Journal of Biogeography. 2019;1–16. wileyonlinelibrary.com/journal/jbi  |  1 Received: 29 November 2018  |  Revised: 18 December 2018  |  Accepted: 1 February 2019 DOI: 10.1111/jbi.13551 S P E C I A L I S S U E The shape of biogeography: Endemism, maps, and classification of fish distributions in the western Pacific Lynne R. Parenti Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, District of Columbia Correspondence Lynne R. Parenti, Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, DC. Email: parentil@si.edu Editor: Dr. Malte Ebach Abstract Aim: The aim of this study is to describe and illustrate with maps the distribution and shape of biotic areas of endemism of fish clades in the western Pacific, including East Asia and Australasia. This study also depicts the allopatric or sympatric distributions of major subclades and makes general statements about the shared history among the areas using information provided by the shape of the distributions, phylogeny, and geology. This study also tests the hypothesis that Australasia is a locus of survival or differentiation for inferred “basal” subclades of global taxa. Location: Western Pacific. Taxa: Teleost fishes. Methods: Clades are mapped to illustrate the shape of their distribution and to infer the shared relationships among distribution, phylogeny, and geology. Phylogenetic hypotheses for groups of taxa are converted into area cladograms which are used to summarize area relationships. This information is compared with that from other clades, ecology, and geology, to make general statements about the biogeographical history of the fish biota. Results: The distribution of each clade has an identifiable shape. The “basal” sub- clades of many widespread teleost fish taxa are endemic to the western Pacific, in- cluding East Asia and Australasia; hence, this is the inferred location of a principal vicariant event for each clade. The “basal” subclade and its sister clade may be al- lopatric and represent remnants of general distributional patterns. The distribution of a clade, including its geography and its relationship with a geographical or geologi- cal feature, conveys historical information about the clade and the area in which it evolved: its area of endemism. These features repeat in unrelated or distantly related taxa in the biota indicating a shared history. Main conclusions: The western Pacific including East Asia and Australasia is a locus of survival or differentiation, the site of the principal breaks in the distribution, for an array of teleost clades. It is not their center of origin. Biology and geology should be integrated in biogeographical analyses. Modern distributions reflect remnants of once more widespread distributions. When biogeography acknowledges the connec- tions and correlations between geology and biology, it implements a guiding principle of the work of Alexander von Humboldt. Published 2019. This article is a U.S. Government work and is in the public domain in the USA 2  |     PARENTI 1  | INTRODUC TION Maps are critical to biogeography: they depict the contours and shapes of biological distributions and contain information about the history of those distributions. To the extent that they coincide with geological or geographical features, past or present, they also illustrate the intimate relationship between biology and geology. Alexander von Humboldt (1850) was among the first scientists to recognize and emphasize the critical role that maps play in biogeog- raphy. He understood and used the power of maps, not just to con- vey his ideas about the relationship between biology and geology, but also to generate them: Central to Humboldt was the idea that spatial dis- tribution might reveal the complex interdependence in the natural world, and he spent much of his life trying to redefine geography along those lines. To a great extent, he invoked maps and graphical visual- izations not only to illustrate his ideas about spatial distribution, but to formulate them. [Italics in original]. Schulten (2018) Along with maps, biogeographical classification—the naming and ranking of areas—engaged biogeographers from the start and natu- rally linked biogeography with evolution. Alfred Russel Wallace (1863) recognized a marked separation between the Asian and Australian biota in the center of the Indo- Australian Archipelago. The division was both biological and geological. Two principles—endemism and terrane fidelity—came together on maps to illustrate that life and Earth evolved together. Endemism is the critical concept in biogeography that links the organism with its habitat (Parenti & Ebach, 2009). An area of ende- mism has a particular shape or morphology that reflects the history of a lineage. An area of endemism has discrete contours and bound- aries and may include breaks or broad gaps, as in disjunct or anti- tropical distributions (e.g., Gill & Mooi, 2017: fig. 6.7). These breaks are not singular, but repeat for many unrelated or distantly related taxa in a biota. An area may be unique to a clade, but it may also reflect general, congruent distributions, such as Pantropical, Holarctic, Trans- Pacific or Trans- Atlantic, or just remnants of those distributional patterns (Humphries & Parenti, 1999; Parenti, 2008). The broad, general dis- tributional patterns shared across taxa in a biota form the basis of a biogeographical classification (e.g., Ebach & Michaux, 2017; Ebach, Morrone, Parenti, & Viloria, 2008) (Figure 1). Biological and geological boundaries that coincide may be hypoth- esized to share a history, to have been formed at the same time and by the same events. For example, Springer (1982; Figure 1) examined in detail the distribution and endemism of marine shore fishes of the Pacific and their relationship with the Pacific, Philippine, and other lithospheric plates. He inferred that formation and activity of the tec- tonic plates effected the distribution of the plate's endemic shore- fishes and other taxa. Concordance between geology and biology was demonstrated also by Hendrickson (1986) for lungless salamanders distributed on accreted terranes in western North America and, more recently, by Gottscho (2016) for zoogeographical patterns that match K E Y W O R D S allopatry, area morphology, biogeographical patterns, dispersal, pantropical, sympatry, terrane fidelity, vicariance F IGURE  1 One example of the correlation of the distribution of a fish taxon with lithospheric plates; the distribution of fishes of the percomorph family Clinidae (blennies). Each dot represents a collection or report of specimens. The star is an unspecified locality in New Guinea. The X marks the approximate location of Palau on the southeastern margin of the Philippine Plate. The large asterisk (*) marks the island of Sulawesi, which is separated from the other islands to indicate its composite geology that was not shown in the original map. Abbreviations: CP, Cocos Plate; I- A Plate, Indian- Australian Plate; PP, Philippine Plate. Modified from Springer (1982: fig. 15)      |  3PARENTI fracture zones of the San Andreas Fault system. Terrane fidelity, the persistence of an organism or a lineage on a geological terrane, links an organism with a discrete geological region. Terrane fidelity, coupled with natal philopatry, the regular and routine return to place of birth (Heads, 2012: 383), characterizes the distributions of many migra- tory taxa, such as birds (e.g., Wolfson, 1948) and fishes (e.g., Parenti, 2008). This method of “spatial correlation” of taxa and tectonics forms the basis of panbiogeography, as expounded by Croizat (1964), Craw, Grehan, and Heads (1999), Michaux (2010), Heads (2005, 2011, 2012, 2014), and Grehan and Mielke (2018), among others. Australasia, Heads (2014) argued, has a special place in biological evolution: it is home to the “basal” members of many global clades such as the New Zealand tuatara, which is sister to all other lizards and snakes. Heads (2014) defined Australasia—the area south of Asia—as comprising Australia and the large islands of New Guinea, New Caledonia, and New Zealand. It is not a natural biogeograph- ical region and we do not know the relationships among the areas considered to lie within it (see Ebach, 2017). Furthermore, many of the “basal” taxa, as identified by Heads (2014), live more broadly in the western Pacific, including Palau and the Philippines. Therefore, I expand the area of interest for bony fishes to the western Pacific, broadly defined. The western Pacific distribution of fishes occupies an arc from the northern limits of East Asia south to New Zealand and the east coast of Australia (Motomura, Alama, Muto, Babaran, & Ishikawa, 2017:16: Figure 1). The aim of this paper is to demonstrate that the western Pacific is home to the “basal” clade, or subclade, of many widespread bony fish taxa, including freshwater, coastal, and marine lineages. A “basal” clade is a small sister of a diverse, widespread group. The break be- tween these sister- groups is “basal.” “Basal” does not mean ancestral, and further, the western Pacific should not be misinterpreted as the center of origin for teleost lineages on Earth, but rather recognized as a locus of survival or differentiation where major breaks or inter- ruptions in widespread distributions occurred. These breaks may be explained by geology and ecology. This approach stands in contrast to applications of the prevailing long- distance dispersal from a center of origin model of biological distribution. In those studies, assump- tions about the age of taxa and their supposed dispersal ability are treated routinely as evidence of particular biogeographical explana- tions, rather than as hypotheses (Parenti & Ebach, 2013). 2  | MATERIAL S AND METHODS Clades from across teleost phylogeny with broad distributions throughout the western Pacific, and global sister groups, are mapped to depict their distributional limits and identify where major breaks or disjunctions in distributions occur. The taxa represent a phyloge- netic range of teleost fishes, Teleostei sensu de Pinna (1996), from the most primitive Osteoglossomorpha to the derived Percomorpha. The teleost clades with the largest number of freshwater (Cypriniformes, about 489 genera and 4,205 species) and marine (Gobioidei or Gobiiformes, about 321 genera and 2,167 species) taxa are included (Nelson, Grande, & Wilson, 2016). Data on distributions are from systematic analyses and the highly informative maps of Berra (1981, 2001, 2007); these are supplemented by catalog records in FishNet2 (http://www.fishnet2.net/.). Phylogenetic hypotheses are converted into area cladograms, following the methods outlined by Nelson and Platnick (1981) and elaborated elsewhere (e.g., Parenti & Ebach, 2009): names of taxa are replaced by the areas in which they live. These area cladograms are used to summarize the information on area relationships contained in the phylogenetic hypotheses. This in- formation is compared with geological hypotheses to make general statements about the distributional history of major clades of fishes in the western Pacific. 3  | RESULTS 3.1 | Distribution and phylogenetic relationships of fish clades 3.1.1 | Osteoglossomorpha (bony- tongues), family Osteoglossidae The osteoglossomorph lineage is one of the most “basal” of all tel- eost fishes (see recent review by Hilton & Lavoué, 2018). Extant and fossil taxa are distributed broadly throughout the tropics and tem- perate Laurasia (Hilton & Lavoué, 2018: fig. 18, top). The family Hiodontidae is the most “basal” of the main osteoglos- somorph lineage: its Recent representatives in the genus Hiodon live in North American freshwaters; fossils in the genus †Eohiodon are known from East Asia and North America. Additional fossil osteoglos- somorphs are represented in several assemblages of Late Jurassic- Early Cretaceous fishes of eastern China (Chang & Chow, 1986: fig. 1). Several of the fossil genera are stem- group osteoglossomorphs; others are more closely related to subclades in the main osteoglossomorph lin- eage. These taxa comprise a boreal or Holarctic group. All other osteo- glossomorphs are pantropical (Hilton & Lavoué, 2018: fig. 18, bottom). Hilton and Lavoué (2018) defined six areas for fossil and liv- ing osteoglossomorphs: Afrotropics (A), Neotropics (B), Orient (C), Australia (D), Nearctic (E), and Northeastern Palearctic (Cretaceous) (F). Areas A through D are pantropical; E and F Holarctic or boreal. Information on relationships among the pantropical areas is con- tained in the osteoglossid clade (Hilton & Lavoué, 2018: fig. 18, bot- tom): Arapaima (B) is sister to Heterotis (A) which together are sister to the fossil genus †Sinoglossus (C). This clade is sister to the trans- Pacific sister genera Osteolgossum (B) and Scleropages (CD). The in- formative relationships among the four pantropical areas are: ((A, B), (C,D)) (Figure 2). A second clade of genera from areas A (families Mormyridae and Gymnarchidae) and C (family Notopteridae) pro- vide no additional information on area relationships. Holarctic areas E and F repeat on the cladogram, as a widespread area EF or with E as sister to one or more pantropical areas. The break between the “basal” temperate fossil assemblages of East Asia (area F) and tropical Scleropages (in area CD) is marked (Figure 3). These East Asian fossils are part of a Holarctic clade (Figure 2). 4  |     PARENTI 3.1.2 | Anguilliformes (freshwater and marine eels) The hypothesized sister group of all other anguilliforms, a broadly distributed taxon, is the deep sea Protanguilla (see Johnson et al., 2012). It was described from a 35 m deep cave in Palau, an is- land chain on the eastern margin of the Philippine tectonic plate (Figures 1 and 4). Protanguilla is estimated to have evolved at least 200 Ma, using a molecular clock analysis of sequence divergence (Johnson et al., 2012). The freshwater eels, the family Anguillidae, comprise some 19 species and subspecies all in the genus Anguilla (Teng, Lin, & Tzeng, 2009). The oldest fossils of Anguilla are from 50 to 55 Ma (Patterson, 1993). Today anguillids live in pantropical and temperate habitats and exhibit remnants of antitropical distributions (Parenti, 2008: fig. 6). A recent molecular analysis of the species of Anguilla concluded that phylogenetic relationships among species were best represented in a polytomy which was congruent with the hypothesis that the genus had undergone several different, potentially rapid, radiations (Teng et al., 2009: fig. 6a). The network was unrooted. One radiation comprises four species: A. japonica, A. reinhardtii, A. celebensis, and A. megastoma. In an earlier, rooted molecular analysis by Lin, Poh, and Tzeng (2001), these four species formed a clade comprising two sister group pairs: ((A. japonica, A. reinhardtii), (A. celebesensis, A. me- gastoma)). Anguilla japonica (north- western Pacific) and A. reinhardtii (south- western Pacific) have an antitropical distribution, whereas A. celebesensis (western Pacific) and A. megastoma (central- Pacific) are tropical (Teng et al., 2009: fig. 6b). F IGURE  2 Area relationships of osteoglossomorph fishes as inferred from information in the analysis of Hilton and Lavoué (2018). See text for explanation F IGURE  3 Distributional limits of the osteoglossomorph family Osteoglossidae, genus Scleropages (following Berra, 2007; Roberts, 2012) and fossil limits of stem- group osteoglossomorphs in East Asia (following Chang & Chow, 1986: fig. 1). A solid line indicates approximate position of the Red River Fault. Two other osteoglossomorph genera, Notopterus and Chitala, in the family Notopteridae, live in Indo- Malaya, allopatric with the East Asia fossils; their distribution is not illustrated here. All other osteoglossomorphs live in North America, South America, and Africa      |  5PARENTI All species of Anguillidae have a catadromous life history with a marine leptocephalus larval phase. Anguilla japonica, the Japanese eel, is distributed along the coast of East Asia (Ege, 1939; Tsukamoto, 2009: Figure 4). Adults migrate from their stream habitats to a spawning ground in the southeastern margin of the Philippine Plate. 3.1.3 | Osmeriformes (freshwater smelts) The smelt suborder Osmeroidei comprises the true smelts and Southern Hemisphere smelts. Three families in the suborder form a clade, following Saruwatari, Oohara, and Kobayashi (2002): Osmeridae (Holarctic in marine and coastal freshwaters), Plecoglossidae (freshwater and marine in East Asia, Japan, and Taiwan), and Salangidae, the icefishes or noodlefishes, a family of small, neotenic, anadromous fishes (East Asia from the Sakhalin Peninsula to Vietnam: Figure 5). The family Plecoglossidae con- tains just one species, Plecoglossus altivelis. In the Saruwatari et al. (2002) analysis, salangids are the “basal” sister group of the smelt families Plecoglossidae and Osmeridae: (Salangidae, (Plecoglossidae, Osmeridae)). The salangids and Plecoglossus are sympatric throughout much of their ranges and both are sympatric with osmerids in the northern part of their ranges (Figure 5). There is an initial break between a western Pacific (East Asia) taxon and the rest of the clade; these are now sym- patric. Area relationships for the suborder are: (East Asia, (East Asia, Holarctic)). 3.1.4 | Ostariophysi (tetras, minnows, catfishes, knifefishes); order Cypriniformes (minnows) Phylogenetic relationships among the Cypriniformes are contro- versial. A superfamily Cobitoidea was recognized by Conway (2011) and Wang et al. (2016) to comprise seven monophyletic families: the algae eaters, Gyrinocheilidae; the suckers, Catostomidae; and the loaches, Cobitidae, Botiidae, Vaillantellidae, Balitoridae, and Nemacheilidae. In a morphological analysis (Conway, 2011: fig. 43), the Gyrinocheilidae and Catostomidae comprise a sis- ter group that is in turn sister to all remaining cobitoids, the five families of loaches. Betancur- R et al. (2017) rejected cobitoid monophyly following a molecular analysis by Stout, Tan, Lemmon, Lemmon, and Armbruster (2016). In the Stout et al. (2016: fig. 2) tree, the Gyrinocheilidae is “basal” to all other Cypriniformes to which it is related as: (Gyrinocheilidae, (Catostomidae, (all other cypriniforms))). F IGURE  4 Distributional limits of adult Anguilla japonica outlined by a dotted line (after Ege, 1939: fig. 32; Tsukamoto, 2009). Anguillid eels are catadromous and migrate to marine spawning grounds. Anguilla japonica migrates to a spawning area (rectangle) on the southeastern margin of the Philippine Plate (approximate limits shaded). Palau (large solid circle) 6  |     PARENTI The Gyrinocheilidae has a restricted distribution in Southeast Asia (Figure 6). The Catostomidae is Holarctic with most of its spe- cies in North America and Russia and a single living taxon in China (Figure 6). These families are allopatric. 3.1.5 | Synbranchiformes (swamp eels) The pantropical or subtropical suborder Synbranchoidei is equiva- lent to the family Synbranchidae, revised by Rosen and Greenwood (1976) to comprise two subfamilies, the Macrotriminae (with one genus, Macrotrema) and Synbranchinae (with three genera, Synbranchus, Ophisternon, and Monopterus). (Macrotriminae was cor- rected to Macrotimatinae by Bailey & Gans, 1998). These fresh and brackish water fishes are not closely related to the Anguilliformes (above). The monotypic Macrotrema is sister to the three other genera of synbranchids (Rosen & Greenwood, 1976: figs. 66, 67). Its single species, Macrotrema caligans, lives in Thailand, Vietnam, and the Malay Peninsula, including Singapore (Kottelat, 1989; Figure 7). Ophisternon has a restricted pantropical distribution (Rosen, 1976: fig. 1). Synbranchus lives in Central and South America. The natural distribution of Monopterus includes South, Southeast, and East Asia, with two species in West Africa (Bailey & Gans, 1998). The cladogram of relationships from Rosen and Greenwood (1976) is: (Macrotrema, (Ophisternon, (Synbranchus, Monopterus))). An area cladogram is: (Thailand, Malaysia, Singapore, (Pantropical, (Central and South America, South, Southeast, and East Asia))) (Figure 7). 3.1.6 | Gasterosteiformes (sticklebacks), family Indostomidae (armored sticklebacks) The Indostomidae comprises a single genus, Indostomus, with three species (Britz & Kottelat, 1999). Relationships of the family are controversial. Betancur- R et al. (2017) classified the Indostomidae in the Synbranchiformes. In contrast, Britz and Johnson (2002) concluded that Indostomus is a gasterosteoid, sister to the family Gasterosteidae. Indostomus is restricted to freshwater habitats in Indo- Burma (Figure 8). Gasterosteids are Holarctic (Figure 8). These two families are allopatric. F IGURE  5 Distributional limits of the families of the Osmeroidei, the freshwater smelts (following Berra, 1981: 30–32). The families Salangidae (black line) and Plecoglossidae (enclosed by dotted line) are sympatric throughout much of their ranges in East Asia. They are sympatric with the Osmeridae (shaded), a Holarctic taxon, in the northern part of their ranges. The families are related as: (Salangidae, (Plecoglossidae, Osmeridae)). Their area relationships are: (East Asia, (East Asia, Holarctic))      |  7PARENTI 3.1.7 | Syngnathiformes (pipefishes and relatives), family Syngnathidae (pipefishes and seahorses) Seahorses, the subfamily Syngnathinae, are a group of some 40 species of shallow- marine fishes distributed throughout the tropical and tem- perate zones from approximately 45° N to 45° S (Lourie, 2016:39). Their pelagic larval phase is relatively short, estimated at 2–4 weeks (Lourie, 2016). The species exhibit discrete, endemic distribution patterns and are notable, but not unique among marine fishes, for adherence to geographical boundaries such as Wallace's Line (Figure 9; Lourie & Vincent, 2004: fig. 1; see also Woodland, 1986; Collette, 2005). One species, H. kuda, is widespread throughout the Indo- Pacific; it is in need of taxonomic revision as its populations are distinct, with some recognized at the subspecific level (Lourie, 2016: 122, map). The pygmy Bargibant's Seahorse, H. bargibanti, is so deeply genet- ically distinct from the others that it is hypothesized to represent “…an ancient divergence…” from the main group of seahorses (Teske, Cherry, & Matthee, 2004:281); it was used as an outgroup taxon in the Teske et al. (2004) molecular analysis of seahorse phylogeny. Bargibant's Seahorse is distributed throughout the western Pacific (including the eastern portion of the Indo- Australian Archipelago) from Japan to New Caledonia (Lourie, 2016: 74, map; Figure 10). The “basal” clade of the main seahorse lineage comprises two species: H. breviceps and H. abdominalis (see Teske et al., 2004) that live in southern Australasia (Heads, 2014:71: Lourie, 2016: 85, 86, maps; Figure 10). These two species overlap in the center of their ranges in south- eastern Australia. Following Teske et al. (2004) and Lourie (2016), relationships of seahorses in the genus Hippocampus are: (H. barg- ibanti, ((H. breviceps, H. abdominalis), (all other species of Hippocampus))). The area cladogram is: (Western Pacific, (southern Australasia, (tropical and temperate zones from approximately 45° N to 45° S))). 3.1.8 | Atheriniformes (silversides). Suborder Atherinoidei (Old World silversides and rainbowfishes) The Atheriniformes is one of three orders in the monophyletic Atherinomorpha, which also comprises the killifishes, ricefishes, fly- ingfishes, and their relatives (Rosen & Parenti, 1981). In a molecular phylogenetic analysis of atheriniform fishes, Campanella et al. (2015: fig. 3) summarized the hypothesized relationships among represent- atives of all lineages in a time- calibrated phylogeny. They classified atheriniforms in two suborders: the New World Atherinopsoidei and F IGURE  6 Distributional limits of the allopatric cypriniform families Gyrinocheilidae (dotted line; following Berra, 2007: 97) and Catostomidae, in part (shaded; following Berra, 2007: 99). There are two hypotheses of relationships: (a) morphological, Gyrinocheilidae and Catostomidae are sister taxa and, in turn, sister to all other cypriniforms (Conway, 2011); (b) molecular, Gyrinocheilidae is the most “basal” cypriniform, and Catostomidae, a Holarctic taxon, is sister to the remaining cypriniforms (Stout et al., 2016) 8  |     PARENTI the Old World Atherinoidei. The family Phallostethidae is the most “basal” lineage in the atherinoid cladogram. Two subfamilies comprise the family Phallostethidae: the Phallostethinae with 24 species and the Dentatherininae with a sin- gle species, Dentatherina merceri (see Parenti, 2014). Campanella et al. (2015) included just two species of phallostethids in their analysis: Neostethus lankesteri and N. bicornis. The phallostethins are coastal freshwater and marine species of Southeast Asia; their sister taxon, the monotypic Dentatherina, is a coastal marine species that is sympatric with the phallostethins in the center of the Indo- Australian Archipelago and has a range that extends along the eastern limit of the western margin of the Pacific Plate to Fiji (Figure 11). The phallostethins and Dentatherina are sister to the remaining Old World atherinoids. The split between phallostethins/Dentatherina and the other atherinoids is estimated at 50–55 Ma (Campanella et al., 2015: fig. 3). 3.1.9 | Gobiiformes, suborder Gobioidei (gobies and sleepers) Gobioid fishes live broadly in pantropical and temperate fresh- water, estuarine, and marine habitats and exhibit a wide range of life- history patterns. Several subgroups are amphidromous, a spe- cialized type of diadromous life- history pattern in which adults live and breed in freshwater. Larvae are transported passively to the sea where they spend weeks or months transforming before migrating to upstream habitats (e.g., Parenti, 2008). Many gobioids, such as those in the subfamily Sicydiinae, are called freshwater fishes be- cause the adults are always taken in freshwater habitats, yet they have a significant marine life- history phase. This is true also of the anguillid eels discussed above. Two allopatric families are the most “basal” of the gobioids: Rhyacichthyidae and Odontobutidae. There are two competing hy- potheses of their relationships with all other gobioids. In the morpho- logical hypothesis of Hoese and Gill (1993: fig. 9), Rhyacichthyidae is at the base of the gobioid tree, with the Odontobutidae sister to all remaining gobioids, then classified in the family Gobiidae: (Rhyacichthyidae, (Odontobutidae, Gobiidae)). In a recent molec- ular hypothesis of Li, He, Jiang, Liu, and Li (2018), these two fam- ilies are sister taxa, and in turn sister to all remaining gobioids: ((Rhyacichthyidae, Odontobutidae), all other gobioids). The Rhyacichthyidae, with two species classified in the genus Rhyacichthys and one species in the genus Protogobius, lives broadly F IGURE  7 Distributional limits of the synbranchid eel genera Macrotrema (dark shading), Ophisternon (dotted line, in part), and Monopterus (light shading, in part). A fourth genus, Synbranchus, does not live in the western Pacific. The genera are related as: (Macrotrema, (Ophisternon, (Synbranchus, Monopterus))), following Rosen and Greenwood (1976). The area cladogram is: (Thailand, Malaysia, Singapore, (Pantropical, (Central and South America, South, Southeast, and East Asia)))      |  9PARENTI throughout the tropical western Pacific, along the eastern portion of the Indo- Australian Archipelago, and extends along the eastern limit of the western margin of the Pacific Plate to New Caledonia (Figure 12). The family Odontobutidae lives in temperate and subtropical waters of East Asia including Japan, Hainan Island, and northern Viet Nam and Laos (Figure 12). They are allopatric. Estimates of molecular divergence times between Rhyacichthyidae and Odontobutidae range from 47.3 Ma (Li et al., 2018) to 98 Ma (Chakrabarty, Davis, & Sparks, 2012). 4  | DISCUSSION The western Pacific, including East Asia and Australasia, is home to the “basal” clade for an array of widespread teleost fish clades. It is not considered the center of origin of these clades, but as a locus of differentiation—a place where many global clades were disrupted. Endemism and allopatry characterize the distributional patterns, but there is also sympatry or overlap of sister clades or other closely related taxa. This is not the sole major area of dif- ferentiation of teleost taxa on Earth; many teleost clades, such as the Characiformes and the Cichlidae, for example, have no representatives—living or fossil—in the western Pacific. But the generality of this pattern will be tested as our understanding of phylogenetic relationships increases. The cichlids are a pantropi- cal and subtropical freshwater family with no natural representa- tives in Indo- Australia outside of India and Sri Lanka (Berra, 2001: 440). Yet, in recent molecular phylogenetic studies (Betancur- R et al., 2017), the sister group of the cichlids is hypothesized to be the marine Pholidichthyidae, broadly distributed throughout the western Pacific (Springer, 1982). One general global biogeographical pattern, ((Austral, Boreal), Pantropical) (see Humphries & Parenti, 1999), is not duplicated wholly in any of the teleost taxa examined here, but there are rem- nants of the pattern in nearly all the distributions (Figure 2). Holarctic (or boreal) clades are sister to, or the next breaks in the cladogram, in osteoglossomorphs (Figure 3), osmeroids (Figure 5), cypriniforms (Figure 6), gasterosteiforms (Figure 8), and gobioids (Figure 12). Austral clades are sister to, or the next breaks in the cladogram, in the seahorses, Hippocampus (Figure 10). In all of these clades except for the osmeroids, the “basal” lineage and its relative are allopatric. Terrane fidelity marks the distribution of the anguilliforms, in- cluding the anguillids (Figure 4), the seahorses (Figures 9 and 10), and the atheriniforms and gobioids (Figures 11 and 12). In these last F IGURE  8 Distributional limits of the family Indostomidae (black shading; following Berra, 2007: 357) and its inferred sister taxon, the Gasterosteidae, in part (gray shading; following Berra, 2007: 351). Gasterosteidae is a Holarctic taxon 10  |     PARENTI two clades, the distributional limit of the easternmost taxon is at the eastern margin of the Indian- Australian lithospheric plate where it meets the western margin of the Pacific lithospheric plate (Figure 1). The “basal” anguilliform, Protanguilla, is marginal on the Philippine lithospheric plate. The age estimate of Protanguilla, at 200 Myr, is “…much older than the Palau- Kyushu Ridge itself, which formed as an island arc at ⁓60–70 Ma. The clade is older than the geomorphic structure to which it is endemic, a pattern shown in many groups” (Heads, 2014:388). Sympatry of sister taxa results from the inferred range expansion of either taxon (Croizat, Nelson, & Rosen, 1974). The osmeroid fam- ilies, related as (Salangidae, (Plecoglossidae, Osmeridae)), are sym- patric throughout much of their ranges in East Asia (Figure 5). The initial break between the Salangidae and the rest of the osmeroids is inferred to have been followed by the expansion of the range of one taxon into the other. The mechanism for these breaks and overlap could have been subsequent changes in sea level and other mac- roenvironmental fluctuations (see Li et al., 2018, for odontobutids in East Asia). The phylogenetic relationships and distribution of syn- branchiform eels (Figure 7), with Macrotrema sister to the remaining taxa, led Rosen (1976:434) to hypothesize that “…the primary event in the history of the Synbranchidae may have occurred in the Old World Tropics.” The overlapping (sympatric) distribution of the sister taxa Phallostethinae and Dentatherininae (Figure 11) was hypothe- sized to have been formed as the widespread ancestral population ‘…was disrupted by reconstruction and rearrangement of land and changes in sea level that may have facilitated expansion of one taxon into the range of the other” (Parenti & Ebach, 2013:815). The clades overlap in regions of accreted terranes, such as the southwestern arm of Sulawesi, a phenomenon seen in the distributions of other taxa (Heads, 2012: 286). What information a phylogeny may contribute to a biogeograph- ical analysis has been the focus of intense methodological debate for decades (Mooi, 2017; Parenti, 2017). The significance of fish distri- butions in the debate is exemplified by the distribution of the osteo- glossid fishes, which was compared to “… the ratites and Nothofagus [southern beeches] combined, with a trans- Pacific distribution extending beyond the shoulder into India and Africa, the heart of Gondwana” (Nelson & Ladiges, 2001:394). Despite the arguments in favor of the discovery of general patterns and repeated components of relationships in a biota, many modern biogeographical studies fol- low another paradigm. The CODA (Centre of Origin, Dispersal, and Adaptation) model of evolution, as defined by Lomolino and Brown (2009), is prominent (Heads, 2014). In this model, the history of each lineage in interpreted separately, starting with the inference of a center of origin followed by an inferred dispersal path, as specified by the phylogeny, to form the modern distribution (viz. Campanella, et al., 2015 for atheriniforms; Teng et al., 2009, for anguillid eels; Thacker, 2015, for gobioids). Teng et al. (2009: 6b) provide small maps of the distributional limits of each species of Anguilla. Thacker (2015) has no maps of taxon distributions, but includes two maps of inferred dispersal routes from a center of origin in the Indo- West F IGURE  9 Geographical distribution of lineages of the seahorse Hippocampus trimaculatus (modified from Lourie & Vincent, 2004: fig. 1) to demonstrate the close relationship between population distribution and Wallace's Line. Symbol size is in proportion to the number of individuals sampled from each location. Continental shelves that would have been subaerial during periods of low sea- level are shaded      |  11PARENTI Pacific of the families Gobionellidae (Thacker, 2015; fig. 2a) and Gobiidae (Thacker, 2015; fig. 2b). Campanella et al. (2015) include no maps. Using a center of origin method (ancestral area analysis; Matzke, 2013; Ree & Smith, 2008), Hilton and Lavoué (2018) concluded that osteoglossomorphs originated in East Asia and dispersed to their present locations. East Asia was “…the main place where the early diversification of the Osteoglossomorpha took place (includ- ing the most recent common ancestor, mrca) of the crown group Osteoglossomorpha, which lived during the Jurassic (about 190– 150 Ma)…” (Hilton & Lavoué, 2018:26). The break between the Holarctic and Pantropical regions of os- teoglossomorphs is illustrated as the break between the East Asian fossils and Scleropages, a representative of the Pantropical areas de- fined by Hilton and Lavoué (2018) as Orient and Australia (Figures 2 and 3). Disruption of a widespread ancestor is inferred here to have formed the pattern of Figures 2 and 3 as an alternative to the CODA hypothesis. Assumptions about dispersal ability also affect biogeographi- cal analyses (Parenti & Ebach, 2013). Some assume the absence of dispersal ability and a global distributional pattern to be at odds. According to Bailey and Gans (1998: 3): “Synbranchids presumably have a limited capacity for active dispersal. Nonetheless the family as a whole and the genus Ophisternon in particular (Rosen, 1976) have a broad pantropical distribution.” On seahorses, the genus Hippocampus, Teske et al. (2004: 274) assumed that “…the circum- global distribution of seahorses reflects major dispersal events.” In contrast, the vicariance assumption about cosmopolitan distribu- tions is that they were formed prior to major continental rearrange- ment (viz. Croizat et al., 1974). These modern biogeographical distributions of fishes reflect just remnants of once more complex and widespread distributions; they are incomplete spatial records, just as the fossil record is an incomplete temporal record. The ages of lineages known from fossils are underestimated, and can only get older with new fossil discov- eries. The Holostei, sister group of the Teleostei, includes the gars (Lepisosteiformes) and the bowfins (Amiiformes). Brito, Alvarado- Ortega, and Meunier (2017) recently described a new gar from the Upper Jurassic (about 157 Ma) of Mexico. The new fossil taxon “… extends the chronological range of lepisosteoids by about 46 Myr and of the lepisosteids by about 57 Myr, and fills a major morpholog- ical gap in current understanding [of] the early diversification of this group” (Brito et al., 2017). Calibrations of molecular clocks using the oldest known fossils of a clade will always result in underestimated ages. Despite this, fossils continue to be used to “anchor” the ages of phylogenies and dictate hypothetical dispersal routes. Teng et al. F IGURE  10 Distributional limits of the seahorses Hippocampus bargibanti (surrounded by a dotted line), H. breviceps (dark shading), and H. abdominalis (light shading), following Lourie (2016). Hippocampus breviceps and H. abdominalis are sympatric in the center of their ranges in south- eastern Australia. Relationships of seahorses in the genus Hippocampus are: (H. bargibanti, ((H. breviceps, H. abdominalis) (all other species of Hippocampus))), following Teske et al. (2004) and Lourie (2016) 12  |     PARENTI (2009: 819) hypothesized that “…The area near the equator of the Indo- West Pacific Ocean accommodates most Anguilla species and is therefore considered to be the center of origin of freshwater eels…”. On the evolution of the antitropical Anguilla japonica (north- western Pacific) and A. reinhardtii (south- western Pacific), Teng et al. (2009: 818) further hypothesized that “One moved northward (A. japonica), one moved southward (A. reinhardtii)…” They further argued that the global distribution of anguillids took place during the past 20 Myr, despite the occurrence of much older (50–55 Ma) fossils (Patterson, 1993). Because the divergence dates among the families of atherini- forms, as estimated using fossil calibrations of molecular sequence data, are all younger than Gondwanan break- up, Campanella et al. (2015) rejected vicariance explanations in favor of long- distance dispersal. They concluded (p. 21): “Ultimately, we do not have conclusive evidence for a vicariance hypothesis in New World sil- versides, but do find that oceanic dispersal was a primary driver in shaping the modern distribution of Atheriniformes.” But, as noted by Heads (2014) and others, rejection of a Gondwanan break- up model does not mean rejection of all vicariance. The overlap of the phallostethid clades (Figure 11) may be explained by the juxtaposi- tion of these areas during assembly of the modern Indo- Australian archipelago over the past 50 My (Hall, 2002; Parenti & Ebach, 2010, 2013). Furthermore, estimates of divergence time using molecular data may differ markedly depending on the method of calibration and the loci. For example, the divergence time between the gobi- oid families Rhyacichthyidae and Odontobutidae was estimated at 98 Ma (Late Cretaceous) by Chakrabarty et al. (2012) and at 47.3 Ma (mid- Eocene) by Li et al. (2018). The geology of the broad western Pacific region treated here is complex. Throughout the Phanerozoic, continental terranes/blocks rifted from the Indian- west Australian margin of eastern Gondwana to form, in successive stages, the Paleo- Tethys, Meso- Tethys, and Ceno- Tethys oceans (Metcalfe, 2011, 2013). In turn, each of these ocean basins closed and are represented on present- day Asia by complex suture zones of accreted terranes. A general break between the Southeast Asian/Australasian and East Asian biota is recognized by endemism and allopatry (e.g., Figures 3, 6, 8, 12). The break is coincident with the Red River Fault, a major strike- slip fault between the South China and Indochina blocks (Zhu, Graham, & McHargue, 2009; Figure 3). This break or disruption between “basal” clades and their sister taxa is hypoth- esized to date at least to the Mesozoic, based on the maximum estimated divergence times between the basal gobioid families. It could be older based on the age of osteoglossomorph fossils, which date to the Late Jurassic- Early Cretaceous. The divergence time between the Southeast Asian and Australian species of the osteo- glossid genus Scleropages (Figure 3) was estimated, using molecular data, at 138 ± 18 Ma by Kumazawa and Nishida (2000). They offered this as support for the hypothesis that part of the fauna was carried on Gondwanan terranes and accreted to the Asian continent in the Early Cretaceous. Biotic distributional patterns are a record of the geological history of the region. The celebrated high biological diversity of the western Pacific has been documented by tallying the high number of marine F IGURE  11 Distributional limits of the atherinoid sister taxa Phallostethinae and Dentatherininae (= Dentatherina); following Parenti & Ebach, 2013). They are related to the rest of the Old World atherinoid fishes as: ((Phallostethinae, Dentatherina), all other Old World atherinoids) (Campanella et al., 2015)      |  13PARENTI species that live there, especially in the central area bounded by New Guinea, the Philippines, and the Malay Peninsula, known as the East Indies Triangle (Briggs, 1966, 2005), the “fertile triangle” (Briggs, 1974), the Indo- Malayan Triangle (Donaldson, 1986) or the Coral Triangle (Barber, 2009; Hoeksema, 2007), among other names. The Philippines have been further characterized as the “center of the center” of marine shore fish biodiversity (Carpenter & Springer, 2005). This high species diversity (or the presence of a “basal” taxon) does not mean that this is the center of origin of teleost fishes or even that it is a fixed center of diversity. As noted by Dornburg, Moore, Beaulieu, Eytan, and Near (2014), Thacker (2015), and oth- ers, the center of diversity of teleosts, as measured in numbers of species, has shifted through time with openings and closures of the succession of Tethyan oceans (Metcalfe, 2011, 2013). Patterns de- scribed here are for clades regardless of the numbers of species in each. Distributional patterns retain the historical component of clades: where and how they evolved, where and how they were disrupted. The distributions for teleost fishes, when combined with informa- tion from phylogenetic analyses, reveal general, repeated patterns. They demonstrate terrane fidelity, allopatry, sympatry on accreted terranes, and other characteristics conveyed readily in maps. These distributional patterns are the shape of biogeography, the shape of distributions and phylogenies that allows us to infer shared histories. To find congruence between biological and geological phenomena as illustrated with maps was at the core of the science of Alexander von Humboldt. Biogeography will move forward when we acknowl- edge the connections and correlations between geology and biol- ogy, especially when we look for them in distributional maps. ACKNOWLEDG EMENTS Malte Ebach, University of New South Wales, Sydney, and Michael Heads, New Zealand, read and commented on a previous draft of this MS. Tim Berra, Ohio State University, Mansfield, granted permission to reproduce data from his published maps, here in Figures 3, 5–6, 8 and 12. Sara Lourie, McGill University, Montreal, granted permission to reproduce the information in Figure 9 and provided the data for Figure 10. Riley Pollom drew raw maps used to create Figure 10. Dan Cole, Smithsonian's National Museum of Natural History, provided F IGURE  12 Distributional limits of the allopatric gobioid families Odontobutidae (light shading; following information in Berra, 2007:460 and Li et al., 2018: fig. 5) and Rhyacichthyidae (dark shading and encircled by dotted line; following Berra, 2007: 458). There are two competing hypotheses of their relationships to all other gobioids: (Rhyacichthyidae, (Odontobutidae, all other gobioids)) and ((Rhyacichthyidae, Odontobutidae), all other gobioids). All other gobioids are pantropical and temperate, and many sympatric with these two families 14  |     PARENTI the base maps used in Figures 3–8, and 10–12. I thank all for their time and interest. ORCID Lynne R. Parenti https://orcid.org/0000-0002-3279-7689 R E FE R E N C E S Bailey, R. M., & Gans, C. (1998). Two new synbranchid fishes, Monopterus roseni from Peninsular India and M. desilvai from Sri Lanka. Occasional Papers of the Museum of Zoology University of Michigan, 726, 1–18. Barber, P. H. (2009). The challenge of understanding the Coral Triangle bio- diversity hotspot. Journal of Biogeography, 36, 260–265. Berra, T. M. (1981). An atlas of distribution of the freshwater fish families of the world. Lincoln, NE: University of Nebraska Press. Berra, T. M. (2001). Freshwater fish distribution. New York, NY: Academic Press. Berra, T. M. (2007). Freshwater fish distribution. Chicago, IL: University of Chicago Press. https://doi.org/10.7208/chicago/9780226044439. 001.0001 Betancur-R, R., Wiley, E. O., Arratia, G., Acero, A., Bailly, N., Miya, M., … Ortí, G. (2017). Phylogenetic classification of bony fishes. BMC Evolutionary Biology, 17, 162. https://doi.org/10.1186/s12862-017-0958-3 Briggs, J. C. (1966). Zoogeography and evolution. Evolution, 20, 282–289. https://doi.org/10.1111/j.1558-5646.1966.tb03366.x Briggs, J. C. (1974). Marine zoogeography. New York, NY: McGraw Hill Book Company. Briggs, J. C. (2005). The marine East Indies: Diversity and spe- ciation. Journal of Biogeography, 32, 1517–1522. https://doi. org/10.1111/j.1365-2699.2005.01266.x Brito, P., Alvarado-Ortega, J., & Meunier, F. J. (2017). Earliest known lepisosteoid extends the range of anatomically modern gars to the Late Jurassic. Scientific Reports, 7, 17830. https://doi.org/10.1038/ s41598-017-17984-w Britz, R., & Johnson, G. D. (2002). “Paradox Lost”: Skeletal ontogeny of Indostomus paradoxus and its significance for the phylogenetic relation- ships of Indostomidae (Teleostei, Gasterosteiformes). American Museum Novitates, 3383, 1–43. https://doi.org/10.1206/0003-0082(2002)383< ;0001:PLSOOI>2.0.CO;2 Britz, R., & Kottelat, M. (1999). Two new species of gasterosteiform fishes of the genus Indostomus (Teleostei: Indostomidae). Ichthyological Exploration of Freshwaters, 10(4), 327–336. Campanella, D., Hughes, L. C., Unmack, P. J., Bloom, D. D., Piller, K. R., & Ortí, G. (2015). Multi-locus fossil-calibrated phylogeny of Atheriniformes (Teleostei, Ovalentaria). Molecular Phylogenetics and Evolution, 86, 8–23. Carpenter, K. E., & Springer, V. G. (2005). The center of the center of marine shorefish biodiversity: The Philippine Islands. Environmental Biology of Fishes, 72, 467–480. https://doi.org/10.1007/ s10641-004-3154-4 Chakrabarty, P., Davis, M. P., & Sparks, J. S. (2012). The first record of a trans- oceanic sister- group relationship between obligate vertebrate troglobites. PLoS ONE, 7, e44083. https://doi.org/10.1371/journal. pone.0044083 Chang, M.-M., & Chow, C.-C. (1986). In T. Uyeno, R. Arai, T. Taniuchi, & K. Matsuura (Eds.), Stratigraphic and geographic distributions of the Late Mesozoic and Cenozoic fishes of China (pp. 529–539). Tokyo, Japan: Ichthyological Society of Japan. Collette, B. B. (2005). Is the east–west division of haplotypes of the three- spot seahorse along Wallace's Line novel among ma- rine organisms? Journal of Biogeography, 32, 1286. https://doi. org/10.1111/j.1365-2699.2005.01277.x Conway, K. W. (2011). Osteology of the South Asian Genus Psilorhynchus McClelland, 1839 (Teleostei: Ostariophysi: Psilorhynchidae), with investigation of its phylogenetic relationships within the order Cypriniformes. Zoological Journal of the Linnean Society, 163, 50–154. Craw, R. C., Grehan, J. R., & Heads, M. J. (1999). Panbiogeography: Tracking the history of life. New York, NY: Oxford University Press. Croizat, L. (1964). Space, time, form: The biological synthesis. Caracas, Venezuela: Author. Croizat, L., Nelson, G., & Rosen, D. E. (1974). Centers of origin and related concepts. Systematic Zoology, 23, 265–287. https://doi. org/10.2307/2412139 de Pinna, M. C. C. (1996). Teleostean monophyly. In M. L. J. S. Stiassny, L. R. Parenti, & G. D. Johnson (Eds.), Interrelationships of fishes (pp. 147–162). New York, NY: Academic Press. https://doi.org/10.1016/ B978-012670950-6/50008-4 Donaldson, T. J. (1986). In T. Uyeno, R. Arai, T. Taniuchi, & K. Matsuura (Eds.), Distribution and species richness patterns of Indo-West Pacific Cirrhitidae: Support for Woodland's hypothesis (pp. 623–628). Tokyo, Japan: Ichthyological Society of Japan. Dornburg, A., Moore, J., Beaulieu, J. M., Eytan, R. I., & Near, T. J. (2014). The impact of shifts in marine biodiversity hotspots on patterns of range evolution: Evidence from the Holocentridae (squirrelfishes and soldierfishes). Evolution, 69, 146–161. Ebach, M. C. (2017). Reinvention of Australasian biogeography. Reform, revolt and rebellion. CSIRO Publishing: Clayton, Victoria. https://doi. org/10.1071/9781486304844 Ebach, M. C., & Michaux, B. (2017). Establishing a framework for a nat- ural area taxonomy. Acta Biotheoretica, 65, 167–177. https://doi. org/10.1007/s10441-017-9310-y Ebach, M. C., Morrone, J. J., Parenti, L. R., & Viloria, Á. L. (2008). International code of area nomenclature. Journal of Biogeography, 35, 1153–1157. https://doi.org/10.1111/j.1365-2699.2008.01920.x Ege, V. (1939). A revision of the genus Anguilla Shaw. A systematic, phy- logenetic and geographical study. Dana Report, 3(16), 1–256, pl. I-VI. Gill, A. C., & Mooi, R. D. (2017). Biogeography of Australian marine fishes. In M. C. Ebach (Ed.), Handbook of Australasian biogeography (pp. 101–128). Boca Raton, FL: CRC Press/Taylor and Francis Group. Gottscho, A. D. (2016). Zoogeography of the San Andreas Fault system: Great Pacific Fracture Zones correspond with spatially concordant phylogeographic boundaries in western North America. Biological Reviews, 91, 235–254. https://doi.org/10.1111/brv.12167 Grehan, J. R., & Mielke, C. G. C. (2018). Evolutionary biogeography and tectonic history of ghost moth families Hepialidae, Mnesarchaeidae, and Palaeosetidae in the Southwest Pacific (Lepidoptera: Exoporia). Zootaxa, 4415, 243–275. https://doi.org/10.11646/zootaxa.4415.2.2 Hall, R. (2002). Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific: Computer- based reconstructions, model and animations. Journal of Asian Earth Science, 20, 353–431. https:// doi.org/10.1016/S1367-9120(01)00069-4 Heads, M. (2005). Towards a panbiogeography of the seas. Biological Journal of the Linnean Society, 84, 675–723. https://doi. org/10.1111/j.1095-8312.2005.00466.x Heads, M. (2011). Old taxa on young islands: A critique of the use of island age to date island- endemic clades and calibrate phylogenies. Systematic Biology, 60, 204–218. https://doi.org/10.1093/sysbio/ syq075 Heads, M. (2012). Molecular panbiogeography of the tropics. Berkeley, CA: University of California Press. https://doi.org/10.1525/ california/9780520271968.001.0001 Heads, M. (2014). Biogeography of Australasia. A molecular analysis. Cambridge University Press: Cambridge, UK. Hendrickson, D. (1986). Congruence of bolitoglossine biogeography and phylogeny with geological history: Paleotransport on dis- placed suspected terranes. Cladistics, 2, 113–129. https://doi. org/10.1111/j.1096-0031.1986.tb00447.x      |  15PARENTI Hilton, E. J., & Lavoué, S. (2018). A review of the systematic biol- ogy of fossil and living bony- tongue fishes, Osteoglossomorpha (Actinopterygii: Teleostei). Neotropical Ichthyology, 16, e180031. https://doi.org/10.1590/1982-0224-20180031 Hoeksema, B. W. (2007). Delineation of the Indo-Malayan centre of max- imum marine biodiversity: The coral triangle. In Biogeography, time and place: Distributions, barriers and islands. Topics in Geobiology, 29, 117–178. https://doi.org/10.1007/978-1-4020-6374-9 Hoese, D. F., & Gill, A. C. (1993). Phylogenetic relationships of eleotr- idid fishes (Perciformes: Gobioidei). Bulletin of Marine Science, 52, 415–440. Humphries, C. J., & Parenti, L. R. (1999). Cladistic biogeography, second edition. Interpreting patterns of plant and animal distributions, 2nd ed. Oxford, UK: Oxford University Press. Johnson, G. D., Ida, H., Sakaue, J., Sado, T., Asahida, T., & Miya, M. (2012). A ‘living fossil’ eel (Anguilliformes: Protanguillidae, fam nov) from an undersea cave in Palau. Proceedings of the Royal Society B, 279, 934– 943. https://doi.org/10.1098/rspb.2011.1289 Kottelat, M. (1989). Zoogeography of the fishes from Indochinese inland waters with an annotated check- list. Bulletin of the Zoölogical Museum of the University of Amsterdam, 12, 1–55. Kumazawa, Y., & Nishida, M. (2000). Molecular phylogeny of osteoglos- soids: A new model for Gondwanian origin and plate tectonic trans- portation of the Asian Arowana. Molecular Biology and Evolution, 17, 1869–1878. https://doi.org/10.1093/oxfordjournals.molbev. a026288 Li, H., He, Y., Jiang, J., Liu, Z., & Li, C. (2018). Molecular systematics and phylogenetic analysis of the Asian endemic freshwater sleepers (Gobiiformes: Odontobutidae). Molecular Phylogenetics and Evolution, 121, 1–11. https://doi.org/10.1016/j.ympev.2017.12.026 Lin, Y.-S., Poh, Y.-P., & Tzeng, C.-S. (2001). A phylogeny of freshwater eels inferred from mitochondrial genes. Molecular Phylogenetics and Evolution, 20, 252–261. https://doi.org/10.1006/mpev.2001.0969 Lomolino, M. V., & Brown, J. H. (2009). The reticulating phylogeny of island biogeography 1459 theory. Quarterly Review of Biology, 84, 357–390. https://doi.org/10.1086/648123 Lourie, S. A. (2016). Seahorses. A life-size guide to every species. Chicago, IL: The University of Chicago Press. https://doi.org/10.7208/ chicago/9780226338552.001.0001 Lourie, S. A., & Vincent, A. C. J. (2004). A marine fish follows Wallace's Line: The phylogeography of the three- spot seahorse (Hippocampus trimaculatus, Syngnathidae, Teleostei) in south- east Asia. Journal of Biogeography, 31, 1975–1985. https://doi. org/10.1111/j.1365-2699.2004.01153.x Matzke, N. J. (2013). Probabilistic historical biogeography: New models for founder-event speciation, imperfect detection, and fossils allow improved accuracy and model-testing [PhD Thesis]. Berkeley, CA: University of California [not seen]. Metcalfe, I. (2011). Palaeozoic-Mesozoic history of SE Asia. In R. Hall, M. A. Cottam, & M. E. J. Wilson (Eds.), The SE Asian gateway: History and tectonics of the Australia-Asia collision (pp. 7–35). The Geological Society of London, Special publication 355, London. Metcalfe, I. (2013). Gondwana dispersion and Asian accretion: Tectonic and palaeogeographic evolution of eastern Tethys. Journal of Asian Earth Sciences, 66, 1–33. https://doi.org/10.1016/j. jseaes.2012.12.020 Michaux, B. (2010). Biogeology of Wallacea: Geotectonic mod- els, areas of endemism, and natural biogeographical units. Biological Journal of the Linnean Society, 101, 193–212. https://doi. org/10.1111/j.1095-8312.2010.01473.x Mooi, R. D. (2017). Evidence, pattern and assumptions. Reintroducing Rosen's empiricism and skepticism to systematics and biogeography. In B. I. Crother & L. R. Parenti (Eds.), Assumptions inhibiting progress in comparative biology (pp. 235–262). Boca Raton, FL: CRC Press/Taylor and Francis Group. Motomura, H., Alama, U. B., Muto, N., Babaran, R. P., & Ishikawa, S. (Eds.). (2017). Commercial and bycatch market fishes of Panay Island, Republic of the Philippines. The Kagoshima University Museum, Kagoshima, University of the Philippines Visayas, Iloilo, and Research Institute for Humanity and Nature, Kyoto. Nelson, J. S., Grande, T. C., & Wilson, M. V. H. (2016). Fishes of the world, 5th ed.. Hoboken, NJ: John Wiley & Sons Inc. https://doi. org/10.1002/9781119174844 Nelson, G., & Ladiges, P. Y. (2001). Gondwana, vicariance biogeography and the New York School revisited. Australian Journal of Systematic Botany, 49, 389–409. https://doi.org/10.1071/BT00025 Nelson, G., & Platnick, N. I. (1981). Systematics and biogeography: Cladistics and vicariance. New York, NY: Columbia University Press. Parenti, L. R. (2008). Life history patterns and biogeography: An in- terpretation of diadromy in fishes. Annals of the Missouri Botanical Garden, 95, 232–247. https://doi.org/10.3417/2006051 Parenti, L. R. (2014). A new species of Neostethus (Teleostei: Atherinomorpha: Phallostethidae) from Brunei Darussalam, with comments on northwestern Borneo as an area of endemism. Raffles Bulletin of Zoology, 62, 175–187. Parenti, L. R. (2017). Raising cain: On the assumptions that inhibit sci- entific progress in comparative biogeography. In B. I. Crother & L. R. Parenti (Eds.), Assumptions inhibiting progress in comparative biol- ogy (pp. 219–234). Boca Raton, FL: CRC Press/Taylor and Francis Group. Parenti, L. R., & Ebach, M. C. (2009). Comparative biogeography: Discovering and classifying biogeographical patterns of a dynamic earth. Berkeley, CA: University of California Press. Parenti, L. R., & Ebach, M. C. (2010). Wallacea deconstructed. In D. M. Williams & S. Knapp (Eds.), Beyond cladistics: The branching of a para- digm (pp. 303–318). Berkeley, CA: University of California Press. Parenti, L. R., & Ebach, M. C. (2013). Evidence and hypothesis in biogeog- raphy. Journal of Biogeography, 40, 813–820. https://doi.org/10.1111/ jbi.12069 Patterson, C. (1993). Chapter 36. Osteichthyes: Teleostei. In M. J. Brenton (Ed.), The fossil record (pp. 1–145). London, UK: Chapman & Hall. Ree, R. H., & Smith, S. A. (2008). Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Systematic Biology, 57(1), 4–14. https://doi. org/10.1080/10635150701883881 Roberts, T. R. (2012). Scleropages inscriptus, a new fish species from the Tananthayi or Tenasserim River basin, Malay Peninsula of Myanmar (Osteoglossidae: Osteoglossiformes). Aqua, 18(2), 113–118. Rosen, D. E. (1976). A vicariance model of Caribbean biogeography. Systematic Zoology, 24(1975), 431–464. Rosen, D. E., & Greenwood, P. H. (1976). A fourth neotropical species of synbranchid eel and the phylogeny and systematics of syn- branchiform fishes. Bulletin of the American Museum of Natural History, 157(1), 70 p. Rosen, D. E., & Parenti, L. R. (1981). Relationships of Oryzias, and the groups of atherinomorph fishes. American Museum Novitates, 2719, 25 p. Saruwatari, T., Oohara, I., & Kobayashi, T. (2002). Salangid fishes: Their past, present and future. Proceeding of the International Commemorative Symposium 70th Anniversary of Japanese Society of Fisheries Science, 68 (Suppl. I), 71–74. Schulten, S. (2018). Alexander von Humboldt: Master of infographics. In: Mapping the nation - a companion site to mapping the nation by Susan Schulten. Retrieved from http://www.mappingthenation.com/. Springer, V. G. (1982). Pacific plate biogeography, with special reference to shorefishes. Smithsonian Contributions to Zoology, 367, 1–182. https://doi.org/10.5479/si.00810282.367 Stout, C. C., Tan, M., Lemmon, A. R., Lemmon, E. M., & Armbruster, J. W. (2016). Resolving Cypriniformes relationships using an anchored 16  |     PARENTI enrichment approach. BMC Evolutionary Biology, 16, 244. https://doi. org/10.1186/s12862-016-0819-5 Teng, H.-Y., Lin, Y.-S., & Tzeng, C.-S. (2009). A new Anguilla species and a reanalysis of the phylogeny of freshwater eels. Zoological Studies, 48, 808–822. Teske, P. R., Cherry, M. I., & Matthee, C. A. (2004). The evolutionary his- tory of seahorses (Syngnathidae: Hippocampus): Molecular data sug- gest a West Pacific origin and two invasions of the Atlantic Ocean. Molecular Phylogenetics and Evolution, 30, 273–286. https://doi. org/10.1016/S1055-7903(03)00214-8 Thacker, C. E. (2015). Biogeography of goby lineages (Gobiiformes: Gobioidei): Origin, invasions and extinction throughout the Cenozoic. Journal of Biogeography, 42, 1615–1625. https://doi.org/10.1111/jbi.12545 Tsukamoto, K. (2009). Oceanic migration and spawning of anguil- lid eels. Journal of Fish Biology, 74, 1833–1852. https://doi. org/10.1111/j.1095-8649.2009.02242.x von Humboldt, A. (1850). Cosmos: A sketch of a physical description of the universe. E. C. Otté (translator). Volume 1. New York, NY: Harper & Brothers. Wallace, A. R. (1863). On the physical geography of the Malay Archipelago. Journal of the Royal Geographical Society, 33, 217–234. https://doi.org/10.2307/1798448 Wang, Y., Shen, Y., Feng, C., Zhao, K., Song, Z., Zhang, Y., … He, S. (2016). Mitogenomic perspectives on the origin of Tibetan loaches and their adaptation to high altitude. Scientific Reports, 6, 29690. https://doi. org/10.1038/srep29690 Wolfson, A. (1948). Bird migration and the concept of continental drift. Science, 108, 23–30. Woodland, D. J. (1986). In T. Uyeno, R. Arai, T. Taniuchi, & K. Matsuura (Eds.), Wallace's Line and the distribution of marine inshore fishes (pp. 453–460). Tokyo, Japan: Ichthyological Society of Japan. Zhu, M., Graham, S., & McHargue, T. (2009). The Red River Fault zone in the Yinggehai Basin, South China Sea. Tectonophysics, 476, 397–417. https://doi.org/10.1016/j.tecto.2009.06.015 BIOSKE TCH Lynne R. Parenti is a Curator of Fishes and Research Scientist at the National Museum of Natural History, Smithsonian Institution, Washington, DC. She studies the systematics, biogeography and reproductive biology of bony fishes, collection- based compara- tive biology, and the theory and methods of biogeography. How to cite this article: Parenti LR. The shape of biogeography: Endemism, maps, and classification of fish distributions in the western Pacific. J Biogeogr. 2019;00:1–16. https://doi.org/10.1111/jbi.13551