16 ± 6 Ma, the other one uniting representatives of the genera Aglaia, Lansium and Reinwardtiodendron, of 33 ± 8 Ma, covering the whole distribution range (areas IND/CHITAI/ INDOCHI/MALSUBO/WALL/GUI/SOLVAN/AUS/FIJ/PAC). Subsequent diversification then gave rise to the crown group of Aglaia sect. Aglaia, distributed from the very west to east (areas IND/CHITAI/INDOCHI/MALSUBO/WALL/GUI/SOLVAN/ AUS/FIJ/PAC) of 28 ± 8 Ma, and a clade uniting Aglaia sect. Neoaglaia as well as Aglaia sect. Amoora, plus genera Lansium and Reinwardtiodendron, of 30 ± 8 Ma, extending from India to Wallacea and Australia (areas IND/CHITAI/INDOCHI/ MALSUBO/WALL/GUI/SOLVAN/AUS). Substitution rate estimates within our ITS data set ranged from 2.85 · 10)9 (Toona) and 3.96 · 10)9 (Melia) to 3.96 · 10)9 (Guarea) substitutions/site/year (s/s/y) in the strict clock approaches based on the MP tree; from 2.83 · 10)9 (Toona) and 3.93 · 10)9 (Melia) to 4.05 · 10)9 (Guarea) s/s/y in the strict clock approaches based on the ML tree; and were 3.0 · 10)9 and 3.5 · 10)9 s/s/y in the different Bayesian approaches based on the MP and the ML tree, respectively (root rate). Ancestral area analyses Table 2 summarizes the results of the ancestral area analysis that used the approach of Bremer (1992). A high gain/loss (G/L) ratio indicates a high probability that a geographical region was part of the ancestral area. Based on this measure, (part of) the area ‘MALSUBO’ (D), comprising the Isthmus of Kra, Peninsular Malaysia, Sumatra, Borneo, Palawan, Java and Bali (area D), probably belonged to the ancestral area of the crown group of Aglaieae, as well as of clades within the tribe – the crown groups of: (1) Aglaia, Lansium and Reinwardtiodendron; (2) Aglaia sect. Aglaia; (3) Aglaia sect. Amoora, Aglaia sect. Neoaglaia, Lansium and Reinwardtiodendron; (4) Aglaia sect. Amoora; and (5) Lansium and Reinwardtiodendron. In diva, the optimal solutions supported the results obtained by applying Bremer’s ancestral area approach (Fig. 5). With the ‘maxareas’ option of the ‘optimize’ command in diva used to impose constraints on the number 12 of unit areas allowed in ancestral distributions, the following results were obtained for all six tree topologies: for the crown group of Aglaieae optimal solutions were areas MALSUBO, IND/MALSUBO, MALSUBO/WALL (maxareas set to 2), IND/ MALSUBO/WALL (maxareas set to 3), IND/CHITAI/INDOCHI/MALSUBO/WALL/GUI/SOLVAN (maxareas set to 8); for all other crown groups (Aglaia, Lansium and Reinwardtiodendron; Aglaia sect. Aglaia; Aglaia sect. Amoora, Aglaia sect. Neoaglaia, Lansium and Reinwardtiodendron; Aglaia sect. Amoora; Lansium and Reinwardtiodendron) the optimal solution was always area MALSUBO (maxareas set to 2, 3, 8 or 10). Area biogeography and analysis of biogeographical patterns through time Table 3 summarizes the results of the analysis of the frequency of dispersal events between single areas and across and within different time segments. Strong dispersal asymmetries between two areas suggest that the flora of one area successfully invaded that of the other area. Asymmetric dispersal was observed between the area comprising Peninsular Malaysia, Sumatra and Borneo (MALSUBO) and Wallacea (WALL) and between the area comprising Peninsular Malaysia, Sumatra and Borneo (MALSUBO) and the area comprising India, Sri Lanka, Bangladesh, Bhutan, Indian Ocean Islands and Burma (IND): frequency of dispersal events for MALSUBO ﬁ WALL was 2.222 vs. 0.555 for WALL ﬁ MALSUBO; likewise 1.600 for MALSUBO ﬁ IND vs. 0.100 for IND ﬁ MALSUBO (Table 3). Multi-area distributions that are common in a particular time interval demonstrate the connection of unit areas. Common multi-area distributions (with a distribution frequency ‡ 0.600 in a time segment; table not shown) were found for MALSUBO/WALL and GUI/FIJ with a distribution frequency of 1.000 each, for IND/INDCHI/WALL/ GUI with a distribution frequency of 0.833, for IND/INDCHI/ MALSUBO/WALL/GUI with a distribution frequency of 0.767, and for IND/MALSUBO with a distribution frequency of 0.600, in the time segment corresponding to the Miocene (23.03–5.34 Ma); for IND/INDCHI/MALSUBO/WALL/GUI with a distribution frequency of 1.000 in the time segment corresponding to the Late Eocene and earlier (> 33.9 Ma). DISCUSSION Our study: (1) provides evidence that the crown group of the tribe Aglaieae dates back at least to the Late Eocene, with major divergence events occurring during the Oligocene and Miocene; (2) implies that part of an area comprising Peninsular Malaysia, Sumatra, Borneo, Palawan, Java and Bali belonged to the ancestral area of the group; (3) suggests that important dispersal routes existed from the ancestral area to Wallacea (i.e. Nusa Tenggara, Philippines, Sulawesi, Maluku), and to India and Indochina, and from the area comprising Aru and Kai Islands, New Guinea, New Ireland and New Britain to the Pacific islands at the peripheries of the distribution range further east (these dispersal routes being most important Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae Figure 5 diva (Ronquist, 1996, 1997) ancestral area reconstruction for Aglaieae on the Bayesian chronogram, with maxareas set to 2. Numbers along the bars indicate million years ago (Ma). Area abbreviations A–J are explained in the text. For delimitation of areas see Figure 2. E, Eocene; O, Oligocene; M, Miocene; P, Pliocene. Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 13 A. N. Muellner et al. during Miocene–Pliocene times); (4) suggests that dispersal has been a major driving force for divergence events in Aglaieae, with some old Gondwanan land masses (e.g. Australia) colonized only recently; and (5) provides evidence that the origin of Fijian species dates back to the Early to Late Pliocene and that these species were probably derived from an ancestor further to the west. These conclusions are based on several lines of evidence. The temporal and geographical origin of Aglaieae Our results imply that Aglaieae originated and began to diversify in the Malesian region. Both the area-based (Bremer analysis) and the event-based (diva) biogeographical approaches indicate that Malesia was part of the ancestral area of the group. The tribe Aglaieae seems to have originated in the western part of its current distribution, with subsequent expansion of its range being predominantly eastwards. The implication that Aglaieae originated and began to diversify in the Malesian region is also consistent with the high levels of diversity found in this area. Concerning the pattern of extant distribution of diversity in the largest genus Aglaia, the greatest number of species is found in regions corresponding to the area MALSUBO in our analyses: 70 species are found in Peninsular Malaysia, Sumatra and Borneo. The number of endemic species in most parts of the range of Aglaia is comparatively low (Pannell, 1992). An exceptionally high proportion of endemic species, 80%, is found in Fiji (8/10); in New Guinea 39% (13/33) of species are endemic and in India 30% (3/10). Of the 12 species found in Australia, 25% (3/12) are endemic, of the 60 species found in Borneo 10% (6/60) are endemic, and of the 35 species in the Philippine Islands nearly 9% are endemic (3/35). Regarding the question of the geographical origin of Aglaia, the location of old endemics is important. Since the sampled endemic species on Fiji are young (Fig. 4, Table 1, Appendices S2 & S3), this area is probably a recent centre of endemism. All clades with endemic species included so far in our phylogenetic analyses are about the same age (c. 5 Ma or less); further inclusion of endemic taxa might provide information about the presence of old endemics in Aglaieae. Recent centres of endemism, as often found on larger islands of the Pacific (e.g. the Solomons, New Caledonia, Fiji), tend to ‘smooth’ the pattern of west–east biotic impoverishment starting east of New Guinea. Considering the family as a whole, Meliaceae are best represented in the Malesian region, although Africa is almost as diverse in terms of genera (Mabberley et al., 1995). The Malay Peninsula alone has more species than the whole of Africa and begins to approach the species richness of the Neotropical region, where only eight genera are found. Thus, the Malesian region is over twice as rich in genera and nearly as rich in species as the Neotropics (Mabberley et al., 1995). Estimation of divergence time for nodes within our phylogenetic trees, by means of molecular clock approaches based on multiple fossil calibrations, illuminates the temporal 14 framework surrounding the origin of the group (Table 1, Appendices S2 & S3). The role of vicariance and long-distance dispersal on the distribution of taxa We postulate that the tribe Aglaieae and its clades are not old enough for major tectonic events in the Southern Hemisphere to have influenced current distribution by vicariance. This conclusion is based on several lines of evidence, including information on the potential ancestral area of Aglaieae along with estimates of age for the different clades in Aglaieae (e.g. Aglaia sect. Aglaia, A. sect. Amoora, Lansium and Reinwardtiodendron; Aphanamixis and Sphaerosacme) based on our fossil-calibrated phylogenetic trees. Our fossil calibrations would have to be at least 40–50 Myr older to make divergence events consistent with continental break-up scenarios (e.g. ‘IND/INDOCHI/MALSUBO/WALL/GUI’ Aglaia cucullata– ‘AUS’ Aglaia australiensis/Aglaia meridionalis divergence; Figs 4 & 5, Table 1). On the contrary, information on the ages of clades and reconstruction of distributional history within our phylogenetic tree strongly suggest that long-distance dispersal has played a major role in shaping the distribution and divergence of Aglaieae taxa. The following picture emerges for the geological history of the regions involved: India separated from Madagascar c. 95–84 Ma and continued its rapid northward migration reaching equatorial latitudes by the Eocene and colliding with southern Asia c. 43 Ma (Barron & Harrison, 1980; Lee & Lawver, 1995; McLoughlin, 2001). Following its separation from eastern Antarctica at about the end of the Eocene (c. 35.5 Ma; McLoughlin, 2001), Australia also began to drift rapidly toward Asia (Sanmartı́n & Ronquist, 2004). This included southern New Guinea, which is the northern part of the stable Australian craton, although it is likely that only the most southern margin of New Guinea was above sea level at that time. More active convergence between the Australian and Pacific plates since the Oligocene (30 Ma) enhanced the tectonic uplift of New Guinea, but by the Early Miocene much of southern New Guinea was submerged again (Sanmartı́n & Ronquist, 2004). Subsequent episodes of uplift started in the Miocene after the collision of the Australian and Asian plates and led to the accretion of numerous terranes to the northern margin of New Guinea (Pigram & Davies, 1987). A second phase of collision between Australia and the Asian Plate in the Late Miocene–Pliocene (10 Ma) resulted in the rise of mountain chains in Southeast Asia, northern Australia, New Guinea and New Zealand (Sanmartı́n & Ronquist, 2004). According to Hall (2001), water gaps were at their narrowest and land most extensive during the Late Miocene. This illustrates the complex geological setting for the area of distribution of Aglaieae. Considering the estimates of divergence time in our study, this makes it likely that current distribution patterns have been shaped by an interplay of dispersal (aided by movement of land and periods of low sea level that reduced water gaps) and extinction events. The Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae complex patterns of morphological variation in extant species of Aglaia point to the likelihood that there were repeated colonization events and incomplete isolation after dispersal to new territories. In general, the resistant dispersal stage of plants (the seed) may facilitate colonization, making plants less constrained by geological history than are many groups of animals (compare Sanmartı́n & Ronquist, 2004). Taking a look at two clades in our tree consisting of endemic taxa (Fig. 5) should demonstrate the importance of dispersal. The first example is the clade uniting species A. australiensis and A. meridionalis, both endemic to Australia. Based on the results from diva (Fig. 5), the species pair is derived from a Malesian ancestor. In addition, the node uniting the two species is of relatively young age: 5 ± 3 Ma, which suggests that some old Gondwanan land masses like Australia have been colonized only recently by means of long-distance dispersal. The recolonization of continents from islands may be an unusual but widespread phenomenon (Heaney, 2007). The second example is the clade uniting the species Aglaia archboldiana, Aglaia vitiensis and Aglaia basiphylla. All three are endemic to Fiji and represent a young lineage: 5 ± 3 Ma is the age of the node uniting all three species, while the node uniting A. archboldiana and A. vitiensis dates back to only 3 ± 2 Ma. As indicated by the results from diva, the extant Fijian species are probably derived from an ancestor which originated further west (Fig. 5). Dispersal routes and their direction Investigations of dispersal asymmetries between areas and biogeographical patterns through time suggest that important dispersal routes existed: (1) From the area comprising Peninsular Malaysia, Sumatra, Borneo, Palawan, Java and Bali: (a) to Wallacea (i.e. Nusa Tenggara, Philippines excluding Palawan, Sulawesi, Maluku); (b) to India; and (c) to Indochina. (2) From the area comprising the Aru and Kai Islands, New Guinea, New Ireland and New Britain during Miocene–Pliocene times to the Pacific islands at the peripheries of the distribution range further east. Examples for these routes and directions for (1) and (2) above are (Fig. 5): (1a) Reinwardtiodendron cinereum (MALSUBO) ﬁ Reinwardtiodendron humile (CHITAI/INDOCHI/MALSUBO/WALL) ﬁ Reinwardtiodendron kinabulense (MALSUBO)/Reinwardtiodendron kostermansii (WALL); (1b) and (1c) Aglaia coriacea (MALSUBO) ﬁ Aglaia odorata (CHITAI/INDOCHI) ﬁ Aglaia oligophylla (IND/INDOCHI/MALSUBO)/Aglaia simplicifolia (IND/INDOCHI/MALSUBO); (2) ancestor of the clade uniting Aglaia archboldiana/vitiensis/basiphylla/samoensis/sapindina ﬁ A. samoensis. As outlined before, based on results of our ancestral area analyses and fossil findings, the tribe Aglaieae seems to have originated in the western part of its current distribution, subsequently dispersing eastwards to inhabit the Aru and Kai islands, New Guinea, New Ireland, New Britain, the Solomons and Vanuatu, Australia, Fiji and other Pacific islands at the eastern end of their modern range. Regions in the east harbour more suitable habitats for Aglaieae now than in the past. As Aglaia is efficiently dispersed by birds and mammals over land and for a medium distance over water, it has probably colonized most suitable habitats in all directions. The size of the genus and the existence within it of seven taxonomically complex species and numerous variable species suggests that it is an adaptable group and has produced new forms to adapt to new conditions (see discussion of cryptic species in Pannell & White, 1988). What it has not done is to adapt to non-tropical environments. Its colonization of India has therefore been limited. It has also not been dispersed for long distances over land or sea to Africa or the islands of the Indian Ocean or far out into the Pacific. This would probably require smaller propagules than the large, recalcitrant seeds of Aglaia, which are adapted to the rain forest environment. West to east attenuation patterns from New Guinea into the eastern Pacific have been documented for virtually every major plant and animal group: for plants generally (Merrill, 1945; Corner, 1963, 1966; Thorne, 1963; van Balgooy et al., 1996; van Welzen et al., 2005); insects (Buxton, 1935; Zimmermann, 1942; Gressit, 1961; Mackerras, 1961; de Boer & Duffels, 1996); land snails (Solem, 1959; Cowie, 1996); corals (Ekman, 1953; Veron, 1995); echinoderms (Clark, 1954); birds (Keast, 1991, 1996a,b); reptiles (Brown, 1965; Adler & Dudley, 1995; Allison, 1996); and mammals (Flannery, 1996). The data set of Sohmer (1990) provided good quantitative examples of the west to east biotic impoverishment for flowering plants. The number of flowering plant species in New Guinea is probably 15,000–20,000 (Johns, 1993; Sekhran & Miller, 1994); those of New Caledonia 3061 (Morat, 1993); Fiji 1769 (Ash, 1992); Samoa c. 550; Tonga 340 (Whistler, 1992); the Societies 427; and the Marquesas 318 (Wagner, 1991). From species-rich New Guinea there is a lesser biotic attenuation westwards into the biologically depauperate Wallacea islands, and a precipitous drop-off to the north into the Bismarcks (Keast, 1996a). Of course, these islands are much smaller and of lower elevations and can accommodate many fewer species than New Guinea. It has to be acknowledged that our analyses only use a subset of all known taxa, and the fine-scale geographical patterns may change if more taxa are included. To understand Indomalesian, Australasian and Pacific island plant dispersal more fully, we will need to include more lineages in future investigations. Animal-vectored dispersal in Aglaia Our knowledge of dispersal in Aglaia is based mainly on field observations in Malaysia (Becker & Wong, 1985; Pannell & Koziol, 1987) and New Guinea (Mack, 1995, 1997). Basically, out of the three sections within Aglaia, two (section Amoora and section Neoaglaia) have dehiscent fruits and are dispersed by birds (seeds with arils rich in lipids), and one (section Aglaia) has indehiscent fruits and is dispersed by mammals (seeds with arils rich in sugars). Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 15 A. N. Muellner et al. Dispersal of species with dehiscent fruits In the western portion of the distribution of Aglaia, the largest seeds of section Amoora (Aglaia cucullata, A. spectabilis, A. macrocarpa, A. malaccensis, A. rubiginosa) can only be swallowed by large birds such as hornbills and fruit pigeons. This limits the possibilities for long-distance dispersal, but studies of hornbills in Africa have revealed that these birds can fly at least 5 km in 2 h over land (Holbrook et al., 2002). Pigeons are also strong fliers, and are known to fly long distances, including over water. They are the most likely agents of dispersal of Aglaia seeds over the sea, especially across the deep oceanic trench between the two continental shelves where there has been no possibility of dry land between the islands during periods of lower sea levels. On the Sahul continental shelf in New Guinea, birds of paradise almost certainly disperse Aglaia seeds. One species in section Amoora, Aglaia mackiana, has enormous seeds which are dispersed by the dwarf cassowary (Pannell, 1997a). This bird is flightless and the known range of A. mackiana is limited to a small area in Chimbu Province. Cassowaries are the only birds large enough to swallow the seeds of A. mackiana and the modified bird-dispersal syndrome exhibited by its fruits appears to have arisen as a response to the presence of these birds as dispersers (Pannell, 1997b; Hopkins et al., 1998). Fallen seeds of species of Aglaia section Amoora are also included in the diet of the cassowary in Australia (Stocker & Irvine, 1983). The smaller seeds of Aglaia lawii and Aglaia teysmanniana in section Neoaglaia are eaten and dispersed by a much wider range of birds than those in section Amoora. These include magpies, mynahs, bulbuls and broadbills as well as the larger birds. In the case of A. lawii, fruit-eating birds of all sizes are presumably attracted by the exceedingly high lipid values in the aril, and the species is one of the most widespread and variable in the genus, being found in India, Indochina, China, through Sundaland, Wallacea and onto the Sahul shelf in New Guinea, New Britain and the Solomon Islands. Dispersal of species with indehiscent fruits In western Malesia, the seeds of indehiscent fruits in section Aglaia are mammal-dispersed. Most field observations are of orang-utans, gibbons and siamang removing the outer inedible pericarp and ingesting the seeds with the sweet, translucent aril that adheres firmly to it (Pannell & Koziol, 1987). Leaf monkeys and macaques also ingest seeds from this section of Aglaia. Dispersal by orang-utans is thought to explain the high density of indehiscent-fruited Aglaia species in Ketambe Reserve, Sumatra, where Aglaia korthalsii and Aglaia speciosa are common and eaten by orang-utans (Rijksen, 1978; Pannell & Koziol, 1987). There is circumstantial evidence that other mammals, including civets, are dispersers of seeds from indehiscent Aglaia fruits. East of Wallace’s line, the only indigenous placental mammals are bats and rodents, and it is likely that marsupials play an important role in dispersal of these mammal-type fruits. We are, however, not aware of any 16 published observations of consumption of Aglaia seeds by these mainly nocturnal marsupials. Bird dispersal and, in the case of A. cucullata, dispersal by sea are the most likely agents for the spread of Aglaia across the deep oceanic trench between the Sunda and the Sahul continental shelves. At least 23 species cross Huxley’s line of 1868 from the Sunda shelf into Wallacea but do not reach the islands of the Sahul continental shelf (Hopkins et al., 1998). At least three species (Aglaia smithii, Aglaia parviflora and A. sapindina) cross Lydekker’s line of 1896 from New Guinea into Wallacea but do not reach the islands of the Sunda continental shelf. Only ten species of Aglaia are found on both the Sunda and Sahul continental shelves, and all but one of these (A. cucullata) are complex or variable species. Wallacea has been reported to form a barrier (Wallace’s line) for plant dispersal for the following reasons: (1) the eastern Malesian elements only rafted during the last 50 Myr as plate fragments towards Southeast Asia, where the western Malesian elements were already in place; (2) most stepping stones for dispersal only emerged during the last 10 Myr, especially in Wallacea; (3) Wallacea has a dry monsoon climate, whereas the Sunda and Sahul shelves have an everwet climate; and (4) no major land bridges were present in Wallacea during glacial periods (van Welzen et al., 2005). The data presented here suggest that the predominant direction of spread of the genus across Wallacea has been from west to east. A relatively small number of dispersal events resulting in successful establishment and low probability of back-crossing with new immigrant populations would have provided the isolation that led to the emergence of new species to the east of Wallacea. Occasionally, lineages appear to have moved in the opposite direction from the general trend, westwards from New Guinea to Wallacea. General implications for the biogeography of oceanic islands In a recent paper, Heaney (2007) presented six hypotheses that summarize aspects of an emerging paradigm for the biogeography of oceanic islands. In the following, we will briefly discuss three of these hypotheses that are supported by our study. Dispersal over oceanic water barriers is common over geological time Many recent studies have presented clear empirical evidence of long-distance dispersal from continents to other continents, from continents to oceanic islands and among oceanic islands (Heaney, 2007). According to Heaney (2007), this does not suggest that geological vicariance is not a relevant process, but rather that both dispersal and vicariance processes contribute to the patterns that we see on islands in oceanic regions. Information on the potential ancestral area of the tribe Aglaieae, along with age estimates for the different clades within it, provided evidence that the group is not old enough for major tectonic events to have influenced distribution Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae patterns by vicariance. On the contrary, our data clearly indicate that long-distance dispersal has played a major role in the distribution and divergence of taxa within the Aglaieae. Kew, to A.N.M. and C.M.P. (OFC project no. 156), and the Senckenberg Research Institute. REFERENCES Diversification within a lineage on islands is frequent, often leading to high species richness and endemism The Fijian islands have the highest proportion (80%) of endemic species in Aglaia anywhere in its range. All eight endemic species share characters unusual in Aglaia. These are a combination of the lateral veins spreading widely at an angle of at least 60 to the midrib in the middle of the leaf and a rounded leaf apex (instead of the acumen that forms a drip-tip in most species in the genus). This may indicate either that they all arose from a common ancestor or that conditions in the islands led to convergence in these characters in species arising from different ancestors. The three species included in this study belong to one clade, suggesting that, on the present evidence, the former explanation is the more likely. To reach a detailed biogeographical appreciation of the Aglaieae in addition to the general patterns elucidated by the present evidence, sampling of additional taxa will be necessary. Species on oceanic islands often persist for very long periods The so-called ‘late Pleistocene model of speciation’ dominated studies of birds and mammals in particular for many decades, but recently this has been largely refuted in favour of a paradigm in which the divergence of living taxa from their closest living relatives occurred in the Miocene and Pliocene, with a probable peak in the Pliocene (Heaney, 2007). This is largely reflected in our study, where almost all of these divergence events date back to c. 5 Ma or earlier. This calls into question the assumption that natural and undisturbed island communities may be intrinsically more vulnerable to extinction and high turnover than communities on continents (Heaney, 2007). ACKNOWLEDGEMENTS We thank Edith Kapinos (Kew) and Heike Henselewski (Frankfurt) for laboratory assistance; Mr Haji Saidin Salleh, Acting Director of the Department of Forestry, Ministry of Industry and Primary Resources, Brunei Darussalam, and Hjh Noralinda Hj Ibrahim, of the same department, for collection and export permissions; Joffree Hj Ali Ahmad and the staff of the Department of Forestry Herbarium (BRUN) and Dr Kamariah Hj Abu Salim of Universiti Brunei Darussalam, for help with organization and in the field; J. Thorne (North Carolina) for help with estbranches; Susanne Renner (Munich) and George Weiblen (Minnesota) for their comments on earlier versions of the manuscript; and three anonymous referees for their valuable comments. Financial support for this study was provided by an EU Marie Curie Fellowship to A.N.M. (project no. MEIF-CT-2003-502194), an Overseas Fieldwork Committee grant from the Royal Botanic Gardens, Adler, G.H. & Dudley, R. (1995) Dispersal and speciation of skinks among archipelagos in the tropical Pacific Ocean. Evolutionary Ecology, 9, 529–541. Allison, A. (1996) Zoogeography of amphibians and reptiles of New Guinea and the Pacific region. The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: patterns and processes (ed. by A. Keast and S.E. Miller), pp. 407–436. SPB Academic Publishing, Amsterdam. Arbogast, B.S., Edwards, S.V., Wakeley, J., Beerli, P. & Slowinski, B. (2002) Estimating divergence times from molecular data on phylogenetic and population genetic timescales. Annual Review of Ecology and Systematics, 33, 707–740. Ash, J. (1992) Vegetation ecology of Fiji: past, present, and future perspectives. Pacific Science, 46, 111–127. van Balgooy, M.M.J., Hovenkamp, P.H. & van Welzen, P.C. (1996) Phytogeography of the Pacific – floristic and historical distribution patterns in plants. The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: patterns and processes (ed. by A. Keast and S.E. Miller), pp. 191–213. SPB Academic Publishing, Amsterdam. Barron, E.J. & Harrison, C.G.A. (1980) An analysis of past plate motions; the South Atlantic and Indian oceans. Mechanisms of continental drift and plate tectonics (ed. by P.A. Davies and S.K. Runcorn), pp. 89–109. Academic Press, London. Becker, P. & Wong, M. (1985) Seed dispersal, seed predation, and juvenile mortality of Aglaia sp. (Meliaceae) in lowland dipterocarp rainforest. Biotropica, 17, 230–237. de Boer, A.J. & Duffels, J.P. (1996) Biogeography of IndoPacific cicadas east of Wallace’s line. The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: patterns and processes (ed. by A. Keast and S.E. Miller), pp. 297–330. SPB Academic Publishing, Amsterdam. Bremer, K. (1992) Ancestral areas: a cladistic reinterpretation of the center of origin concept. Systematic Biology, 41, 436– 445. Brenner, G.J. (1996) Evidence for the earliest stage of angiosperm pollen evolution: a paleoequatorial section from Israel. Flowering plant origin, evolution and phylogeny (ed. by D.W. Taylor and L.J. Hickey), pp. 91–115. Chapman and Hall, New York. Brown, W.C. (1965) The distribution of terrestrial reptiles in the islands of the Pacific basin. Proceedings of the 8th Pacific Science Congress, 3, 1479–1491. Brown, G.K., Nelson, G. & Ladiges, P.Y. (2006) Historical biogeography of Rhododendron section Vireya and the Malesian Archipelago. Journal of Biogeography, 33, 1929–1944. Buxton, P.A. (1935) Summary. Insects of Samoa. British Museum, 9, 33–159. Carlquist, S. (1996) Plant dispersal and the origin of Pacific island floras. The origin and evolution of Pacific island biotas, Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 17 A. N. Muellner et al. New Guinea to eastern Polynesia: patterns and processes (ed. by A. Keast and S.E. Miller), pp. 153–164. SPB Academic Publishing, Amsterdam. Chandler, M.E.J. (1964) The lower Tertiary floras of southern England. IV. A summary and survey of findings in the light of recent botanical observations. Bulletin of the British Museum (Natural History), Geology Series, 12, 1–151 (pl. 1–4). Clark, A.H. (1954) Records of Indo-Pacific echinoderms. Pacific Science, 8, 243–263. Coleman, A.W. (2003) ITS2 is a double-edged tool for eukaryote evolutionary comparisons. Trends in Genetics, 19, 370–375. Coleman, A.W. (2007) Pan-eukaryote ITS2 homologies revealed by RNA secondary structure. Nucleic Acids Research, 35, 3322–3329. Coleman, A.W., Preparata, R.M., Mehrotra, B. & Mai, J.C. (1998) Derivation of the secondary structure of the ITS-1 transcript in Volvocales and its taxonomic correlations. Protist, 149, 135–146. Corner, E.J.H. (1963) Ficus in the Pacific region. Pacific Basin biogeography (ed. by J.L. Gressit), pp. 233–245. Bishop Museum Press, Honolulu. Corner, E.J.H. (1966) The natural history of palms. University of California Press, Berkeley. Cowan, R.S., Chase, M.W., Kress, J. & Savolainen, V. (2006) 300,000 species to identify: problems, progress, and prospects in DNA barcoding of land plants. Taxon, 55, 611–616. Cowie, R.H. (1996) Pacific island land snails: relationships, origins, and determinants of diversity. The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: patterns and processes (ed. by A. Keast and S.E. Miller), pp. 347–372. SPB Academic Publishing, Amsterdam. Darwin, C. (1859) On the origin of species by means of natural selection, or, the preservation of favoured races in the struggle for life. John Murray, London. Ekman, S. (1953) Zoogeography of the sea. Sidgwick and Jackson, London. Felsenstein, J. (1985) Confidence limits on phylogenetics: an approach using the bootstrap. Evolution, 39, 783–791. Fitch, W.M. (1971) Toward defining the course of evolution: minimal change for a specific tree topology. Systematic Zoology, 20, 406–416. Flannery, T.F. (1996) Mammalian zoogeography of New Guinea and the southwestern Pacific. The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: patterns and processes (ed. by A. Keast and S.E. Miller), pp. 399–406. SPB Academic Publishing, Amsterdam. Ghosh, P.K. & Roy, S.K. (1979) Chisochetonoxylon benegalensis gen. et sp. nov., a new fossil wood of Meliaceae from the tertiary beds of Birbhum District, West Bengal, India. Current Science, 48, 737–739. Gottschling, M., Hilger, H.H., Wolf, M. & Diane, N. (2001) Secondary structure of the ITS1 transcript and its application in a reconstruction of the phylogeny of Boraginales. Plant Biology, 3, 629–636. Gradstein, F.M., Ogg, J.G. & Smith, A.G. (2004) A geologic time scale 2004. Cambridge University Press, Cambridge. 18 Graham, A. & Jarzen, D.M. (1969) Studies in Neotropical paleobotany. I. The Oligocene communities of Puerto Rico. Annals of the Missouri Botanical Garden, 56, 308–357. Gressit, J.L. (1961) Problems in the zoogeography of Pacific and Antarctic insects. Pacific Insects, 2, 1–94. Hall, R. (2001) Cenozoic reconstructions of SE Asia and the SW Pacific: changing patterns of land and sea. Faunal and floral migrations and evolution in SE Asia–Australasia (ed. by I. Metcalfe, J.M.B. Smith, M. Morwood and I. Davidson), pp. 35–36. A.A. Balkema, Lisse. Heaney, L.R. (2007) Is a new paradigm emerging for oceanic island biogeography? Journal of Biogeography, 34, 753–757. Hershkovitz, M.A. & Zimmer, E.A. (1996) Conservation patterns in angiosperm rDNA ITS2 sequences. Nucleic Acids Research, 24, 2857–2867. Ho, S.Y.W., Phillips, M.J., Drummond, A.J. & Cooper, A. (2005) Accuracy of rate estimation using relaxed-clock models with a critical focus on the early Metazoan radiation. Molecular Biology and Evolution, 23, 1355–1363. Holbrook, K.M., Smith, T.B. & Hardesty, B.D. (2002) Implications of long-distance movements of frugivorous rain forest hornbills. Ecography, 25, 745–749. Hopkins, H.C.F., Huxley, C.R., Pannell, C.M., Prance, G.T. & White, F. (1998) The biological monograph: the importance of field studies and functional syndromes for taxonomy and evolution of tropical plants, Kew Bulletin Additional Series. Continental Printing, Belgium. Hughes, N.F. (1994) The enigma of angiosperm origins, Cambridge Palaeobiology Series 1. Cambridge University Press, Cambridge. Jobes, D.V. & Thien, L.B. (1997) A conserved motif in the 5.8S ribosomal RNA (rRNA) gene is a useful diagnostic marker for plant internal transcribed spacer (ITS) sequences. Plant Molecular Biology Reporter, 15, 326–334. Johns, R.J. (1993) Biodiversity and conservation of the native flora of Papua New Guinea. New Guinea conservation needs assessment 2 (ed. by B.M. Beehler), pp. 15–73. Department of Environment and Conservation, Papua New Guinea. Karanth, K.P. (2006) Out-of-India Gondwanan origin of some tropical Asian biota. Current Science, 90, 789–792. Keast, A. (1991) Ornithogeography of the Pacific region. Acta XX Congressus Internationalis Ornithologici, Christchurch, pp. 420–446. New Zealand Ornithological Congress Trust Board, Wellington. Keast, A. (1996a) Pacific biogeography: patterns and processes. The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: patterns and processes (ed. by A. Keast and S.E. Miller), pp. 477–512. SPB Academic Publishing, Amsterdam. Keast, A. (1996b) Modern Pacific biogeography: an introduction. The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: patterns and processes (ed. by A. Keast and S.E. Miller), pp. 1–5. SPB Academic Publishing, Amsterdam. Keast, A. & Miller, S.E. (1996) The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae patterns and processes. SPB Academic Publishing, Amsterdam. Kroenke, L.W. (1996) Plate tectonic development of the western and southwestern Pacific: Mesozoic to the present. The origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: patterns and processes (ed. by A. Keast and S.E. Miller), pp. 19–34. SPB Academic Publishing, Amsterdam. Lee, T.-Y. & Lawver, L.A. (1995) Cenozoic plate reconstruction of the southeast Asia region. Tectonophysics, 251, 85–138. Liu, J.S. & Schardl, C.L. (1994) A conserved sequence in internal transcribed spacer 1 of plant nuclear rRNA genes. Plant Molecular Biology, 26, 775–778. Mabberley, D.J., Pannell, C.M. & Sing, A.M. (1995) Meliaceae. Flora Malesiana, Series I, 12, 1–407. Mack, A.L. (1995) Distance and non-randomness of seed dispersal by the dwarf cassowary Casuarius bennetti. Ecography, 18, 286–295. Mack, A.L. (1997) Spatial distribution, fruit production and seed removal of a rare, dioecious canopy tree species (Aglaia aff. flavida Merr. et Perr.) in Papua New Guinea. Journal of Tropical Ecology, 13, 305–316. Mackerras, I.M. (1961) The zoogeography of western Pacific Tabanidae. Pacific Insects Monograph, 2, 101–106. Mai, J.C. & Coleman, A.W. (1997) The internal transcribed spacer 2 exhibits a common secondary structure in green algae and flowering plants. Journal of Molecular Evolution, 44, 258–271. Mathews, D.H., Sabina, J., Zuker, M. & Turner, D.H. (1999) Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. Journal of Molecular Biology, 288, 911–940. McLoughlin, S. (2001) The breakup history of Gondwana and its impact on pre-Cenozoic floristic provincialism. Australian Journal of Botany, 49, 271–300. Merrill, E.D. (1945) Plant life of the Pacific world. Macmillan, New York. Morat, P. (1993) Our knowledge of the flora of New Caledonia: endemism and diversity in relation to vegetation types and substrates. Biodiversity Letters, 1, 72–81. Muellner, A.N., Samuel, R., Johnson, S.A., Cheek, M., Pennington, T.D. & Chase, M.W. (2003) Molecular phylogenetics of Meliaceae based on nuclear and plastid DNA sequences. American Journal of Botany, 90, 471–480. Muellner, A.N., Samuel, R., Chase, M.W., Pannell, C.M. & Greger, H. (2005) Aglaia (Meliaceae): an evaluation of taxonomic concepts based on DNA data and secondary metabolites. American Journal of Botany, 92, 534–543. Muellner, A.N., Savolainen, V., Samuel, R. & Chase, M.W. (2006) The mahogany family ‘out-of-Africa’: divergence time estimation, global biogeographic patterns inferred from plastid rbcL DNA sequences, extant, and fossil distribution of diversity. Molecular Phylogenetics and Evolution, 40, 236– 250. Muellner, A.N., Vassiliades, D.D. & Renner, S. S. (2007) Placing Biebersteiniaceae, a herbaceous clade of Sapindales, in a temporal and geographic context. Plant Systematics and Evolution, 266, 233–252. Pannell, C.M. (1992) A taxonomic monograph of the genus Aglaia Lour (Meliaceae). Kew Bulletin Additional Series, XVI. Her Majesty’s Stationery Office, London. Pannell, C.M. (1997a) A new, cassowary-dispersed, species of Aglaia (Meliaceae, section Amoora) from Papua New Guinea. Kew Bulletin, 52, 715–717. Pannell, C.M. (1997b) Solving problems in the taxonomy of Aglaia (Meliaceae): functional syndromes and the biological monograph. Proceedings of the Third International Flora Malesiana Symposium (ed. by J. Dransfield, M.J.E. Coode and D.A. Simpson), pp. 163–170. Kew Publishing, Royal Botanic Gardens, Kew, London. Pannell, C.M. & Koziol, M.J. (1987) Arillate seeds and vertebrate dispersal in Aglaia (Meliaceae): a study of ecological and phytochemical diversity. Philosophical Transactions of the Royal Society B: Biological Sciences, 316, 303–313. Pannell, C.M. & White, F. (1988) Patterns of speciation in Africa, Madagascar, and the tropical Far East: regional faunas and cryptic evolution in vertebrate-dispersed plants. Proceedings of the AETFAT Congress, 1985. Monographs in Systematic Botany from the Missouri Botanical Garden, 25, 639–659. Pennington, T.D. & Styles, B.T. (1975) A generic monograph of the Meliaceae. Blumea, 22, 419–540. Pigram, C.J. & Davies, P.J. (1987) Terranes and accretion history of the New Guinea orogen. BMR Journal of Australian Geology and Geophysics, 10, 193–211. Posada, D. & Crandall, K.A. (1998) Modeltest: testing the model of DNA substitution. Bioinformatics, 14, 817–818. Rambaut, A. & Charleston, M. (2000) Phylogenetic tree editor, version 1.0 alpha 4-61, http://evolve.zoo.ox.ac.uk/software/ TreeEdit. Reid, E.M. & Chandler, M.E.J. (1933) The London Clay flora. British Museum (Natural History), London. Renner, S.S. (2005) Relaxed molecular clocks for dating historical plant dispersal events. Trends in Plant Science, 10, 550–558. Rijksen, H.D. (1978) A field study on Sumatran orang-utans (Pongo pygmaeus abelii Lesson 1827). Ecology, behaviour and conservation. Thesis, Landbouwhogeschool Wageningen, Wageningen. Ronquist, F. (1996) DIVA version 1.1. Computer program and manual available from Uppsala University (http://www. ebc.uu.se/systzoo/research/diva/diva.html). Ronquist, F. (1997) Dispersal–vicariance analysis: a new approach to the quantification of historical biogeography. Systematic Biology, 46, 195–203. Ronquist, F. & Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19, 1572–1574. Rutschmann, F. (2006) Molecular dating of phylogenetic trees: a brief review of current methods that estimate divergence times. Diversity and Distributions, 12, 35–48. Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 19 A. N. Muellner et al. Salard-Cheboldaeff, M. (1978) Sur la palynoflore Maestrichtienne et Tertiaire du basin sédimentaire littoral du Cameroun. Pollen et Spores, 20, 215–260. Sanderson, M.J. (1997) A nonparametric approach to estimating divergence times in the absence of rate constancy. Molecular Biology and Evolution, 14, 1218–1232. Sanderson, M.J. (1998) Estimating rate and time in molecular phylogenies: beyond the molecular clock? Molecular systematics of plants II: DNA sequencing (ed. by D.E. Soltis, P.S. Soltis and J.J. Doyle), pp. 242–264. Kluwer, Boston. Sanmartı́n, I. & Ronquist, F. (2004) Southern Hemisphere biogeography inferred by event-based models: plant versus animal patterns. Systematic Biology, 53, 216–243. Sanmartı́n, I., Enghoff, H. & Ronquist, F. (2001) Patterns of animal dispersal, vicariance and diversification in the Holarctic. Biological Journal of the Linnean Society, 73, 345–390. Sekhran, N. & Miller, S. (1994) Papua New Guinea country study on biological diversity. Department of Environment and Conservation, Conservation Resource Centre, Waigani, Papua New Guinea. Sohmer, S.J. (1990) Elements of Pacific phytodiversity. The plant diversity of Malesia (ed. by P. Baas, K. Kalkman and R. Geesink), pp. 293–304. Kluwer Academic Publishers, Dordrecht. Solem, A. (1959) Zoogeography of the land and fresh-water mollusca of the New Hebrides. Fieldiana: Zoology, 43, 239– 359. Soltis, D.E. & Soltis, P.S. (1998) Choosing an approach and an appropriate gene for phylogenetic analysis. Molecular systematics of plants II: DNA sequencing (ed. by D.E. Soltis, P.S. Soltis and J.J. Doyle), pp. 1–42. Kluwer, Boston. Stamatakis, A. (2006) RAxML-VI-HPC: maximum likelihoodbased phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics, 22, 2688–2690. Stocker, G.C. & Irvine, A.K. (1983) Seed dispersal by cassowaries (Casuarius casuarius) in North Queensland’s rainforest. Biotropica, 15, 170–176. Swofford, D.L. (2002) PAUP*. Phylogenetic analysis using parsimony (*and other methods), version 4. Sinauer Associates, Sunderland, MA. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. & Higgins, D.G. (1997) The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research, 24, 4876– 4882. Thorne, R.F. (1963) Biotic distribution patterns in the tropical Pacific. Pacific basin biogeography (ed. by J.L. Gressitt), pp. 311–350. Bishop Museum Press, Honolulu. Thorne, J.L. & Kishino, H. (2002) Divergence time estimation and rate evolution with multilocus data sets. Systematic Biology, 51, 689–702. Thorne, J.L., Kishino, H. & Painter, I.S. (1998) Estimating the rate of evolution of the rate of molecular evolution. Molecular Biology and Evolution, 15, 1647–1657. Veron, J.E.N. (1995) Corals in space and time: the biogeography and evolution of the Scleractinia. University of New South Wales Press, Sydney. 20 Wagner, W.L. (1991) Evolution of waif floras: a comparison of the Hawaiian and Marquesan archipelagos. The unity of evolutionary biology. Proceedings of the Fourth International Congress of Systematics and Evolutionary Biology (ed. by E.C. Dudley), pp. 267–284. Dioscorides Press, Portland, OR. van Welzen, P.C., Slik, J.W.F. & Alahuhta, J. (2005) Plant distribution patterns and plate tectonics in Malesia. Biologiske Skrifter, 55, 199–217. Whistler, W.A. (1992) Vegetation of Samoa and Tonga. Pacific Science, 46, 159–178. Won, H. & Renner, S.S. (2006) Dating dispersal and radiation in the gymnosperm Gnetum (Gnetales) – clock calibration when outgroup relationships are uncertain. Systematic Biology, 55, 610–622. Yang, Z. (1997) PAML: a program package for phylogenetic analysis by maximum likelihood. Computer Applications in BioSciences, 13, 555–556 (http://abacus.gene.ucl.ac.uk/software/paml.html). Yuan, Y.M., Wohlhauser, S., Moller, M., Klackenberg, J., Callmander, M.W. & Kupfer, P. (2005) Phylogeny and biogeography of Exacum (Gentianaceae): a disjunctive distribution in the Indian Ocean Basin resulted from long distance dispersal and extensive radiation. Systematic Biology, 54, 21–34. Zimmermann, E.C. (1942) Distribution and origin of Pacific Ocean insects. The American Naturalist, 76, 280–307. Zuker, M., Mathews, D.H. & Turner, D.H. (1999) Algorithms and thermodynamics for RNA secondary structure prediction: a practical guide. RNA biochemistry and biotechnology (ed. by J. Barciszewski and B.F.C. Clark), pp. 11–43. Kluwer Academic Publishers, Boston. SUPPLEMENTARY MATERIAL The following supplementary material is available for this article: Appendix S1 Table with taxa, voucher information, origin, distribution (Aglaieae), and GenBank accession numbers for the investigated plant material. Appendix S2 Table with age estimates derived from the maximum parsimony tree under the assumption of a strict molecular clock, and based on the branch lengths observed in the non-parametric rate smoothing (NPRS) approach. Appendix S3 Table with age estimates derived from the maximum likelihood tree under the assumption of a strict molecular clock, and based on the branch lengths observed in the non-parametric rate smoothing (NPRS) approach. This material is available as part of the online article from: http://www.blackwell-synergy.com (This link will take you to the article abstract). Please note: Blackwell Publishing is not responsible for the content or functionality of any supplementary materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article. Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd Origin and evolution of Aglaieae BIOSKETCHES Alexandra N. Muellner is a research scientist and laboratory manager at the Department of Botany and Molecular Evolution, Senckenberg Research Institute, and lecturer at the Institute for Ecology, Evolution and Diversity, Goethe University. Her current research focuses on the biogeography of selected angiosperm taxa in geologically complex regions of South America and Southeast Asia. Caroline M. Pannell is a visiting researcher in the Daubeny Herbarium, Oxford, and Honorary Research Associate of the Royal Botanic Gardens, Kew. Her research is primarily in the taxonomy and ecology of the Southeast Asian rain forest mahogany trees, Aglaia. Annette Coleman is a botanist and phycologist at Brown University, Providence, RI. Her most recent work has been on the ITS2 RNA folding characteristics of eukaryotes. Mark W. Chase is Keeper of the Jodrell Laboratory at the Royal Botanic Gardens, Kew. His research interests include angiosperm classification and evolutionary biology, particularly of orchids and Nicotiana (Solanaceae). Editor: Pauline Ladiges Journal of Biogeography ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 21

RELATED PAPERS

RELATED TOPICS

Log In

The origin and evolution of Indomalesian, Australasian and Pacific island biotas: insights from Aglaieae (Meliaceae, Sapindales)

The origin and evolution of Indomalesian, Australasian and Pacific island biotas: insights from Aglaieae (Meliaceae, Sapindales)

Related Papers

RELATED PAPERS

RELATED TOPICS