Comparative biogeography of Southeast Asia and the West Pacific region

1 2Title 3Comparative biogeography of Southeast Asia and the West Pacific region 5Visotheary UNG 1*, 2, René ZARAGUETA-BAGILS 1, 2, and David M WILLIAMS 61 CNRS UMR 7205 (CNRS-UPMC-MNHN), 57 rue Cuvier CP43 75005 Paris, France 72 CNRS UMR 7207 (CNRS-UPMC-MNHN), 57 rue Cuvier CP43 75005 Paris, France 83 UPMC Univ Paris 06 94 Department of Life Sciences, the Natural History Museum, Cromwell Road, London, SW7 105BD, United Kingdom 11* Corresponding author: visotheary.riviere-ung@snv.jussieu.fr 12 13Short running title: Comparative biogeography of Southeast Asia 14Abstract (1630 words) 15 The relationship between areas located in Southeast Asia and the West Pacific region, 16is still debated because of its complex historical geology and the enormous diversity of taxa 17occupying the region Cladistic methods have previously been used to reconstruct the 18relationships between areas in the region but never with such a high number of unrelated taxa 19(35) We use a compilation of phylogenies to investigate area relationships among Southeast 20Asia and the West Pacific region, run the comparative analysis with LisBeth (based on the 21three-item analyses approach, i.e 3ia) and compare the results with recently published 22geological reconstructions of the region and discuss the relevance of such an approach to the 23interpretation of general pattern The main questions addressed are: how to explain actual 24distributions of taxa in Southeast Asia and the West Pacific region? Is there an emerging 25common pattern? Three-item analysis found 27 most optimal trees An intersection tree, i.e 26an area cladogram, reconstructed from the taxon-area statements (common to all 27) had an 27overall retention index 84.8% and retrieved 13 nodes with two major branches congruent with 28a separation between Southeast Asia and the West Pacific region Congruent patterns revealed 29by the combination of unrelated taxa should reflect a common cause The extraction of 30information on area relationships contained in phylogenetic analyses of taxa consists of 31testing for area homologs We obtained an area cladogram from this region based on an 32empirical dataset which give account for new insights regarding area classification in the 33region 34 35Key-words: Areas of endemism, Cladistics, Comparative biogeography, Southeast Asia, West 36Pacific, Three-item analysis (3ia), Pattern, Process 37 38Introduction 39 40 Comparative biogeography is the analysis of taxon distributions with respect to 41understanding Earth history (Parenti and Ebach 2009) It involves diverse taxon cladogram 42related together via their distributions The ichthyologist Donn Eric Rosen (1929—1986; 43Nelson et al 1987) first proposed the idea, stating “…that biological and geological patterns 44can shed light on one another (by the process of “reciprocal illumination”) but cannot test, and 45therefore cannot reject, one another” (Rosen 1978; see also Parenti 2006) 46 47 Cladistics, as applied to area relationships, makes no a priori assumptions about the 48process (or processes) responsible for patterns of distribution However, vicariance – the 49diversification of biotas following the creation of a barrier (most commonly a geological 50fragmentation) – may more parsimoniously explain the congruence of distributional patterns 51found among unrelated taxa rather than any other processes For example, De Boer has 52written that “when several monophyletic groups of species comply with one and the same 53generalized area cladogram, we can safely assume that these groups did not acquire their 54distribution patterns independently by chance dispersals, but that they responded similarly to 55the same geological events Area cladistic analysis should therefore always be based on two 56or more, preferably unrelated, groups” (de Boer 1995d) 57 58 Although historical biogeography is sometimes considered to be the same as 59vicariance biogeography, it is not designed to provide explanations about processes involved 60in taxon diversification and current distributions of taxa Nonetheless, finding congruence 61amongst a large number of non-related taxa may indeed imply a common cause Of course, 62dispersal as a cause of taxon distribution may be an explanation it remains a problem 63interpreting congruent patterns of non-related taxa which have different dispersal capabilities 64 65 The Southeast Asia and the West Pacific region (SEA-WP) have long been noted as a 66centre of biodiversity for both marine and terrestrial taxa (Michaux 2010; Woodruff 2010) 67The complex geological history of the SEA-WP region makes it a compelling area for 68biogeographical studies (Hall 1998, 2002, 2011) To this end, a large dataset was compiled 69composed of 35 non-related taxa and a general area cladogram of the region was constructed 70as one is currently lacking (Ung 2013) 71 72 The complexity of the geo-tectonic activity makes SEA-WP a challenging area for 73historical biogeographical studies, since many of the changes that occurred in the relative 74positions of the various islands and their parts suggest that a great amount of vicariant 75speciation has taken place Therefore, as in systematics, where characters are discussed 76according to their discovered status (synapomorphy versus homoplasy), the obtained 77areagram will be interpreted in terms of the combination of taxa used to reconstruct it 78 The purpose of this study is to perform a comparative biogeographical analysis of a 79compiled dataset composed of cladograms of taxa (plants and animals) The comparison with 80recently published geological reconstructions will give new clues concerning relationships 81between areas of endemism in the region Finally we will discuss the relevance of such an 82approach to the interpretation of the general pattern retrieved 83 84Material and Methods 85Data 86Areas of endemism 87 As SEA-WP biogeography is a highly debated topic, we begin by focusing on the earlier 88study of Turner et al (2001) These authors collected (for the first time) a large number of 89cladograms for both plants and animals Their dataset consists of the phylogenies and 90distribution patterns for 29 monophyletic groups These were selected according to the 91following criteria: 92 93 94 Availability of a cladistic hypothesis at species level, constructed using Wagner parsimony (Kluge & Farris 1969); 95 Availability of detailed information for the distribution patterns of all terminal species; 96 Species exclusively – or at least predominantly – occurring in the region of interest; 97 The degree of confidence in the accuracy of the cladogram 98 99 From this collection of 29 cladograms of taxa, 22 were retained according to the number 100of informative characters generated (i.e once analysed some cladograms don’t produce any 101informative TAC, thus are excluded, see explanation bellow) Thirteen recently published 102cladograms of taxa distributed over SEA-WP have been selected to increase the dataset 103following the same criteria as described above 104 105 The areas of endemism used are illustrated in Fig They have been delimited by the 106presence of a unique combination of taxa (Axelius, 1991) 107 108 Following the recommendations made in Turner et al., all taxa occurring outside the 109regions of interest were discarded That is, terminal taxa were merely deleted and the branch 110pruned from the cladogram In total, 18 areas of endemism were defined (Table 1) 111 112Taxon cladograms 113 114 Tables presents the list of taxa used by Turner et al (2001) (all details are available 115here: 116http://www.blackwellpublishing.com/products/journals/suppmat/jbi/jbi526/jbi526sm.htm) 117Table presents the additional dataset The new dataset consists of 35 cladograms with a total 118of 839 species (see Appendix S1 for all distributions) 119 120Three-item analysis 121 122 The taxon cladograms were constructed using three-item analysis (3ia), a method 123designed to represent taxon relationships directly, rather than as binary variables Thus, for an 124area cladogram AB(CD)), there are two three-item statements (3is), A(CD) and B(CD), for the 125area cladogram A(B(CD)), there are four three-item statements, A(BC), A(BD), A(CD), 126B(CD) and so on For each area cladogram, the suite of three-item statements obtained are 127fitted to an optimal tree, that which accommodates the greatest number of statements 128 A new computer program, LisBeth v.1.3 (Zaragüeta-Bagils et al 2012) 129(http://infosyslab.fr/downloadlisbeth/LisBeth.exe), has been developed in order to analysis 130three-item statements for biogeographical characters using a new method of implementation 131of three-item analysis (3ia) (Nelson & Ladiges 1991a, b, 1992; Nelson & Platnick 1991) 132LisBeth finds optimal trees by applying a compatibility analysis (Estabrook et al., 1976) to the 133suites of three-item statements (Cao 2008; Zaragüeta-Bagils et al 2012) In biogeography, 134there are two well-known problems: redundant areas due to taxic paralogy (i.e taxon 135diversification prior to area duplication) and widespread taxa (or Multiple Areas in Single 136Terminals [MASTs] according to Ebach et al 2005) are respectively resolved using Paralogy137free Subtree analysis (Nelson and Ladiges 1996) and the ‘transparent method’ (Ebach et al 1382005), the latter implements a version of Assumption (Nelson and Platnick 1981) 139 LisBeth builds a summary tree, called an intersection tree (Cao et al 2009), by 140combining the three-item statements present in all optimal trees; the intersection tree can be 141viewed as the minimal tree (for more details, see Zaragüeta-Bagils et al 2012) and is rooted 142since it is reconstructed from 3is LisBeth uses taxon-area cladograms (TAC) as ‘characters’ 143rather than the usual binary characters displayed in a tabular matrix Thus, LisBeth utilises the 144paralogy-, MAST-free TACs (Cao et al 2007) 145 It is not the intention of this paper to described all the features of LisBeth Here we 146simply draw attention to one new tool which traces the distribution of taxa that ‘support’ each 147node (see Fig 5, Table and Appendix for full algorithm); this ‘support’ is analogous to the 148concept of synapomorphy in systematics 149The large number of areas analysed (18) was accommodated by exporting three-area 150statements into a NEXUS matrix for analysis in PAUP* 4b10 (Swofford, 2003; Zaragüeta151Bagils et al., 2012), which was used to find optimal taxon-area cladograms These were then 152imported into LisBeth to reconstruct an intersection tree (Zaragüeta-Bagils et al., 2012) We 153offer one word of caution The result from the intersection tree method implemented in 154LisBeth 1.3 cannot be interpreted as a cladogram whenever polytomies are present We are 155currently working on an implementation for a new tree calculation that will fix this 156inconsistency (in version 1.4, in prep., Zaragüeta-Bagils et al in prep.) However, this 157problem is not relevant in this case as there are only two polytomies, one which is perfectly 158explained as artifact (see Results) due to sampling and the other concerns clade C, which is a 159terminal clade The full procedure to run a biogeographical analysis with LisBeth 1.3 is 160described in Appendix S2 of Hoagstrom et al (2014) 161 162Results 163 Fig shows the optimal intersection tree derived from the analysis of the 35 taxa 164cladograms selected: 1016 characters were extracted and 27 compatible trees were found 165For the intersection tree (Fig 2), the overall retention index (RI) = 0.848 and the 166Completeness Index (Compl) = 56,6% (see Zaragüeta-Bagils et al 2012 for an explanation of 167the completeness index) The 3ia analysis yields a reasonably quite resolved cladogram of the 168areas in SEA-WP: thirteen nodes (clades) are identified (Table 4) 169 170 A general pattern emerges from the 3ia analysis (Fig 3): The tree is clearly divided in 171two parts separating the Pacific islands (Fig 3, ‘Pacific’ clade and Table 4) from the rest of 172Southeast Asia (Fig 3, ‘Southeast’ Asian clade—‘Australian’ clade) The first part is named 173the ‘Pacific’ clade in reference to its location The second part is named the “Australian” 174clade That is, the clade separate from the rest of the Archipelago Apart from the unresolved 175basal position of the Lesser Sunda Islands (due to the few number of taxa distributed there), 176Weber’s (1902) line is clearly identified, which is said to trace ‘the boundary of the North 177Moluccas, separating first Timor and Australia and then, between the islands of the Babar and 178Tanimbar groups, west around Buru and Halmahera and to the Pacific’ (van Oosterzee, 1997: 17900) (Fig 4) In all, five clades are distinguishable: ‘Pacific’ (clade K), ‘Australian’ (clade I), 180‘Indonesian’ (clade H), ‘Southeast Asian’ (clade C) and ‘Wallacean’ (clade G) The latter is 181closer to the ‘Indonesian’ clade than to the ‘Southeast Asian’ Both are included in the clade D 182which in turn is sister clade to the ‘Southeast Asian’ clade The ‘Australian’ clade (clade I) is 183closer to clade B than to the ‘Pacific’ clade despite its geographical proximity 184 185DISCUSSION 186 The aim of this study is to find the general pattern for the area SEA-WP and illustrated 187it by areagram The analyses were conducted as a way of testing biogeographical hypotheses 188Van Welzen et al (2003) reassessed the Turner et al dataset in a unrooted analysis and they 189gained a result similar to our They concluded that the pattern retrieved was related more to 190proximity than to cladistics relatedness Although their result is similar to our general pattern, 191we cannot interpret them similarly With an unrooted tree, nothing can be said regarding 192relationships between the areas as there is no root Nonetheless, their result is still relevant for 193our purpose since it gives the same results 194 195 The clades found are due to specific combinations of taxa For example, clade G 196(grouping the Moluccas islands with Sulawesi, ‘Wallacean’ clade in Fig 3) is supported by 197the combined presence of species from the following plants and insects: Megarthrus, 198Rhysotoechia, Xenobates (the three insect groups), Rhododendron, Chlorocystini, and Parkia 199the three plant groups (Table 5, Fig 5) The best-supported nodes are those with the highest 200number of taxa, those which are can be termed ‘synapomorphic taxa’ Hence, clade D is the 201best supported with 15 taxa; clade I, the ‘Australian’ clade, and clade E are supported by 13 202taxa each; 12 taxa support clade F; clades B and J are supported by 11 taxa each; clade A is 203supported by taxa; clade L is supported by taxa; clade G supported by taxa and clades C 204(the ‘Southeast asian’ clade), K (the ‘Pacific’ clade) and M are supported by just 205‘synapomorphic’ taxa 206 Clade K is interesting because despite the number of taxa distributed over those areas 207(13), it gains support from only five (Cosmopsaltriina, Cyrtandra, Cupianopsis, Gehyra and 208Halobates princeps) 209 In Table 5, we have indicated in bold taxa that support only one node: Erismanthus, 210Dundubia jacoona assemblage, Fordia, whose distributions are unambiguous 211‘synapomorphies’ of clades B, D, E Calicmeniinae, Cycas and Haloveloides support clade I 212Table shows distributions of taxa and nodes supported 213 Fig illustrates their distributions to which we add the Varanus distribution because it 214supports clades J and I (J being included in I) It is noteworthy that although these taxa are 215broadly distributed over the region, each distribution supports only one area, Erismanthus 216appears as a ‘synapomorphy’ of area B, the Dundubia jaccona assemblage of area D, Fordia 217distribution supports area E, and I has the distributions of Chlorocyphidae, Cycas and 218Halobates as unambiguous ‘synapomorphies’ These results highlight the fact that, in spite of 219what seems complex biogeographical relationships, the distributions of these taxa may be 220assigned to a single area The identification of distribution of taxa as ‘synapomorphies’ of 221biogeographical areas allows focusing the discussion on the pattern of relationships 222 223Comparison with the geology 224 Exploring cladistics relationships between areas of endemism leads to a consideration 225of the geological history of a region of interest In our case, regarding SEA-WP, different 226interpretations regarding its evolution because of its complex geology were discovered (e.g 227Hamilton 1979; Holloway 1979; Duffels 1986; Hall 1998, 2002, 2011 among others) Some 228of the complexity is captured by Hall: 229 230 231 232 233 “…it is clear from the geology of the region that the snapshot we see today is no less complicated than in the past The region has developed by the interaction of major lithospheric plates, principally those of the Pacific, India-Australia and Eurasia, but at the present day a description only in terms of these three plates is a very great oversimplification.” (Hall 1998: 99) 234 235 236 “ The abrupt division between faunas and floras in Indonesia first recognized by Wallace in the nineteenth century, has its origin in the rapid plate movements and reorganisation of land-masses in SE Asia” (Hall 1998: 100) 237 238 According to Hall, therefore, much of the evidence that must be used in a regional 239tectonic model of SE Asia is based on the interpretation of geological data from the small 240ocean basins, their margins, and from the geologically more complicated land areas around 241them He goes on to note: 242 243 244 245 246 “The reader should be aware that, as in other areas of science, geologists differ in their interpretations of these data, and much of the information does not lend itself to unambiguous reconstruction Nonetheless, a complete tectonic history can only be deduced from the geology on land combined with data from the oceans.” (Hall 1998: 105) 247 248 In another paper from the same book, Holloway and Hall (1998) claim that to a 249geologist the dismissal of the role of dispersal is odd Judging from the geological history of 250SE Asia it seems highly unlikely that any understanding of the biogeographic patterns can be 251achieved without considering both vicariance and dispersal Certainly, the Cenozoic 252development of the region is characterised more by amalgamation than fragmentation 253Nevertheless, this reasoning misses that the source of relationships of areas is based on 254relationships of distribution of taxa These are tree-like independently of the geological 255history of the region Moreover, a tree of areas derived from biological evidence could 256reasonably be interpreted as: the fragmentation of a previously continuous land area such as 257Gondwanaland, or Pleistocene division of Sundaland through marine transgression (Ruedi 2581995); the approach and accretion of terranes (and thus dispersal to) onto a larger land mass; 259or the slow dispersal of organisms, with speciation, through an archipelago with stable 260geography 261 262 The Gondwana origins of all component continental blocks of SE Asia, i.e the core of 263Sundaland, is now widely accepted (Hall 2009, 2011, 2012; Hall et al 2011; Metcalfe 2011) 264Sundaland was assembled from continental blocks that separated from Gondwana in the 265Paleozoic and amalgamated with Asian blocks in the Triassic (Hall 2011; Hall et al 2011; 266Metcalfe 2011) The first event that shaped the region (about 180 Ma ago) is the break-up of 267the Gondwana which led to the separation of India and Australia-Antarctica from Africa The 268Indian subcontinent moved northward towards Asia and collided later (at about 50 Ma ago) 269(de Boer and Duffels 1996 and references herein) Australia, which had become separated 270from Antarctica by the opening of the Tasman Sea (95 Ma ago), changed its course 271northwards (de Boer and Duffels 1996) The floor of the Thethys sea was forced to subducted 272under the Pacific plate which gave rise to a volcanic island arc (the West Pacific island arc, 273also named the Outer Melanesian arc by Duffels (1986) and Holloway (1979) although not 274entirely recognized as the same (Michaux 1994; de Boer & Duffels 1996) and simply the 275Melanesian arc system by Hall (2002)) Remnants of this West Pacific island arc have been 276recognized and are as follows: the central Philippines, northern, central, and southeastern New 277Guinea, and the Bismarck Archipelago The collision between the Pacific plate and the Asian 278continent must have occurred about 40-42 Ma ago and caused the fracture of the West Pacific 279island with its northern western part rotating clockwise which made the central Philippines 280collide with the continental western Philippines (Rangin et al 1990a, b; Daly et al 1991; 281Honza 1991) 282 283 To the east, the continuous South-West Pacific island arc (also known as the East 284Melanesian arc) is composed of current Vanuatu linked to Solomon and Fiji (de Boer 1995d 285and references therein) At about 9-12 Ma ago, it collided with the Australian continent in the 286Solomon area and simultaneously in the New Guinea area, which makes it broken up giving 287rise to Vanuatu, Fiji and the Tonga-Kermode (de Boer 1995d) By Ma ago Fiji was totally 288isolated The palaeogeographic reconstruction presented here can be summarized in the 289“cladogram-like” graph for the West Pacific island arc (Fig 7) 290 291 Some geological events can be drawn from examination of the geological area 292cladogram (Fig.7) and the intersection tree (Fig 2) The main one is the vicariant event that 293separated the areas emerged from the East Melanesian Arc (sensu de Boer, 1995d) 294That scheme is consistent with our areagram Node 11 of Fig shows that after the first 295vicariant event that has separated Samoa, successive events may be reconstructed: Vanuatu 296and New Caledonia, then Fiji and Tonga Furthermore, this clade is revealed by the presence 297of only four taxa: Cosmopsaltriina, Cyrtandra, Gehyra, Cupaniopsis and Halobates princeps 298whose dispersal abilities varied a lot from one to another Hence, we could hypothesize that 299those taxa were present on the Melanesian arc and that its fracture led to this distribution 300Nonetheless, the rest of the general areagram suggest relationships emerging from 301geographical proximity (clade I, clade C, clade H and clade G) Thus, dispersal events would 302be a more probable explanation despite the different dispersal capabilities of the taxa under 303study However, one has to keep in mind that the areagram shows relationships between areas 304of endemism that are biogeographic areas and not geographic areas even if they are given 305names from geography Many islands in that region should be considered as separate 306fragments because they are geologically composite For instance Sulawesi should be 307separated in three distinct areas: north, west and south (Hall 1996; Michaux 2010; Hall 2011), 308New Guinea in at least two areas, north and south (Hall 2002) The main impediment remains 309the availability of precise distributional data since most published cladograms of taxa not 310present accurate distributions 311 312Cladistic biogeography 313 314 Biogeography and geology are often opposed, although these two disciplines are 315definitively complementary (Holloway & Hall 1998; Hall 2002, 2011; Michaux 2009) If 316there is a contradiction between them, then one discipline should not prevail Congruence 317between geological reconstructing/modelling history and biogeographical reconstruction of 318relationships between areas of endemism emphasize the fact that both can indeed be true 319 320 From 35 different taxa, including both animals and plants, we have shown that a 321pattern does emerge from the analysis and, to a certain extent, fits with the most common 322geological reconstructions of SEA-WP evolution (Hall 2002, 2011, Hall et al 2011; Metcalfe 3232011) This result emphasizes the accuracy of the 3ia method in revealing a biogeographical 324pattern that does not only consider dispersal as the main process of organism distribution 325 326 327CONCLUSIONS 328 329 Our analysis indicates that a vicariant event separated Fiji, Tonga, Vanuatu, New 330Caledonia and Samoa from the rest, which corroborates the results of de Boer and Duffels 331who examined the historical biogeography of the cicadas of Wallacea, New Guinea and the 332West pacific (de Boer & Duffels, 1996) They compared each cicada area-cladogram with the 333geological cladogram (Fig 7) and concluded that: 334 335 336 “from the high degree of congruence between the geological cladogram and the two taxon-area cladograms we conclude it is highly probable that vicariance caused by the fragmentation of a West and 337 338 a South-West Pacific island arc is responsible for most of the present-day generic diversification” (de Boer & Duffels 1996:174) 339 340 They intended at first to use the cladistic method of Platnick and Nelson (1978) and 341Humphries & Parenti (1986), i.e combining taxon-area cladograms based on several 342unrelated groups into a general area cladogram Nevertheless, they considered that the area 343cladograms for the two groups of cicadas provided an insufficient basis for the construction of 344such a general cladogram We agree: congruence among two cladograms may provide the 345weakest support Indeed, their general argument strengthens our case, as we combined 35 346different taxon cladograms from both plants and animals and obtained the general areagram 347shown in Fig where relationships between the areas of endemism can be discussed in term 348of combination of taxa that supported the nodes of the tree Our analysis has recovered five 349‘monophyletic’ clades : ‘Pacific’ (clade K), ‘Australian’ (clade I), ‘Indonesian’ (clade H), 350‘Southeast asian’ (clade C) and ‘Wallacean’ (clade G) supported by a certain assemblage of 351taxa Furthermore, Weber’s Line is revealed which confirms that drawing lines is informative 352 353Systematic biogeography is similar to phylogenetic systematics in that proposal of area 354relationships to be tested and accepted or rejected in an effort to form relevant area 355classifications (Parenti & Ebach, 2009) 3ia is a straightforward method that serves historical 356biogeographic purposes The hierarchical representation of knowledge available in LisBeth 357authorise new discussions in term of relationships between areas towards a natural 358classification 359 360 361ACKNOWLEDGEMENTS 362Visotheary UNG is grateful to Peter van Welzen for reviewing a draft of the manuscript and 363for fruitful discussions about the biogeography of the region 364The authors would like to thank several anonymous referees for constructive critiques of 365previous version of the manuscript 366This research is supported by a grant (I3iaSEAB, AAP-IN-2009-033) from the Fundation for 367Research on Biodiversity (FRB) 368 369REFERENCES 370Andersen NM 1998 Marine water striders (Heteroptera, Gerromorpha) of the Indo-Pacific: 371 cladistic biogeography and Cenozoic paleogeography In: Hall, R, Holloway, JS, eds 372 Biogeography and geological evolution of SE Asia Leiden: Backhuys Publishers, 341- 373 354 374Ast JC 2001 Mitochondrial DNA evidence and evolution in Varanoidea (Squamata) 375 Cladistics 17: 211-226 376Axelius B 1991 Areas of distribution and areas of endemism Cladistics 7: 197-199 377Birks SM, Edwards SV 2002 A phylogeny of the megapodes (Aves: Megapodiidae) based 378 on nuclear and mitochondrial DNA sequences Molecular Phylogenetics and Evolution 23: 379 408-421 380Brown GK, Nelson G, Ladiges P 2006 Historical biogeography of Rhododendron section 381 Vireya and the Malesian Archipelago Journal of Biogeography 33: 1929-1944 382Cao N 2008 Analyse trois élements et anatomie du bois des Fagales, Engl Thèse de 383 doctorat Université Pierre et Marie Curie 384Cao N, Zaragüeta-Bagils R, Vignes-Lebbe R 2007 Hierarchical representation of the 385 hypotheses of homology Geodiversitas 29: 5-15 386Cao N, El Azawi M, Zaragüeta-Bagils R 2009 Three-item analysis and parsimony, 387 intersection tree and strict consensus: a biogeographical example Bulletin de la Société 388 Géologique de France 180: 13-15 389Cronk QCB, Khien M, Wagner, WL, Smith JF 2005 Evolution of Cyrtandra 390 (Gesneriaceae) in the Pacific Ocean: the origin of a supertramp clade American Journal of 391 Botany 92:1017-1024 392Daly MC, Cooper MA, Wilson I, Smith DG, Hooper BGD 1991 Cenozoic plate tectonics 393 and basin evolution in Indonesia Marine and Petroleum Geology 8: 2-21 394de Boer AJ 1995 Islands and cicadas adrift in the West-Pacific Biogeographic patterns 395 related to plate tectonics Tijdschrift Voor Entomologie 138: 169-244 396de Boer AJ, Duffels JP 1996 Historical biogeography of the cicadas of Wallacea, New 397 Guinea and the West Pacific: a geotectonic explanation Palaeogeography, 398 Palaeoclimatology, Palaeoecology 124(1-2): 153-177 399Dekker RWRJ 2007 Distribution and speciation of Megapodes (Megapodiidae) and 400 subsequent development of their breeding behaviour In: Renema W, ed Biogeography, 401 Time, and Space: Distributions, Barriers, and Islands Springer, 45-91 402Duffels JP 1986 Biogeography of indo-pacific cicadoidea: a tentative recognition of areas of 403 endemism Cladistics 2: 318-336 404Duffels JP, Turner H 2002 Cladistic analysis and biogeography of the cicadas of the Indo405 Pacific subtribe Cosmopsaltriina (Hemiptera: Cicadoidea: Cicadidae) Systematic 406 Entomology 27: 235-261 407Ebach MC, Humphries CJ, Newman RA, Williams DM, Walsh SA 2005 Assumption 2: 408 opaque to intuition? Journal of Biogeography 32: 781–787 409Estabrook GF, Johnson CS, McMorris FR 1976 A mathematical foundation for the 410 analysis of cladistic character compatibility Mathematical Biosciences 29: 181–187 411Ewart A 1988 Geological history of the Fiji-Tonga-Samoan region of the S.W Pacific and 412 some palaeogeographic and biogeographic implications In: Duffels JP, eds The Cicadas 413 of Fiji Samoa and Tonga Islands, their Taxonomy and Biogeography (Homoptera, 414 Cicadoidea) Entomonograph 10: 15-23 415Gasmann D 2005 The phylogeny of Southeast Asian and Indo-Pacific Calicnemiinae 416 (Odonata, Platycnemididae) Bonner Zoologische Beiträge 53: 37-80 417Gemmill CEC, Allan GJ, Wagner WL, Zimmer A 2002 Evolution of insular pacific 418 Pittosporum (Pittosporaceae): origin of the Hawaiian radiation Molecular Phylogenetics 419 and Evolution 22: 1-42 420Hall R 1998 The plate tectonics of Cenozoic SE Asia and the distribution of land and sea In 421 Hall R, Holloway JD, eds Biogeography and Geological evolution of SE Asia Leiden: 422 Backhuis Publishers, , 99-131 423Hall R 2002 Cenozoic geological and plate tectonic evolution of SE Asia and the SW 424 Pacific: computer-based reconstructions, model and animations Journal of Asian Earth 425 Sciences 20: 353-434 426Hall R 2009 Southeast Asia's changing palaeogeography Blumea: Journal of Plant 427 Taxonomy and Plant Geography 54: 148-161 428Hall R 2011 Australia-SE Asia collision: plate tectonics and crustal flow In: Hall R, Cottam 429 MA, Wilson MEJ, eds The SE Asian gateway: history and tectonics of Australia-Asia 430 collision Geological Society, Special publications, London, 355: 75-109 431Hall R 2012 Sundaland and Wallacea: geology, plate tectonics and palaeogeography In: 432 Gower DJ, Richardson JE, Rosen BR, Rüber L, Williams ST, eds Biotic Evolution and 433 Environmental Change in Southeast Asia Cambridge: Cambridge University Press, 32-78 434Hall R, Cottam MA, Wilson MEJ 2011 The SE Asian gateway: history and tectonics of the 435 Australian-Asia collision In: Hall R, Cottam MA, Wilson MEJ, eds The SE Asian 436 gateway: history and tectonics of Australia-Asia collision Geological Society, Special 437 publications, London, 355, 1-6 438Hamilton W 1979 Tectonics of the Indonesian region United States Geological Survey 439 Professional Paper, 1078 440Heinicke MP, Greenbaum E, Jackman TR, Bauer AM 2011 Phylogeny of a trans441 Wallacean radiation (Squamata, Gekkonidae, Gehyra ) supports a single early colonization 442 of Australia Zoologica Scripta 40: 584-602 443Hoagstrom CW, Ung V, Taylor K (2014) Comparative biogeography and Geographic 444 Speciation of North American highland fishes from the Mesa Central to the Canadian 445 Shield, Journal of Biogeography 41: 644-658 446Holloway JD 1979 A survey of the Lepidoptera, Biogeography and Ecology of New 447 Caledonia Series Entomologica 15 The Hague: Junk W 448Holloway JD, Hall R 1998 SE Asian geology and biogeography: an introduction In Hall R, 449 Holloway JD, eds Biogeography and Geological evolution of SE Asia Leiden: Backhuys 450 Publishers, 1-23 451Honda M, Ota H, Murphy RW, Hikida T 2006 Phylogeny and biogeography of water skinks of the 452 genus Tropidophorus (Reptilia: Scincidae): a molecular approach Zoologica Scripta 35: 85-95 453Honza E 1991 The Tertiary arc chain in the western Pacific Tectonophysics 187: 285-303 454Humphries CJ, Parenti LR 1986 Cladistic biogeography Oxford monographs on 455 biogeography no Oxford: Clarendon Press 456Jønsson KA, Iresredt M, Bowie ACK, Christidis L, Fjeldså 2011 Systematics and 457 biogeography of Indo-Pacific ground-doves Molecular Phylogenetics and Evolution 59: 458 538-543 459Keppel G, Hodgskiss PD,Plunkett GM 2008 Cycas in the insular South-west Pacific: 460 dispersal or vicariance? Journal of Biogeography 35: 1004-1015 461Kluge AG, Farris JS 1969 Quantitative phyletics and the evolution of anurans Systematic 462 Zoology 18: 1-32 463Metcalfe I 2011 Paleozoic-Mesozoic history of SE Asia The SE Asian gateway: history and 464 tectonics of Australia-Asia collision In: Hall R, Cottam MA, Wilson MEJ, eds The SE 465 Asian gateway: history and tectonics of Australia-Asia collision Geological Society, 466 Special publications, London 355: 7-35 467Michaux B 1994 Land movements and animal distributions in east Wallacea (eastern 468 Indonesia, Papua New guinea and Melanesia) Palaeogeography, Paleoclimatology, 469 Paleoecology 112: 323-343 470Michaux B 2009 Reciprocality between biology and geology: Reconstructing polar 471 Gondwana Gondwana Research 16: 655-668 472Michaux B 2010 Biogeology of Wallacea: geotectonic models, areas of endemism, and 473 natural biogeographical units Biological Journal of the Linnean Society 101: 193–212 474Nelson G, Atz JW, Kallman KD., Smith CL 1987 Donn Eric Rosen, 1929-1986 Copeia: 475 541-547 476Nelson GJ, Ladiges PY 1991a Standard assumptions for biogeographic analysis, Australian 477 Systematic Botany 4: 41-58 478Nelson GJ, Ladiges PY 1991b Three-area statements: Standard assumptions for 479 biogeographic analysis, Systematic Zoology 40: 470-485 480Nelson GJ, Ladiges PY 1992 Information Content and Fractional Weight of Three-Item 481 Statements, Systematic Biology 41: 490-494 482Nelson GJ, Ladiges PY 1996 Paralogy in Cladistic Biogeography and Analysis of Paralogy483 Free Subtrees, American Museum Novitates 3167: 1-58 484Nelson G Platnick NI 1981 Systematics and Biogeography Cladistics and Vicariance New 485 York: Columbia University Press 486Nelson GJ, Platnick NI 1991 Three-taxon statements: A more precise use of parsimony? 487 Cladistics 7: 351-366 488Parenti LR 2006 Common cause and historical biogeography In: Ebach MC, Tangney R, 489 eds Biogeography in a Changing World CRC Press, 61-82 490Parenti LR, Ebach MC 2009 Comparative Biogeography: Discovering and Classifying 491 Biogeographical Patterns of a Dynamic Earth Berkeley: University of California Press 492Parenti LR, Ebach MC 2010 Wallacea Deconstructed In: Williams DM, Knapp S, eds 493 Beyond Cladistics: The Branching of a Paradigm Berkeley: University of California 494 Press, 303-318 495Parenti LR, Ebach MC 2013 Evidence and hypothesis in biogeography Journal of 496 Biogeography 40: 813-820 497Pigram CJ, Davies HJ 1987 Terranes and the accretion history of the New Guinea Orogen 498 B.M.R.J Aust Geol Geophys 10: 193-212 499Platnick NI, Nelson G 1978 A method of analysis for historical biogeography Systematic 500 Zoology 27: 1-16 501Rangin C, Jolivet L., Pubellier M 1990a A simple model for the tectonic evolution of 502 southeast Asia and Indonesia region for the past 43 Ma Bull Geol Soc France, 8, VI: 503 889-905 504Rangin G, Pubellier M, Azema J, Briais A, Chotin P, Fontaine H, Huchon P, Jolivet L, 505 Maury R, Muller C, Rampnoux JP, Stephan JF, Tournon J, Cottereau N, Dercourt J, 506 Ricou LE 1990b The quest for Tethys in the western Pacific palaeogeodynamic maps 507 for Cenozoic time Bull Soc Geol Fr.6: 907-913 508Riddle BR, Hafner DJ 2010 Integrating pattern with process at biogeographic boundaries: 509 the legacy of Wallace Ecography 33: 321-325 510Rosen DE 1978 Vicariant patterns and historical explanation in biogeography Systematic 511 Zoology 27: 159-188 512Ruedi M 1995 Taxonomic revision of shrews of the genus Crocidura from the Sunda Shelf 513 and Sulawesi with description of two new species (Mammalia: Soricidae) Zoological 514 Journal of the Linnean Society 115: 211-265 515Swofford, D L 2003 PAUP*: phylogenetic analysis using parsimony (*and other methods) 516 Version Sunderland, MA: Sinauer Associates 517Turner H, Hovenkamp P, van Welzen PC 2001 Biogeography of Southeast Asia and the 518 West pacific Journal of Biogeography 28: 217-230 519Ung V 2013 [Review of] Biotic Evolution and Environmental Change in Southeast Asia -520 Edited by David Gower, Kenneth Johnson, James Richardson, Brian Rosen, Lukas Ruber 521 and Suzanne Williams Systematic Biology 62: 913-915 522van Oosterzee P 1997 Where worlds collide: the Wallace line Ithaca and London: Cornell 523 University Press 524van Tol J 1998 The Odonata of Sulawesi and adjacent islands Part A new genus and 525 species of Chlorocyphidae from South-East Sulawesi, Zoölogische Verhandelingen 323: 526 441-448 527van Tol J, Gasmann D 2007 Zoogeography of freshwater invertebrates of Southeast Asia, 528 with special reference to Odonata In: Renema W, eds Biogeography, Time, and Space: 529 Distributions, Barriers, and Islands Springer, 45-91 530Weber MCW 1902 Der Indo-Australische Archipel und die Geschichte Seiner Tierwelt 531 Jena : Gustav Fischer, Jena 532Woodruff David S 2010 Biogeography and conservation in Southeast Asia: how 2.7 million 533 years of repeated environmental fluctuations affect today’s patterns and future of the 534 remaining refugial–phase biodiversity Biodiversity Conservation 19: 919-941 535Zaragüeta-Bagils R, Ung V, Grand A, Vignes-Lebbe R, Cao N, Ducasse J 2012 LisBeth: 536 new cladistics for phylogenetics and biogeography Comptes Rendus de Palevol 11: 563- 537 566 538 539 540SUPPORTING INFORMATION 541Appendix S1: Distributions of all taxa in the corresponding areas 542Appendix S2: Algorithm for traceability of characters 543 544 ... boundary of the North 177Moluccas, separating first Timor and Australia and then, between the islands of the Babar and 178Tanimbar groups, west around Buru and Halmahera and to the Pacific? ?? (van... analysis (Fig 3): The tree is clearly divided in 171two parts separating the Pacific islands (Fig 3, ? ?Pacific? ?? clade and Table 4) from the rest of 17 2Southeast Asia (Fig 3, ? ?Southeast? ?? Asian clade—‘Australian’... 2001 Biogeography of Southeast Asia and the 518 West pacific Journal of Biogeography 28: 217-230 519Ung V 2013 [Review of] Biotic Evolution and Environmental Change in Southeast Asia -520 Edited