Biogeography of the social wasp genus Brachygastra (Hymenoptera: Vespidade: Polistinae).

Journal of Biogeography (J. Biogeogr.) (2014)

ORIGINAL ARTICLE

Biogeography of the social wasp genus Brachygastra (Hymenoptera: Vespidade: Polistinae) Marjorie da Silva* and Fernando Barbosa Noll

Depto Zoologia e Bot^anica, Instituto de Bioci^encias, Letras e Ci^encias Exatas, University Estadual Paulista, S~ao Jose do Rio Preto, SP, Brazil

ABSTRACT

Aim The aim of this study was to understand the biogeography of Brachygastra. As the spatial component of evolution is of fundamental importance to understanding the processes shaping the evolution of taxa, the known geological history of the Neotropical region was used together with the current phylogeny and distribution of species to investigate questions concerning the biogeography of Brachygastra: the ancestral ranges of Brachygastra species; their areal relationships and their congruence with previously published hypotheses; the possible associated vicariance events and the influence of land bridges between North and South America, and the split between the Amazon and Atlantic forests. Location Neotropical region, from Mexico to central Argentina and southern USA. Methods Statistical dispersal–vicariance analysis (S-DIVA) was used to reconstruct the possible ancestral ranges of Brachygastra species based on their phylogeny (divided into three groups, lecheguana, scuttelaris and smithii). A Brooks parsimony analysis (BPA) and component analysis were performed to reconstruct the areal relationships of these species within the Neotropics. Results S-DIVA suggested a widespread, South American ancestral region for Brachygastra. The ancestral B. azteca probably reached the Nearctic before a posterior vicariance event separated it from the species groups ((lecheguana (scutellaris + smithii))), that stayed in the Atlantic forest. The ancestral (scutellaris + smithii groups) possibly reached the Amazon by dispersal, and the subsequent vicariance event splitting the Atlantic forest and Amazon separated the groups into scutellaris in the Atlantic forest and smithii in the Amazon. BPA and component analyses suggested that the Nearctic was a sister area to other regions, the Andes and Mesoamerica was a sister area to the Neotropical regions and the Amazon was closely related to the Atlantic forest.

*Correspondence: Marjorie da Silva, Lab de Aculeata, Depto Zoologia e Bot^anica, Instituto de Bioci^encias, Letras e Ci^encias Exatas, UNESP, Crist ov~ao Colombo 2265, 15054–000 S~ao Jose do Rio Preto, SP, Brazil. E-mail: [email protected]

ª 2014 John Wiley & Sons Ltd

Main conclusions The phylogeny and distribution of Brachygastra suggest the influence of a land bridge between the Northern and Southern Hemispheres affecting the cladogenesis of B. azteca and the importance of the formation of the two blocks of forests in South America to the cladogenesis of the main groups of Brachygastra. Future comparisons between the distribution patterns of other taxa should enable a more precise identification of the possible events and outcomes, adding robustness to the hypothesized areal relationships. Keywords Brachygastra, cladistic biogeography, Epiponini, historical biogeography, Neotropical region, S-DIVA, vicariance.

http://wileyonlinelibrary.com/journal/jbi doi:10.1111/jbi.12417

1

M. da Silva and F. Barbosa Noll INTRODUCTION The Neotropical region, including Central and South America, is well known for its remarkable biodiversity (Morrone, 2013); it is perhaps the most species-rich terrestrial biogeographical region on Earth (Condamine et al., 2012). The heterogeneity of the abiotic conditions and the complex geological history are two factors that presumably have played a role in the patterns of species distribution and diversification over geological time (Amorim & Pires, 1996; Morrone, 2013). The spatial structure of Neotropical biodiversity has long been studied by evolutionary biologists, particularly those interested in understanding the processes that could explain the origin of this diversification (Sigrist & Carvalho, 2009). Although the age and geographical origin of many Neotropical groups are well established, less is known of their diversification patterns (Condamine et al., 2012). Similarly, the factors that have shaped such a high level of species richness over time have scarcely been investigated within a large-scale temporal framework, which is especially true for the processes that underlie the assembly and evolution of Amazonian biodiversity (Hoorn & Wesselingh, 2010; Viens et al., 2011). Furthermore, the temporal and geographical diversification of Neotropical insects remains poorly understood because of the complex changes in geological and climatic conditions that occurred during the Cenozoic (Condamine et al., 2012). Social wasps (Epiponini) are common representatives of the Neotropical fauna (Jeanne, 1991) and the species belonging to Brachygastra Perty, 1833 are widely distributed in this region (from southern USA to Argentina). This genus comprises 17 species broadly distributed in South America, absent only in Chile, Uruguay and central and south Argentina. The species are most commonly distributed in humid and forested environments (Fig. 1, and see Appendices S1 and S2 in the Supporting Information), with only B. augusti, B. mouleae, B. moebiana and B. lecheguana occurring in areas with open vegetation (B. lecheguana is found in drier areas, i.e. Caatinga). Brachygastra lecheguana and B. augusti are widespread in South America and Central America (Panama and Costa Rica). Two species, B. borellii and B. baccaraulea, are present in upland areas. With the exception of Trinidad and Tobago, there are no other records for the Antilles. Summarizing the distribution of Brachygastra according to the current phylogeny and the division of the genus into three main groups (Andena & Carpenter, 2012; see Appendix S3), the lecheguana group is found in the Atlantic rain forest and Nearctic region, while the scutellaris group occurs predominantly in the Atlantic forest and the smithii group is restricted to the Amazon rain forest. Brachygastra azteca, a sister species to all other species of the genus, occurs in Mexico. Biogeography provides a historical perspective for understanding both the biota and geological evolution of the planet (Ebach & Humphries, 2002). Biogeographical analyses based on the distribution of South American monophyletic groups can be used to corroborate or refute hypothesized areal 2

relationships and provide additional evidence to help understand the biogeography of other Neotropical taxa and the history of this region. From the current phylogeny and distribution of Brachygastra, two points are relevant to understanding the biogeography of this group: the presence of the most basal species, B. azteca, in North America, and the influence of the split of the two blocks of forests of South America on the cladogenesis of the scutellaris and smithii groups, with B. scutellaris presenting a disjunct distribution in these two blocks. This work aimed to investigate four major questions. (1) What are the ancestral ranges of Brachygastra species? (2) What areal relationships can be uncovered based on the known distributional and phylogenetic information? (3) Are the areal relationships consistent with previously published hypotheses? (4) Is it possible to identify the vicariance events associated with the ancestral range and areal relationships of the species’ distributions, in particular the influence of the land bridges between North and South America and the split between the Amazon and Atlantic forests? MATERIALS AND METHODS Data collection and map preparation A distributional dataset was produced based on a compilation of literature and collection records from museums (see Appendix S1). The literature records were taken from the revision of the genus, a checklist and the Global Biodiversity Information Facility (Appendix S1). The museum collections are listed in Appendix S1. If there were no geographical coordinates associated with the specimen records, the specimens were georeferenced online with the aid of gazetteers or Google Earth. In cases of missing or inaccurate locality data, the record was excluded from the analysis. The geographical coordinates were converted to decimal degrees and imported into Qgis 2.2.0 Valmiera (Nanni et al., 2012) to produce distribution maps. Based on the distribution maps (see Appendix S2) and Amorim & Pires (1996), we used the following areas for the biogeographical analyses: Nearctic (NEA), Andes and Mesoamerica (AnME), northern Amazonia (NAm), south-western Amazonia (SWAm), south-eastern Amazonia (SEAm) and Atlantic forest (AtFor) (Fig. 2). Biogeographical investigation To determine the possible scenarios of Brachygastra diversification, a statistical dispersal–vicariance analysis (S-DIVA) (Ronquist, 1997; Yu et al., 2010) was employed. To reconstruct the relationships among the different areas of the Neotropics, Brooks parsimony analysis (BPA) (Brooks, 1981, 1990; Wiley, 1988a,b) and a component analysis (Platnick & Nelson, 1978; Nelson & Platnick, 1981) were applied. The current phylogenetic hypothesis for Brachygastra (Andena & Carpenter, 2012) was used for these analyses, and the three main clades (lecheguana, scutellaris and Journal of Biogeography ª 2014 John Wiley & Sons Ltd

Biogeography of the social wasp genus Brachygastra

Figure 1 The biomes of the Neotropics (according to the World Wide Fund for Nature; http://www.worldwildlife.org/ publications/ terrestrial-ecoregions-of-the-world) used to study the distribution of the social wasp genus Brachygastra.

Figure 2 The biogeographical areas postulated by Amorim & Pires (1996) and the relationships among them, used to study the biogeography of the social wasp genus Brachygastra. AnME, Andes and Mesoamerica; NWAm, north-western Amazon; NAm, northern Amazon; SWAm, south-western Amazon (north-west component); SEAm, south-eastern Amazon; AtlFor, Atlantic forest (south-east component) (modified from Nihei & Carvalho, 2007).

smithii groups) proposed by Andena & Carpenter (2012) were used to facilitate interpretation of the biogeographical history of the group. Journal of Biogeography ª 2014 John Wiley & Sons Ltd

Based on the phylogenetic tree, possible ancestral ranges of Brachygastra were investigated using S-DIVA in rasp 2.0 (Yu et al., 2011). This method reconstructs ancestral ranges while 3

M. da Silva and F. Barbosa Noll also accounting for phylogenetic uncertainty and multiple solutions during DIVA optimization (Yu et al., 2010). The analysis was performed using the ‘maxareas = 4’ option in S-DIVA, which is equivalent to assuming that the ancestors of the group in question have the same ability to disperse as their extant descendants and, therefore, that the ancestral ranges were similar in size to the extant ranges (Sanmartın, 2003; Nylander et al., 2008). Data from Chartergus, a sister genus of Brachygastra, were used as an outgroup for S-DIVA analysis. The primary BPA involved constructing an individual area cladogram for the taxa by replacing the species names in the phylogeny with their endemic area. All of the internal and terminal nodes were numbered for later representation in data matrices that were analysed using a parsimony algorithm. Using this procedure, area cladograms were established to show the general pattern of areal relationships (Brooks et al., 2001). A hypothetical ancestral area with a total absence of species was added to the data matrices to allow grouping by presence rather than absence of taxa, which was employed to root the general area cladograms (Crisci et al., 2003). A parsimony analysis was performed using tnt 1.1 (tree analysis using new technology; Goloboff et al., 2000, 2008) with the traditional search algorithm, 500 replications, tree bisection–reconnection (TBR) branch swapping and holding 10 trees per replication. For the component analysis, taxon cladograms were converted into fully resolved area cladograms (if necessary using assumptions 0, 1 and 2) by searching for congruence in the patterns of relationships between areas and attempting to find a greater number of components (which are basically monophyletic groups of areas) in the area cladograms. There is still a notion of hierarchy underlying the components such that one component includes smaller components (Nihei, 2012). If a general area cladogram is not found through intersection, or if the intersection contains multiple cladograms, a consensus tree can be constructed (Morrone & Crisci, 1995). This analysis was performed using component 2.0 (Page, 1993) with heuristic searching with four criteria: minimize leaves added; minimize losses; minimize leaves added plus duplications; and minimize losses plus duplications. Although the use of only one group represented a shortcoming, we decided to perform the component analysis and BPA with just Brachygastra because the aim of the study was not to present a model for Neotropical species but to compare the areal relationships with previously established patterns for other Neotropical taxa, which has been performed by other authors (Pires & Marinoni, 2010; de Carvalho et al., 2013). RESULTS S-DIVA S-DIVA produced a scenario with four vicariance and 25 dispersion events that have shaped the current distribution patterns of Brachygastra (Fig. 3). This analysis indicated five possible ancestral areas for Brachygastra + Chartergus, all of 4

which were widespread in South America. The ancestral area of B. azteca and all other species included NEA and AtFor. A posterior vicariance event separated these areas, resulting in the origin of B. azteca in NEA and the permanence of the ancestor of the remaining species in AtFor. The ancestral area of the lecheguana group and the other groups was present in AtFor. After an apparent dispersal event, the ancestor of the scutellaris and smithii groups reached SWAm. With a subsequent vicariance event, the ancestral scutellaris group remained in AtFor while the ancestral smithii group remained in SWAm. A second connection/separation event between AtFor and SWAm occurred within the scutellaris group. The sister species B. augusti and B. mouleae remained in AtFor, while the ancestor of the clade comprising (B. fistulosa, (B. cooperi, (B. scutellaris and B. myersi))) dispersed to SWAm and occupied an ancestral area corresponding to AtFor and SWAm. Thereafter, a vicariance event separated these two areas, with B. fistulosa in AtFor and (B. cooperi, (B. scutellaris and B. myersi)) in SWAm. After the division of the ancestral areas of the scutellaris and smithii groups, the species within the smithii group remained in the Amazonian regions, with B. smithii reaching Central America secondarily. BPA and component analysis BPA resulted in a single hypothesis regarding the relationships of the Neotropical areas (Fig. 4a). The area cladogram suggested that NEA was a sister area to all the other regions, and AnME was a sister area to the Neotropical regions. The Amazonian regions appeared to be a clade that was closely related to AtFor. With the component analysis, the use of the minimizing losses option generated two area cladograms and was the only criterion that presented a high-resolution consensus relationship. The topology of the consensus suggested that AtFor was a sister area to SWAm and revealed a polytomy formed by AnME, NAm and SEAm (Fig. 4b). The use of other criteria resulted in a low-resolution consensus and suggested that NEA was a sister area to a polytomy that included all the other areas (Fig. 4c). DISCUSSION Ancestral ranges It is important to note certain aspects of this methodology. S-DIVA is unable to distinguish between contiguous and non-contiguous areas, so some ancestral areas may not be apparent. Similarly, the presence of a species in adjacent areas is probably the result of an expansion of the population in periods that had gaps in or non-existent vicariant barriers as a result of dispersal events. The vicariance event separating NEA and AtFor appears to be an incongruent result because these two areas are widely separated and have never been in contact. The ancestor of B. azteca probably Journal of Biogeography ª 2014 John Wiley & Sons Ltd

Biogeography of the social wasp genus Brachygastra Areal relationship The areal relationships that we found are incongruent with those found by Amorim & Pires (1996). Differing results for the same areas have been found in other studies as well (Nihei & Carvalho, 2007; Sigrist & Carvalho, 2009; Pires & Marinoni, 2010; Morrone, 2013). Although Amorim & Pires (1996) considered Amazonia to consist of two non-related areas (north-western Amazonia and SEAm), they formed a single clade in the present study, with SEAm being closely related to NAm and SWAm. Despite the use of just one genus, the areal relationship we found is congruent with Sigrist & Carvalho (2009), who used a total of 114 species from 12 unrelated genera [Araneae (2 genera), Coleoptera (2), Squamata (1), Diptera (3), Hemiptera (2), Heteroptera (1) and Lepidoptera (1)] and discovered a basal split between the Amazonian and Atlantic forest regions, suggesting an ancient division between these two areas. The Amazon and Atlantic forests have been separated by open vegetation formations since the Miocene (BatalhaFilho et al., 2013; Nascimento et al., 2013) and have been completely isolated since at least the Pliocene (DaSilva & Pinto da Rocha, 2012). Sigrist & Carvalho (2009) also found that the Amazon areas formed a clade with AnME, which is congruent with our results from the component analysis. Figure 3 Statistical dispersal–vicariance analysis (S-DIVA) graphical representation of the ancestral distribution of the phylogenetic hypothesis proposed for Brachygastra in the Neotropics by Andena & Carpenter (2012). The numbers correspond to the nodes of the cladogram, and coloured circles correspond to the hypothetical ancestral area. A circle with more than one colour indicates more than one possibility with the same probability. The letters before the names of the terminal taxa represent the current distribution. Biogeographical regions: A, Nearctic; B, Andes and Mesoamerica; C, northern Amazon; D, south-western Amazon; E, south-eastern Amazon; F, Atlantic forest.

reached North America through a landmass that linked South and North America before the formation of the Isthmus of Panama (see below). According to Batalha-Filho et al. (2013), a connection between the Amazon forest and AtFor appears to have existed during much of the Miocene (approximately 15– 5 Ma) and is congruent with a vicariance event between these two biomes leading to the cladogenesis of the scutellaris and smithii groups. The same authors also suggested a more recent connection between these regions, but it is not congruent with the separation of (B. augusti + B. mouleae)/ (B. fistulosa, (B. cooperi, (B. scutellaris + B. myersi))) within the scutellaris group. However, the distribution of B. lecheguana indicates an expansion of this species that is congruent with the younger pathway from the northern AtFor to the lower Amazon in the northern Para and eastern Amazonian states of Brazil. The emergence of the Isthmus of Panama enabled the expansion of B. augusti and B. smithii to Panama in Central America. Journal of Biogeography ª 2014 John Wiley & Sons Ltd

Origin and diversification of the Brachygastra Few social wasps are known from the fossil record. The absence of Vespidae from fossil resins, in which bees and ants are relatively abundant, is probably a result of their relatively large size, which reduces the likelihood of their entrapment in the sticky tree resin (Burnham, 1978). Furthermore, behavioural characteristics and paper nest structures do not lend themselves to fossilization, which explains their scarcity in sedimentary deposits (Spradbery, 1973). In a revision of the social insects in the fossil record, Burnham (1978) stated that no taxa prior to the Oligocene could be definitively assigned to the sister group Vespinae and Polistinae (Pickett & Carpenter, 2010). However, two wasp fossils from the late Eocene (38–33.9 Ma) have been found recently in the Monteils Formation in the south of France (Perrard et al., 2014). These wasps have been assigned to a new genus of Polistinae wasps, Palaepolistes jattioti, which is probably closely related to the tribe Ropalidiini, but its relationship with other Polistinae tribes remains uncertain (Perrard et al., 2014). Furthermore, a fossil nest of Vespinae dated to the Late Cretaceous was described by Wenzel (1990), and eusociality probably evolved in these wasps around the time eusociality arose in ants, in the mid-Cretaceous (Grimaldi & Engel, 2005). Carpenter & Grimaldi (1997) described Agelaia electra, a representative fossil of the Epiponini, which is also the tribe to which Brachygastra belongs. That species was found in Dominican amber (from the early Miocene to the late Oligocene, 16–25 Ma) and it is the first Antillean representative of 5

M. da Silva and F. Barbosa Noll

(a)

(b)

(c)

Figure 4 Areal relationship analysis of the distribution of Brachygastra in the Neotropics. (a) A cladogram generated via Brooks parsimony analysis (length, 40; consistency index, 87; retention index, 82). (b) A consensus of area cladograms generated via component analysis using the minimizing losses criterion. (c) A consensus cladogram obtained via component analysis using three different criteria: minimizing leaves added; losses plus duplications; leaves added plus duplications. NEA, Nearctic; AnMe, Andes and Mesoamerica; AtFor, Atlantic forest; NAm, northern Amazon; SEAm, south-eastern Amazon; SWAm, south-western Amazon.

this tribe of paper wasps. Although Agelaia is a basal genus in the Epiponini, the fossil suggests that this tribe was present on Greater Antilles and became extinct (Carpenter & Grimaldi, 1997). Possible explanations for this finding include the former connection between the North American and South American landmasses via the drift of the protoAntilleans from the Early Jurassic to the Early Cretaceous (Morrone, 2013) and the extensive land bridges that connected many islands at various times, particularly in the Oligocene (Carpenter & Grimaldi, 1997). Gutierrez-Garcıa & Vazquez-Domınguez (2013) recently proposed that species exchange between North and South America and Central America included three separate events. First, an initial migration during the Late Cretaceous–early Palaeocene. Second, dispersal along a terrestrial corridor preceding the formation of the Isthmus of Panama, an event defined by a high differentiation between the highlands and lowlands that are associated with intense volcanic activity. Third, a migration involving a major dispersion through the Isthmus of Panama. The origination and diversification of Brachygastra probably began around 32 Ma. At that time, the Amazonian forests extended north to the Caribbean coast of South America, where a diverse rain forest existed (Hoorn & Wesselingh, 2010) and a land bridge (the Greater Antilles and Aves Ridge land bridge, GAARlandia; Iturralde-Vinent, 2006) possibly connected the Nearctic and Neotropics (Hoorn & Wesselingh, 2010). The GAARlandia was probably not a complete land bridge, as it probably had water gaps. Moreover, how well it functioned as a land bridge remains uncertain. However, although GAARlandia is a hypothesized connection albeit with an uncertain degree of continuity, Iturralde-Vinent (2006) speculated that it permitted the overland dispersal of various elements of the land biota from northern South America, and Condamine et al. (2012) suggested that some butterflies used this land bridge to reach North and South America. The Andes orogenesis had begun, and the Pozo embayment separated the northern and central Andes. These land bridges preceded the formation of the Isthmus of Panama in the Oligocene and were probably the route by which the ancestors of B. azteca reached Mexico. This 6

species, a sister species to the others, occupied a central region of Mexico located between the Sierra Madre Occidental and Oriental ranges. According to Morrone (2006), many insect species with origins in the Neotropics dispersed from South America, diversifying in the area of the Mexican plateau during the Eocene–Pliocene (c. 35–2 Ma) and disappearing from other areas. Potential causes of these extinction events include intense volcanism in the southern areas caused by subduction among the plates that form Central America, the high elevations of the Sierra Madre mountain chains that were formed via tectonism during the Cretaceous, volcanism in the Tertiary, Oligocene and Miocene, and significant elevation of the crests in the middle Pleistocene that resulted in the formation of the modern standard drainage (Barbou, 1973). These events are congruent with the Montane Mesoamerican cenocron (sets of taxa that share the same biogeographical history) of the Mexican transition zone, which have South American affinities and are hypothesized to have diversified in this area in the Oligocene (Morrone, 2010). The ancestor of the lecheguana group probably reached North America by way of Central America through the western part of that continent and remained in its central and southern portions. From 23 to 10 Ma, there was an intensification of the Andean uplift and subsequent changes in the Amazonian landscape, with the formation of the Pebas system (Hoorn & Wesselingh, 2010). The Pebas mega-wetland extended from the northern subAndean basins eastwards and occupied part of western Amazonia during the early Miocene and most of the region during the middle Miocene. At its maximum, this vast network of lakes and swamps covered an area of more than 1 million km2 (Hoorn & Wesselingh, 2010). Because of this barrier, the ancestor of B. lecheguana probably inhabited areas further to the south and centre of South America. The GAARlandia land bridge disappeared from 32 to 23 Ma. The degree of connectivity between North and South America peaked between 35 and 32 Ma and was subsequently reduced as many Caribbean lowlands were inundated, and the terrestrial organisms that colonized GAARlandia in the Eocene–Oligocene transition became isoJournal of Biogeography ª 2014 John Wiley & Sons Ltd

Biogeography of the social wasp genus Brachygastra lated from the continent. Elements of this biota with a clear South American signature have been found in Puerto Rico, Hispaniola and Cuba, and in younger deposits of many Antillean islands (Iturralde-Vinent, 2006); this event probably separated the ancestral areas of B. lecheguana + B. mellifica, which are currently found in South America and North America, respectively. Brachygastra lecheguana is widespread in South America (from Panama to the central part of Argentina) and because it also occurs west of the Andean chain in Ecuador, its expansion probably occurred before the uplift of the Northern Andes (23–10 Ma) (Hoorn et al., 2010). A posterior migration is unlikely because the mountains in this region reach 5000 m. Brachygastra borelli, a sister species of B. lecheguana and B. mellifica, is currently restricted to a range of the Andes in southern Bolivia and northern Argentina, but it probably occupied a larger area in South America that has been shaped over time, mainly during the Miocene, which witnessed extensive marine transgressions in the basins of the lowland south (Lundberg et al., 1998). The Paranense marine transgressions (c. 15–13 Ma) resulted from a 150-m increase in sea level (Haq et al., 1987). Sediments assigned to this event are limited to a latitude 17 °S north-west of Santa Cruz, Bolivia (Lundberg et al., 1998), but have been identified as far south as 26 °S (Hernandez et al., 2005). However, between 22 °S and 26 °S there was a marine gap that resulted from a topographical barrier in the region (Hernandez et al., 2005), which is congruent with the current distribution of B. borelli. The posterior retraction of sea level allowed this species to expand its area to the east, and the isolation that resulted from an increase in sea level is also congruent with the occurrence of this species in the highlands. The cladogenesis between the scutellaris and smithii groups is congruent with a vicariance event separating the two forest areas of SWAm and AtFor, which appears to have been influenced by the aridization of South America. The gradual uplift of the Andes that began in the late Oligocene and the higher elevation period during the Pliocene decreased the effect of winds arriving from the Pacific and resulted in a large arid area in South America. Around the middle Miocene, subtropical environments were displaced northwards and a broad open vegetation area developed (Pascual & Ortiz Jaureguizar, 1990). Morrone & Coscar on (1998) suggested that this was the vicariance event that separated what had once been a continuous forest area into two blocks. The presence of B. cooperi, B. myersi and B. scutellaris in NAm and SWAm reinforces the idea that the Amazon region was formed by different unrelated components. In a study of birds, Batalha-Filho et al. (2013) identified species that originated within AtFor and colonized the Amazon only during the Miocene. Brachygastra scutellaris still occurred in the two forest blocks and probably became extinct in the open vegetation areas. Species of the smithii group are almost exclusively Amazonian; none of them occurs in AtFor. A similar pattern was found for the Neotropical lianas of Bignonieae. Journal of Biogeography ª 2014 John Wiley & Sons Ltd

In this tribe, the earliest divergence occurred about 50 Ma in eastern South American, in the same geographical region that is currently occupied by the Atlantic tropical forests of Brazil, followed by the colonization and diversification of lowlands in Amazonia. Reversals involving both eastern South America and Amazonia have also been observed (Lohmann et al., 2013; Alcantara et al., 2014). Since the formation of the South American dry vegetation diagonal in the Late Cretaceous to middle Oligocene (Zanella, 2012), contact between the Atlantic and Amazon forests through this open area has occurred numerous times, but the spatio-temporal dynamics of this contact remain to be investigated. In a study using molecular data, Batalha-Filho et al. (2013) found three putative pathways for birds connecting the Atlantic forest and Amazon (two from the middle to late Miocene and one from the Pliocene to Pleistocene); this is a pattern that is congruent with other groups (Oliveira et al., 1999; Costa, 2003; Auler et al., 2004; Wang et al., 2004). Nevertheless, the distributions of species of the scutellaris and smithii groups are strongly congruent with the northwestern and south-western components of the Neotropics. However, if this is assumed to be a vicariance event between these groups, a much older origin for the genus would be implied because it would have occurred during the Late Cretaceous (Morrone, 2013). The presence of B. augusti and B. smithii in Central America is secondary. There is evidence that the current biota of Neotropical South America expanded to the north as far as Central America and Mexico and to the south reaching Patagonia during pre-Quaternary times (Morrone, 2004). Both species originated in South America and reached Central America when the formation of the Panama Isthmus was complete, enabling significant faunal exchange. CONCLUSIONS The present work has addressed historical aspects of Brachygastra and represents an initial effort towards achieving a better understanding of the biogeography of the Epiponini. Additionally, it is the first work to examine the biogeography of this tribe and one of a limited number of studies concerning social insects (Camargo & Moure, 1996; Grimaldi & Agosti, 2000; Camargo & Pedro, 2003; Solomon et al., 2008). Regarding the ancestral ranges of the genus and the possible vicarinace events involved, our data indicate that the ancestor of Chartergus + Brachygastra probably inhabited northern South America and areas of a ‘proto’ Central America, reaching North America and, posteriorly, central and southern South America when the Pebas mega-wetland, which was caused by the intensification of the uplift of Andes, covered the Amazonian landscape. The phylogeny and distribution clearly show the importance of the land bridges connecting the Northern and Southern Hemispheres to the cladogenesis of the basal species B. azteca and B. mellifica. 7

M. da Silva and F. Barbosa Noll The emergence of the Isthmus of Panama also enabled the expansion of species to Central America. The influence of the formation of open vegetation areas in South America dividing the two blocks of forest is evident in the distribution of species of the scutellaris and smithii groups and appears to have influenced their cladogenesis. However, a division of South America into north-western and southeastern components resulting from an older vicariance event cannot be discounted because the time of vicariance cannot be determined using these data. However, this would be a strongly incongruent result if the distribution and phylogeny of Brachygastra are considered as a whole. An origin in the Neotropics and an absence of species in southern South America add additional evidence to the hypothesis that there is a Gondwanan connection in the Epiponini and that this tribe is closely related to groups from tropical Africa (Carpenter, 1993). This is also consistent with the phylogenetic hypothesis that postulates Epiponini as a sister tribe to Ropalidiini (Wenzel & Carpenter, 1994) because Ropalidiini occurs in the Afrotropical, Oriental and Australotropical regions. The areal relationship showed the AtFor and Amazonia regions to be closely related and, even though the current study was based on a single taxon, the results of the comparison are consistent with the literature (Sigrist & Carvalho, 2009). In future studies, comparisons between the distribution patterns of other taxa, including other social wasps, and sister-group relationships at higher taxonomic levels, may enable a more precise identification of possible events and the patterns they have generated. Furthermore, comparisons with other taxa will provide greater robustness to hypotheses regarding areal relationships. Even though the current study was of only one taxon, the comparison suggests results that are consistent with the literature (Sigrist & Carvalho, 2009). ACKNOWLEDGEMENTS The authors would like to thank Sergio Andena and James Carpenter for providing data on the Brachygastra phylogeny, and would also like to thank all the curators of the museums that provided material for analysis. This research was supported by the Fundacß~ao de Amparo a Pesquisa do Estado de S~ao Paulo–FAPESP (grant numbers 2009/12997-4 and 2011/ 06058-5).

REFERENCES Alcantara, S., Ree, R.H., Martins, F.R. & Lohmann, L.G. (2014) The effect of phylogeny, environment and morphology on communities of a lianescent clade (Bignonieae–Bignoniaceae) in Neotropical biomes. PLoS ONE, 9, e90177. Amorim, D.S. & Pires, M.R.S. (1996) Neotropical biogeography and a method for maximum biodiversity estimation. Biodiversity in Brazil, a first approach (ed. by C.E.M. Bicudo and N.A. Menezes), pp. 183–219. CNPq, S~ao Paulo, Brazil. 8

Andena, S.R. & Carpenter, J.M. (2012) A phylogenetic analysis of the social wasp genus Brachygastra Perty, 1833, and description of a new species (Hymenoptera: Vespidae: Epiponini). American Museum Novitates, 3753, 1–38. Auler, A.S., Wang, A., Edwards, R.L., Cheng, H., Cristalli, P.S., Smart, M.L. & Richards, D.A. (2004) Quaternary ecological and geomorphic changes associated with rainfall events in presently semi-arid northeastern Brazil. Journal of Quaternary Science, 19, 693–701. Barbou, C.D.A. (1973) Biogeographical history of Chirostoma (Pisces: Atherinidae): a species flock from the Mexican plateau. Copeia, 3, 533–556. Batalha-Filho, H., Fjelds a, J., Fabre, P.H. & Miyaki, C.Y. (2013) Connections between the Atlantic and the Amazonian forest avifaunas represent distinct historical events. Journal of Ornithology, 154, 41–50. Brooks, D.R. (1981) Hennig’s parasitological method: a proposed solution. Systematica Zoology, 30, 229–249. Brooks, D.R. (1990) Parsimony analysis in historical biogeography and coevolution: methodological and theoretical update. Systematic Zoology, 39, 14–30. Brooks, D.R., Vanveller, M. & McLennan, D.A. (2001) How to do BPA, really. Journal of Biogeography, 28, 343–358. Burnham, L. (1978) Survey of social insects in the fossil record. Psyche, 85, 85–133. Camargo, J.M.F. & Moure, J.S. (1996) Meliponini Neotropicais: o g^enero Geotrigona Moure, 1943 (Apinae, Apidae, Hymenoptera), com especial refer^encia a filogenia e biogeografia. Arquivos de Zoologia, 33, 95–161. Camargo, J.M.F. & Pedro, S.E.M. (2003) Meliponini neotropicais: o g^enero Partamona Schwarz, 1939 (Hymenoptera, Apidae, Apinae): bionomia e biogeografia. Revista Brasileira de Entomologia, 47, 311–372. Carpenter, J.M. (1993) Biogeographic patterns in the Vespidae (Hymenoptera): two views of Africa and South America. Biological relationships between Africa and South America (ed. by P. Goldblatt), pp. 138–155. Yale University Press, New Haven and London. Carpenter, J.M. & Grimaldi, D.A. (1997) Social wasps in amber. American Museum Novitates, 2303, 2–7. de Carvalho, A.L.G., de Britto, M.R. & Fernandes, D.S. (2013) Biogeography of the lizard genus Tropidurus Wied-Neuwied, 1825 (Squamata: Tropiduridae): distribution, endemism, and area relationships in South America. PLoS ONE, 8, 1–14. Condamine, F.L., Silva-Brand~ao, K.L., Kergoat, G.J. & Sperling, F.A.H. (2012) Biogeographic and diversification patterns of Neotropical Troidini butterflies (Papilionidae) support a museum model of diversity dynamics for Amazonia. BMC Evolutionary Biology, 12, 2–16. Costa, L.P. (2003) The historical bridge between the Amazon and the Atlantic Forest of Brazil: a study of molecular phylogeography with small mammals. Journal of Biogeography, 30, 71–86. Crisci, J.V., Katinas, L. & Posadas, P. (2003) Historical biogeography: an introduction. Harvard University Press, Cambridge, MA. Journal of Biogeography ª 2014 John Wiley & Sons Ltd

Biogeography of the social wasp genus Brachygastra DaSilva, M.B. & Pinto da Rocha, R. (2012) Hist oria biogeografica da Mata Atl^antica: opili~ oes (Arachnida) como modelo para sua infer^encia. Biogeografia da America do Sul: padr~oes e processos (ed. by C.J.B. Carvalho and E.A.B. Almeida), pp. 99–122. Editora Roca, S~ao Paulo, Brazil. Ebach, M.C. & Humphries, C.J. (2002) Cladistic biogeography and the art of discovery. Journal of Biogeography, 29, 427–444. Goloboff, P.A., Farris, J.S. & Nixon, K.C. (2000) TNT (tree analysis using new technology) (BETA). Published by the authors, Tucuman, Argentina. Goloboff, P.A., Farris, J.S. & Nixon, K.C. (2008) TNT (tree analysis using new technology), a free program for phylogenetic analysis. Cladistics, 24, 1–13. Grimaldi, D. & Agosti, D.A. (2000) Formicine in New Jersey Cretaceous amber (Hymenoptera: Formicidae) and early evolution of the ants. Proceedings of the National Academy of Sciences USA, 97, 13678–13683. Grimaldi, D.A. & Engel, M.S. (2005) Evolution of the insects. Cambridge University Press, New York. Gutierrez-Garcıa, T.A. & Vazquez-Domınguez, E. (2013) Consensus between genes and stones in the biogeographic and evolutionary history of Central America. Quaternary Research, 79, 311–324. Haq, U.B., Handerbol, J. & Vail, P.R. (1987) Chronology of fluctuating sea levels since the Triassic. Science, 235, 1156–1167. Hernandez, R.M., Jordam, T.B., Dalenz Farjat, A., Echavarrıa, L., Idleman, B.D. & Reynolds, J.H. (2005) Age, distribution, tectonics, and eustatic controls of Paranense and Caribbean marine transgressions in southern Bolivia and Argentina. Journal of South American Earth Sciences, 19, 495–512. Hoorn, C.M. & Wesselingh, F.P. (2010) Amazonia, landscape and species evolution: a look into the past, 1st edn. WileyBlackwell, London. Hoorn, C., Wesselingh, F.P., ter Steege, H., Bermudez, M.A., Mora, A., Stevink, J., Sanmartın, I., Sanchez-Meseguer, A., Anderson, C.L., Figueiredo, J.P., Jaramillo, C., Riff, D., Negri, F.R., Hooghiemstra, H., Lundberg, J., Stadler, T., S€arkinen, T. & Antonelli, A. (2010) Amazonia through time: Andean uplift, climate change, landscape evolution, and biodiversity. Science, 330, 927–931. Iturralde-Vinent, M.A. (2006) Meso-Cenozoic Caribbean paleogeography: implications for the historical biogeography of the region. International Geology Review, 48, 791–827. Jeanne, R.L. (1991) The swarm-founding Polistinae. The social biology of wasps (ed. by K.G. Ross and R.W. Matthews), pp. 191–231. Cornell University Press, Ithaca, NY. Lohmann, L.G., Bell, C.D., Cali o, M.F. & Winkworth, R.C. (2013) Pattern and timing of biogeographical history in the Neotropical tribe Bignonieae (Bignoniaceae). Botanical Journal of the Linnean Society, 171, 154–170. Lundberg, J.G., Marshall, L.G., Guerrero, J., Horton, B., Malabarba, M.C.S.L. & Wesselingh, F. (1998) The stage for neotropical fish diversification: a history of tropical South American rivers. Phylogeny and classification of neotropical Journal of Biogeography ª 2014 John Wiley & Sons Ltd

fishes (ed. by L.R. Malabarba, R.E. Reis, R.P. Vari, Z.M. Lucena and C.A.S. Lucena), pp. 603. Edipucrs, Porto Alegre, Brazil. Morrone, J.J. (2004) Biogeografıa, componentes bi oticos y zonas de transici on. Revista Brasileira de Entomologia, 48, 149–162. Morrone, J.J. (2006) Biogeographic areas and transitions zones of Latin America and the Caribbean islands based on panbiogeographic and cladistic analyses of the entomofauna. Annual Review of Entomology, 51, 467–494. Morrone, J.J. (2010) Fundamental biogeographic patterns across the Mexican Transition Zone: an evolutionary approach. Ecography, 33, 355–361. Morrone, J.J. (2013) Cladistics biogeography of the Neotropical region: identifying the main events in the diversification of the terrestrial biota. Cladistics, 29, 1–13. Morrone, J.J. & Coscar on, M. del C. (1998) Cladistics and biogeography of the assassin bug genus Rasahus Amyot & Serville (Heteroptera: Reduviidae: Peiratinae). Zoologische Mededelingen, 72, 73–87. Morrone, J.J. & Crisci, J.V. (1995) Historical biogeography: introduction to methods. Annual Review of Ecology, 23, 373–401. Nanni, A.S., Descovi Filho, L., Virtuoso, M.A., Montenegro, D., Willrich, G., Machado, P.H., Sperb, R., Dantas, G.S. & Calazans, Y. (2012) Quantum GIS: guia do usuario, vers~ao 1.7.4 ‘Wroclaw’. Available at: http://qgis.org/downloads/ manual/guia_do_usuario_174_pt_br.pdf. Nascimento, F.F., Lazar, A., Menezes, A.N., Durans, A.M., Moreira, J.C., Salazar-Bravo, J., D’Andrea, P.S. & Bonvicino, C.R. (2013) The role of historical barriers in the diversification processes in open vegetation formations during the Miocene/Pliocene using an ancient rodent lineage as a model. PLoS ONE, 8, e61924. Nelson, G. & Platnick, N.I. (1981) Systematics and biogeography: cladistics and vicariance. Colombia University Press, New York. Nihei, S.S. (2012) Biogeografia cladıstica. Biogeografia da America do Sul: padr~oes e processos (ed. by C.J.B. Carvalho and E.A.B. Almeida), pp. 99–122. Editora Roca, S~ao Paulo, Brazil. Nihei, S.S. & Carvalho, C.J.B. (2007) Systematics and biogeography of Polietina Schnabl & Dziedzicki (Diptera, Muscidae): neotropical area relationships and Amazonia as a composite area. Systematic Entomology, 32, 477–501. Nylander, J.A.A., Olsson, U., Alstrom, P. & Sanmartın, I. (2008) Accounting for phylogenetic uncertainty in biogeography: a Bayesian approach to dispersal–vicariance analysis of the thrushes (Aves: Turdus). Systematic Biology, 57, 257–268. Oliveira, P.E., Barreto, A.M.F. & Suguio, K. (1999) Late Pleistocene/Holocene climatic and vegetational history of the Brazilian caatinga: the fossil dunes of the middle S~ao Francisco River. Palaeogeography, Palaeoclimatology, Palaeoecology, 152, 319–337. Page, R.D.M. (1993) Component, version 2.0. The Natural History Museum, London. 9

M. da Silva and F. Barbosa Noll Pascual, R. & Ortiz Jaureguizar, E. (1990) Evolving climates and mammal faunas in Cenozoic South America. Journal of Human Evolution, 19, 23–60. Perrard, A., Nel, A. & Carpenter, J.M. (2014) A new paper wasp from Late Eocene of France (Hymenoptera: Vespidae: Polistinae). Zootaxa, 3753, 542–548. Pickett, K.M. & Carpenter, J.M. (2010) Simultaneous analysis and the origin of eusociality in the Vespidae (Insecta: Hymenoptera). Arthropod Systematics and Phylogeny, 68, 3–33. Pires, A.C. & Marinoni, L. (2010) Historical relationships among Neotropical endemic areas based on Sepedonea (Diptera: Sciomyzidae) phylogenetic and distribution data. Zoologia, 27, 681–690. Platnick, N.I. & Nelson, G. (1978) A method of analysis for historical biogeography. Systematic Zoology, 27, 1–16. Ronquist, F. (1997) Dispersal–vicariance analysis: a new approach to the quantification of historical biogeography. Systematic Biology, 46, 195–203. Sanmartın, I. (2003) Dispersal vs. vicariance in the Mediterranean: historical biogeography of the Palearctic Pachydeminae (Coleoptera, Scarabaeoidea). Journal of Biogeography, 30, 1883–1897. Sigrist, M.S. & Carvalho, C.J.B. (2009) Historical relationships among areas of endemism in the tropical South America using Brooks parsimony analysis (BPA). Biota Neotropica, 9, 79–91. Solomon, S.E., Bacci, M., Jr, Martins, J., Jr, Vinha, G.G. & Mueller, U.G. (2008) Paleodistribution and comparative molecular phylogeography of leafcutter ants (Atta spp.) provide new insight into the origins of Amazonian diversity. PLoS ONE, 3, e2738. Spradbery, J.P. (1973) Wasps: an account of biology and natural history of social and solitary wasps. University of Washington Press, Seattle. Viens, J.J., Pyron, R.A. & Moen, D.C. (2011) Phylogenetic origins of local-scale diversity patterns and causes of Amazonian megadiversity. Ecology Letters, 14, 643–652. Wang, X.F., Auler, A.S., Edwards, R.L., Cheng, H., Cristalli, P.S., Smart, P.L., Richards, D.A. & Shen, C.C. (2004) Wet periods in northeastern Brazil over the past 210 kyr linked to distant climate anomalies. Nature, 432, 740–743. Wenzel, J.W. (1990) A social wasp’s nest from the Cretaceous period, Utah, USA, and its biogeographical significance. Psyche, 97, 21–19.

10

Wenzel, J.W. & Carpenter, J.M. (1994) Comparing methods: adaptive traits and tests of adaptation. Phylogenetics and Ecology (ed. by P. Eggleton and R. Vane-Wright), pp. 79– 101. Academic Press, London. Wiley, E.O. (1988a) Parsimony analysis and vicariance biogeography. Systematic Zoology, 37, 271–290. Wiley, E.O. (1988b) Vicariance biogeography. Annual Review of Ecology, Evolution, and Systematics, 19, 513–542. Yu, Y., Harris, A.J. & He, X.J. (2010) S-DIVA (Statistical Dispersal-Vicariance Analysis): a tool for inferring biogeographic histories. Molecular Phylogenetics and Evolution, 56, 848–850. Yu, Y., Harris, A.J. & He, X.J. (2011) RASP (Reconstruct Ancestral State in Phylogenies) 2.0b. Sichuan University, Chengdu, China. Zanella, F.C.V. (2012) Evolucß~ao da biota da diagonal de for~es abertas secas da America do Sul. Biogeografia da macßo America do Sul: padr~oes e processos (ed. by C.J.B. Carvalho and E.A.B. Almeida), pp. 198–220. Editora Roca, S~ao Paulo, Brazil. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Appendix S1 Current distribution of Brachygastra species. Appendix S2 Maps of the current distribution of Brachygastra species. Appendix S3 Current phylogeny of Brachygastra species. BIOSKETCHES Marjorie da Silva is broadly interested in Neotropical social wasps and primarily interested in the systematics, biogeography and phylogeography of the Epiponini. This study is a result of her thesis from the Aculeata Laboratory at the Universidade Estadual Paulista. Fernando Barbosa Noll has an interest in the cladistics and behaviour of Neotropical swarm-founding wasps and was the adviser to Marjorie da Silva.

Editor: Liliana Katinas

Journal of Biogeography ª 2014 John Wiley & Sons Ltd