Multiple origins of the Phaenonotum beetles in the Greater Antilles (Coleoptera: Hydrophilidae): phylogeny, biogeography and systematics

Multiple origins of the Phaenonotum beetles in the Greater Antilles (Coleoptera: Hydrophilidae):... Abstract The systematics and the phylogenetic position of the Caribbean representatives of Phaenonotum Sharp (Coleoptera: Hydrophilidae) are investigated to understand the composition of the Caribbean fauna and its origin. Phylogenetic analysis based on mitochondrial and nuclear genes has revealed the Caribbean species to be situated in three deeply nested clades, inferring multiple colonization of Caribbean islands from the continent. Time-tree analysis and BioGeoBEARS analyses of ancestral ranges estimated the oldest clade, consisting of wingless single-island endemics of Cuba (P. delgadoi), Jamaica (P. ondreji sp. nov.) and Hispaniola (P. laterale sp. nov.), to have diverged c. 46.6 Mya from the South American ancestor and subsequently colonizing the Caribbean most likely via the GAARlandia land bridge connecting South America with the Greater Antilles. The remaining three Caribbean species, including the Puerto Rican endemic, P. borinquenum sp. nov., are of more recent (Miocene to Pliocene) origin and colonized the Greater Antilles by over-water dispersal. All the Caribbean species are illustrated and diagnosed, and three new species are described. The genus Phaenonotum, excluding P. caribense Archangelsky, is confirmed as a monophylum. We demonstrate that species-level taxonomy of Phaenonotum is difficult to solve by morphology alone and ideally requires the combination of morphology and molecular markers. ancestral range analysis, Caribbean biogeography, endemics, GAARlandia, molecular dating, new species, phylogeny, systematics, taxon cycle INTRODUCTION The Greater Antilles (i.e. islands of Cuba, Hispaniola, Jamaica and Puerto Rico in the Caribbean Region) are known for their high species diversity and endemism and are considered one of the world biodiversity hotspots (Mittermeier et al., 2005). The complex geological history of the region, characterized by emergence and sinking of particular islands and changes in their interconnections, has played a crucial role in generating these diverse and endemic faunas. The Antilles first emerged as the Antillean volcanic arc system in the Cretaceous in the gap between North and South America and repeatedly emerged and submerged while moving eastward into the Proto-Caribbean Sea later during the Cenozoic. The uplift of the core Greater Antilles started during the Middle Eocene and reached its maximum at the Eocene–Oligocene boundary. The area of the islands was reduced during the higher sea-level period in the Late Oligocene to Middle Miocene, but the islands remained emerged, possibly with the exception of parts of today’s Jamaica and Hispaniola, which are believed to have emerged permanently only during the Neogene (e.g. Iturralde-Vinent & MacPhee, 1999; Bartolini, Lang & Spell, 2003; Iturralde-Vinent, 2006). The origin of the present day terrestrial and freshwater biota of the archipelago has been explained by three alternative scenarios. The first scenario assumes that ancestors of extant lineages colonized the Antillean volcanic arc system from southern North America (via land bridges or by short-distance over-water dispersal) during the Late Cretaceous and Early Paleogene and survived there as relicts to the present (e.g. Rosen, 1975, 1985). This scenario was originally proposed for solenodons and Cricosaura lizards (e.g. Hedges, 2006), but only the Cricosaura clade has been confirmed to be old enough (Noonan et al., 2013; Sato et al., 2016). Under this scenario, the ancient Antillean fauna would have had to survive the collision of the large bolide with the Earth at the Yucatan Peninsula 65 Mya and its consequences that exterminated most of the Cretaceous terrestrial life forms in the Caribbean region (Iturralde-Vinent, 1982; Iturralde-Vinent & MacPhee, 1999; Tada et al., 2003). The second scenario suggests long-distance over-water dispersal from the surrounding landmasses during the Cenozoic (e.g. Hedges, Hass & Maxson, 1992; Hedges, 2006), which was recently documented, for example, for solenodons (Sato et al., 2016) and urocoptid snails (Uit de Weerd, Robinson & Rosenberg, 2016). In contrast to over-water dispersal, Iturralde-Vinent & MacPhee (1999) introduced the GAARlandia theory (GAAR = Greater Antilles + Aves Ridge) assuming a land bridge connection of the Greater Antilles with northern South America resulting from a tectonic uplift and sea-level fall at the Eocene–Oligocene boundary. This third scenario assumes that numerous terrestrial and freshwater organisms colonized the archipelago during a relatively short period (c. 35–33 Mya), as was suggested, for example, for Peltophryne toads (Alonso, Crawford & Bermingham, 2012), poecilid fishes (Weaver et al., 2016) or heroine cichlid fishes (Říčan et al., 2013). The GAARlandia hypothesis is still a subject of debate: Hedges (2006) and Ali (2012) point to the absence of geological evidence and lack of any clear signal in vertebrate phylogenetic analyses and hypothesize GAARlandia as no more than a chain of widely spaced islands situated between northern South America and the Greater Antilles. The above scenarios have been tested by numerous phylogenetic studies, which make the Caribbean one of the model regions for understanding historical island biogeography. Paradoxically, the vast majority of these studies are focused on vertebrates (i.e. relatively species-poor clades with rather young modern crown-groups; e.g. Roelants et al., 2007; Claramount & Cracraft, 2015; Foley, Springer & Teeling, 2016) and plants (i.e. groups with resistant seeds facilitating long-distance dispersal; e.g. Sanmartín & Ronquist, 2004). Little is known about the origin and biogeography of Caribbean insects and other terrestrial or freshwater arthropods, that is species-rich old groups that may provide numerous independent examples of evolutionary histories corroborating or contradicting the above biogeographic scenarios. Dated molecular phylogenetic studies have been performed only for selected groups of Caribbean butterflies (Wahlberg, 2006; Wahlberg & Freitas, 2007; Matos-Maraví et al., 2014; Lewis et al., 2015), wasps (Ceccarelli & Zaldívar-Riverón, 2013; Rodriguez, Pitts & von Dohlen, 2015) and arachnids (e.g. Crews & Gillespie, 2010; Zhang & Maddison, 2013; McHugh et al., 2014; Dziki et al., 2015; Esposito et al., 2015; Agnarsson et al., 2016). Only a single very recent study is available for Caribbean beetles (Coleoptera), which is the most speciose and most intensively studied insect group (Zhang et al., 2017). The beetle family Hydrophilidae comprises c. 3350 species inhabiting both aquatic and terrestrial habitats and is distributed worldwide (Hansen, 1999; Short & Fikáček, 2013; Lawrence & Slipinski, 2014; Seidel, Arriaga-Varela & Fikáček, 2016). In the Caribbean, the family is represented by 16 genera, 11 of which contain at least one species endemic to the Greater Antilles (Hansen, 1999; Short, 2004; Spangler & Short, 2008; Deler-Hernández, Cala-Riquelme & Fikáček, 2013a; Deler-Hernández, Fikáček & Cala-Riquelme, 2013b; Deler-Hernández, Cala-Riquelme & Fikáček, 2014; Arriaga-Varela et al., 2017). Among them, the genus Phaenonotum Sharp, 1882 contains 18 described and many undescribed species in the Neotropical and southern Nearctic Regions, and one East-African species (Hansen, 1999), all inhabiting semiaquatic or terrestrial habitats. A single widespread species, Phaenonotum exstriatum (Say, 1835), was originally recorded from the Caribbean, until the surprising discovery of P. delgadoi Deler-Hernández et al., 2013 endemic to eastern Cuba (Deler-Hernández et al., 2013a). Our subsequent fieldwork revealed additional candidates for endemic species in Jamaica, Hispaniola and Puerto Rico, indicating that Phaenonotum possibly underwent an island radiation in the Greater Antilles. In this paper, we analyse the systematics and evolutionary history of the Caribbean Phaenonotum in order to understand to what extent it corroborates the aforementioned scenarios explaining the origin of Caribbean faunas. MATERIAL AND METHODS DNA sequences The molecular study is based on Phaenonotum specimens collected in all main islands of the Greater Antilles during our fieldwork in 2010–2016. Specimens were collected either manually from aquatic habitats, rotten plant debris and at the light traps, or by sifting of forest leaf litter and extracting the beetles from sifting samples using Berlese and/or Winkler funnels. All specimens were collected in 96% ethanol and stored at −20 °C in the lab. To understand the origin of the Caribbean fauna, we included Phaenonotum species from Costa Rica, Guatemala, Venezuela, Suriname, Guyana, Ecuador, Peru and the USA, that is all continental DNA-grade material available to us. To test the monophyly of Phaenonotum, we also included all available DNA-grade specimens of the genera Phaenostoma Orchymont, 1937 and Lachnodacnum Orchymont, 1937, which form a strongly supported clade with Phaenonotum (V. Sýkora et al., unpubl. data). Outgroup taxa consist of selected genera representing the main clades of the tribe Coelostomatini (Cyclotypus Sharp, 1882, Dactylosternum Wollaston, 1854 and Coelostoma Brullé, 1835) and the sphaeridiine tribes Protosternini (Sphaerocetum Fikáček, 2010) and Sphaeridiini (Sphaeridium Fabricius, 1775). Most specimens used for this study were newly extracted and sequenced, but in a few cases, we adopted the sequences published by Short & Fikáček (2013). DNA was extracted from complete specimens cut into two parts between prothorax and mesothorax, using the commercial DNeasy Tissue Kit (Qiagen GmbH, Hilden, Germany) following the manufacturer’s instructions. Voucher specimens and DNA extractions are kept in the collection of the Department of Entomology of the National Museum, Prague (NMPC). Our gene selection corresponds with those used in previous phylogenetic studies of the family Hydrophilidae (e.g. Short & Fikáček, 2013) and contains three mitochondrial genes (3′ end cytochrome c oxidase I, cytochrome c oxidase II and 16S rRNA), and two nuclear genes (18S rRNA and 28S rRNA); sequences of 18S rRNA were amplified in two parts (3′ end and 5′ end) and subsequently combined; for primers and PCR conditions, see Supporting Information, Supplement S1. Ten microlitres of each PCR product were purified by adding 1.0 µL Exonuclease 1 [Exo1 (20 U/µL)] (ThermoFisherScientific) and 2.0 µL Thermosensitive Alkaline Phosphatase [FastAP (1 U/µL)] (ThermoFisherScientific) and incubating the mixture for 15 min at 37 °C followed by a 15 min inactivation step at 80 °C. Sanger sequencing was performed by BIOCEV (Vestec, Czech Republic). DNA sequences were assembled and edited using Geneious 7.1.9 and submitted to GenBank under the accession numbers shown in Supporting Information, Supplement S1. In total, we obtained sequences for 36 terminal taxa (25 specimens of Phaenonotum, four of Phaenostoma, one of Lachnodacnum and six outgroup taxa). Phylogenetic analyses Our molecular data set includes sequences of three ribosomal genes, which are non-trivial for alignment especially in case of distantly related taxa (e.g. Bocák et al., 2014), and a rather large proportion of missing data for cox2 and 16S genes, which we failed to amplify despite repeated effort. To test the impact of these problems, we compared the results of analyses of three different data sets: (1) 35 taxa data set: contains all Phaenonotum + Phaenostoma + Lachnodacnum specimens and all outgroups; (2) 30 taxa data set: contains all Phaenonotum + Phaenostoma + Lachnodacnum specimens, but is only rooted with closely related Cyclotypus (eliminates the problem with alignment of distantly related taxa); (3) 22 taxa data set: contains only Phaenonotum + Phaenostoma + Lachnodacnum with three and more genes, and is rooted by Cyclotypus (eliminates the problem with the alignment of non-related taxa and the effect of missing cox2 and 16S data). Specimens MF1736 and MF1741 from Venezuela were genetically very close (2.1% in cox1), corresponding to intraspecific variation in cox1 found in Phaenonotum exstriatum (0.0‒3.1%), and were hence considered as conspecific; they were combined for our analyses and included in all three data sets. Sequences were aligned with MAFFT algorithm as implemented in Geneious 7.1.9 software, using the default settings (score matrix = 200PAM/k = 2, gap open penalty = 1.53, offset value = 0.123). The final alignment has the total length of 4827 bp, consisting of the following gene fragments: cox1 (778 bp), cox2 (694 bp), 16S (530 bp), 18S (1798 bp) and 28S (1027 bp). The data set was divided into nine partitions (by genes, plus cox1 and cox2 were both divided by codon positions). Each data set was analysed using Bayesian inference and maximum likelihood. Bayesian analyses were performed in MrBayes 3.2.6 (Ronquist et al., 2012) using four chains of 25000000 generations and sampling the chain every 1000 generations. We sampled across the substitution model space in the Bayesian MCMC analysis itself (Huelsenbeck, Larget & Alfaro, 2004; Ronquist et al., 2012) instead of prior testing for an appropriate model for each of nine partitions. Results were examined in Tracer v.1.6 (Rambaut et al., 2014) to check for the proper effective sample size, proper mixing of chains and reaching the stationary phase; 25% burn-in was used for the construction of the final consensus tree. Maximum likelihood analyses were performed using RAxML 8.0 (Stamatakis, 2014) with a GTR substitution model and 1000 bootstrap replicates. Resulting trees were visualized in FigTree 1.4.3 (Rambaut, 2012). Divergence dating We performed a divergence dating analysis using the full data set of 35 taxa in BEAST 2.4.5 (Bouckaert et al., 2014) with fixed tree topology as revealed by the Bayesian analysis of the 35 taxa data set; the internal topology of P. exstriatum was manipulated to be bifurcate in agreement with results of the maximum likelihood analyses of 35 and 30 taxa. We divided the data set into five partitions corresponding to each gene (cox1, cox2, 16S rRNA, 18S rRNA and 28S rRNA) and used PartitionFinder v1.1.1 (Lanfear et al., 2012) to estimate the evolutionary model that best fit the data for each partition separately. The following substitution models were selected using Bayesian information criterion and used for particular genes in the divergence dating analysis: GTR + I + G for cox1, cox2 and 28S; GTR + G for 16S; and TrNef + I + G for 18S. Due to the absence of fossils belonging to the tribe Coelostomatini, we used the combination of a rate dating and node dating constraining the age of the most recent common ancestor (MRCA) of the Coelostomatini. Molecular clock models were linked into two partitions (mtDNA: cox1, cox2, 16S rRNA; nDNA: 18S rRNA, 28S rRNA) and substitution rates were set to 0.0133 (mtDNA) and 0.0017 (nDNA) substitutions per million years following Papadopoulou, Anastasiou & Vogler (2010). The age of the MRCA of the Coelostomatini was constrained to 152.5 Mya [95% highest posterior density (confidence interval): 134–170 Mya], following the results of the time tree analysis of the whole family Hydrophilidae performed by Bloom, Fikáček & Short (2014) based on a wide spectrum of fossil calibrations (for list of fossils used see supplementary material in Bloom et al., 2014). We performed a Bayes Factor comparison to test alternative clock models (non-clock, strict clock, relaxed clock) using a stepping-stone method (Xie et al., 2011) as implemented in MrBayes 3.2.6 (Ronquist et al., 2012) (see Supporting Information, Supplement S1). The birth–death model was used for the tree prior as this model is commonly used to model speciations and extinctions in inferring phylogenies using Bayesian methods; thus, at any point in time, every lineage can undergo speciation at rate λ or go extinct at rate μ (Stadler, 2009). Due to the problems with convergence of parameters, we performed two separate runs each with MCMC chain length set to 500 million generations and sampling frequency of every 25000 generations. We combined both runs using LogCombiner 2.4.0 and applied a 10% burn-in fraction after checking the convergence of parameters in TRACER 1.6 (Rambaut et al., 2014). TreeAnnotator 2.4.1 was used to summarize the information of all trees in the sample onto the maximum clade credibility tree. The resulting tree was visualized in FigTree 1.4.3 (Rambaut, 2012). Historical biogeography reconstruction For historical biogeography estimation, we used the time tree resulting from the divergence dating analysis as the input tree, from which we excluded all outgroups except Cyclotypus. Four specimens of P. exstriatum were included into the analysis, each representing a geographically distinct population. An alternative set of analyses with a single terminal for P. exstriatum was also performed. The distribution ranges were divided into the following eight areas: A ‒ North America incl. Mexico; B ‒ Central America; C ‒ northern South America (corresponding to Guiana Shield + northwestern Venezuela and northern Colombia, i.e. regions on the Caribbean coast and in direct contact with Central America); D ‒ South America (remaining continent south of the former region); E ‒ Cuba; F ‒ Jamaica; G ‒ Hispaniola; H ‒ Puerto Rico. As no inter-island divergences are present in our phylogeny, we coded the Caribbean islands in their current shape, without considering their historical composition of multiple paleoislands. Islands of the Lesser Antilles were not considered as separate areas, as only the youngest and widespread P. exstriatum is known to occur in some of them; instead, the presence/absence of the Lesser Antilles was considered when setting dispersal multipliers for the time-stratified analysis (see below). The distribution of terminal taxa is based on examined specimens in all undescribed species; distribution of P. exstriatum and P. laevicolle complex on the continent follows Smetana (1978); Hansen (1999); Oliva, Fernández & Bachmann (2002) and Gonzáles-Rodríguez, García-Hernández & Clarkson (2017). We carried out the historical biogeography analyses in the R package BioGeoBEARS (Matzke, 2014) to estimate the timing of Phaenonotum colonization of the Greater Antilles archipelago. This package contains three models implemented in a maximum likelihood framework: DEC model (Ree & Smith, 2008), DIVALIKE model (likelihood version of the DIVA model: Ronquist, 1997) and BAYAREALIKE model (likelihood version of BayArea model: Landis et al., 2013). Moreover, each model is available in its original version and with an additional parameter j representing the founder event, that is speciation following long-distance dispersal; six different models are hence available in total. We conducted non-time-stratified and time-stratified analyses to estimate ancestral area distribution on given nodes, each with all six models. Unconstrained non-stratified analyses were done using default parameter values. For time-stratified analyses, time periods were defined as follows, to reflect different paleogeography of the area in each period (Iturralde-Vinent, 2006; O’Dea et al., 2016): (1) 0–2.8 Mya: from present to the closing of Isthmus of Panama; (2) 2.8–9 Mya: the Greater Antilles emerged, the chain of islands (Lesser Antilles) present between the Greater Antilles and South America; (3) 9–32 Mya: the Greater Antilles significantly reduced in area and widely separated from South America by deep sea; (4) 32–38 Mya: the Greater Antilles connected to northern South America by GAARlandia land bridge; (5) 38–55 Mya: the Greater Antilles started to form and were well separated from surrounding continents; (6) 55–110 Mya: prior to the formation of the Greater Antilles. Dispersal probabilities were set to reflect the paleogeography as follows: 0.8 for adjacent continental areas; 0.5 for non-adjacent continental areas; 0.2 for adjacent islands (or island-continent) separated by less than 200 km from sea; 0.05 for connection by island chain (e.g. Lesser Antilles) or intermediate island (e.g. Hispaniola between Cuba and Puerto Rico); 0.001 for long-distance dispersal (areas separated by more than 200 km from sea); 0.000001 when dispersal was not possible (i.e. when the area was submerged; we followed the BioGeoBEARS manual in setting extremely low rather than zero probabilities in such cases). Areas not present during particular time slice (e.g. Cuba in the Late Cretaceous) were not allowed for the reconstruction using the Areas Allowed matrix. Models of both non-time-stratified and time-stratified analyses were compared using likelihood values and Akaike information criterion corrected for small sample sizes (AICc) (Matzke, 2013). All input files for the biogeography analyses are available in Supporting Information, Supplement S2. Morphology and taxonomy Examined specimens are deposited in the following collections: BMNH Natural History Museum, London, United Kingdom (M.V.L. Barclay); CMN Canadian Museum of Nature, Ottawa, Canada (R. Anderson); DZRJ Coleção Entomológica Prof. José Alfredo Pinheiro Dutra, Instituto de Biologia, Universidade Federal de Rio de Janeiro, Rio de Janeiro, Brazil (B. Clarkson); MCZ Museum of Comparative Zoology, Cambridge, USA (P. Perkins); MNHNSD Museo Nacional de Historia Natural, Santo Domingo, Dominican Republic (C. Suriel); NHMW Naturhistorisches Museum, Wien, Austria (M. Jäch, A. Komarek); NMPC National Museum, Prague, Czech Republic (M. Fikáček); SBNM Santa Barbara Museum of Natural History, California, USA (M. L. Gimmel); SBPC Stewart Peck Personal Collection, Ottawa, Canada; UPRM University of Puerto Rico, Mayagüez, Puerto Rico (A. Segarra); ZMHB Museum für Naturkunde der Humboldt-Universität, Berlin, Germany (J. Frisch, B. Jäger). Habitus photographs were taken using Canon EOS 550D digital camera with attached Canon MP-E65mm f/2.8 1–5× macro lens and subsequently adapted in Adobe Photoshop CS5 and CorelDRAW Home & Student X8. Photographs of genitalia were taken using Canon EOS 1100D digital camera attached to Olympus BX41 compound microscope and subsequently combined with Helicon Focus software. Scanning electron micrographs of the holotype of the new species were taken using a Hitachi S-3700N environmental electron microscope at the Department of Paleontology, National Museum in Prague. General morphological terminology follows Smetana (1978), Archangelsky (1989) and Hansen (1991). All Caribbean species were compared with type specimens of the Central and South American species deposited in the Natural History Museum, London (Sharp collection; see Deler-Hernández & Fikáček, 2016) and in the Institut Royal des Sciences Naturelles de Belgique, Brussels (Orchymont collection), and with unidentified material from Mexico, Costa Rica, Ecuador and Peru deposited in National Museum, Prague, in order to reveal whether or not they might be conspecific with continental species. Complete data of all examined specimens in DarwinCore-formatted Excel spreadsheet are available in Supporting Information, Supplement S3 and in the data set submitted to Zenodo. Depository of primary data All primary data and results of all analyses were uploaded as a.zip file into the Zenodo depository (https://zenodo.org/) under doi: 10.5281/zenodo.850595. Parts of the data were also uploaded to specialized archives as specified below. RESULTS Phylogenetic analyses All analyses performed revealed a strongly supported clade consisting of Neotropical genera Phaenostoma, Phaenonotum and Lachnodacnum, but do not support the current status of the genera (Fig. 1). Phaenonotum caribenseArchangelsky, 1989 is revealed as not related to the remaining Phaenonotum species and is revealed as a sister group to the rest of the Neotropical clade in most analyses (strongly supported by Bayesian analysis with 30 and 22 taxa, weakly supported by maximum likelihood analyses); in the Bayesian analysis with 35 taxa, it is revealed as sister to Phaenostoma kontax Gustafson & Short, 2010 and Lachnodacnum, with moderate support. When P. caribense is excluded, the genus Phaenonotum (= core Phaenonotum hereafter) is revealed as strongly supported monophylum in all analyses. The genus Phaenostoma is revealed as sister to the core Phaenonotum in most analyses (moderately supported by Bayesian analysis except that with 35 taxa, weakly supported by all maximum likelihood analyses), with Lachnodacnum always nested within Phaenostoma. In the Bayesian analysis with 35 taxa, Phaenonotum caribense + Phaenostoma kontax + Lachnodacnum clade is revealed as sister to the Phaenostoma posticatum + core Phaenonotum. Figure 1. View largeDownload slide The phylogenetic position and polyphyly of Phaenonotum as revealed by our maximum likelihood (A) and Bayesian analyses (B). Branch support is indicated for analyses with 35, 30 and 22 terminal taxa, respectively. Figure 1. View largeDownload slide The phylogenetic position and polyphyly of Phaenonotum as revealed by our maximum likelihood (A) and Bayesian analyses (B). Branch support is indicated for analyses with 35, 30 and 22 terminal taxa, respectively. Within the core Phaenonotum clade, all six analyses revealed the same five strongly to moderately supported clades and two undescribed species (MF1739 from Guatemala and MF861 from Peru) not closely related to other species and forming deeply divergent separate clades (Fig. 2). The relationships between these seven principal clades are the same in all analyses performed, with the exception of MF1739, which is revealed as early branching in Bayesian analyses and more deeply nested in maximum likelihood analyses. All Caribbean species are part of the strongly supported deeply nested monophylum containing three of the principal clades: the clade of the endemic Cuban, Jamaican and Hispaniolan species; the clade containing Puerto Rican endemic P. borinquenum sp. nov., P. laevicolle complex and an undescribed species from Guyana (MF1061); and the clade consisting of P. exstriatum and undescribed species from Venezuela and Suriname. Within the Caribbean endemic clade, the internal topology is the same in all analyses, with the Cuban P. delgadoi diverging first, and the Jamaican P. ondreji sp. nov. and Hispaniolan P. laterale sp. nov. as sister taxa. The internal topology of the P. borinquenum + laevicolle clade varies the most among analyses, with either P. borinquenum or the undescribed species from Guyana (MF1061) revealed as the earliest diverging taxon. The internal topology of P. exstriatum clade differs between Bayesian and maximum likelihood analyses, with Puerto Rican (MF1728) and Costa Rican (MF1738) revealed as sister taxa in maximum likelihood ones; the Cuban specimen (MF654) is always revealed as sister to the one from USA: Delaware (MF1063). Figure 2. View largeDownload slide Summary of the internal topology of the core Phaenonotum clade revealed by Bayesian and maximum likelihood analyses of all three data sets (35 taxa, 30 taxa and 22 taxa, see Material and methods for details). A, topology resulting from Bayesian inference of 35 taxa data set, with the presence and support of particular clades in all analyses mapped by colour squares; B, alternative topology of the borinquenum‒laevicolle clade obtained from maximum likelihood analysis of the 30 taxa data set; C, alternative internal topology of P. exstriatum clade obtained from maximum likelihood analyses of 35 and 30 taxa data sets. Figure 2. View largeDownload slide Summary of the internal topology of the core Phaenonotum clade revealed by Bayesian and maximum likelihood analyses of all three data sets (35 taxa, 30 taxa and 22 taxa, see Material and methods for details). A, topology resulting from Bayesian inference of 35 taxa data set, with the presence and support of particular clades in all analyses mapped by colour squares; B, alternative topology of the borinquenum‒laevicolle clade obtained from maximum likelihood analysis of the 30 taxa data set; C, alternative internal topology of P. exstriatum clade obtained from maximum likelihood analyses of 35 and 30 taxa data sets. Divergence dating The relaxed clock model was identified as the best-fitting clock model based on the Bayes factor comparison (see Supporting Information, Supplement S1), and it was hence implemented in BEAST time tree analysis. The core Phaenonotum clade was estimated to originate during the middle Cretaceous (c. 102 Mya), with the diversification of modern clades starting in Late Cretaceous (c. 83 Mya). The clade composed of the endemic Cuban, Jamaican and Hispaniolan Phaenonotum diverged during the Eocene (c. 47 Mya), with P. delgadoi from Cuba diverging at about Eocene–Oligocene boundary (c. 36 Mya), and P. ondreji (Jamaica) from P. laterale (Hispaniola) during the Oligocene (c. 26.4 Mya). The Puerto Rican endemic P. borinquenum diverged in the early Miocene (c. 19 Mya). The divergence of the Cuban and Venezuelan specimens of P. laevicolle complex was dated to the Pliocene–Pleistocene (c. 4.7 Mya), and the modern populations of the widespread P. exstriatum diverged at about Pliocene–Pleistocene boundary (c. 2.4 Mya). Precise ages and 95% confidence intervals are listed in Table 1 for the principal Phaenonotum clades and illustrated in Supporting Information, Supplement S1 for all clades. Table 1. Divergence ages and their confidence intervals for major Phaenonotum clades including all Caribbean ones Age (Mya) 95% confidence interval (Mya) Phaenonotum stem 101.7 81.7‒121.9 Phaenonotum crown 82.5 65.4‒100.2 Caribbean Phaenonotum stem 46.6 35.6‒58.1 P. delgadoi 35.7 25.5‒47.2 P. ondreji and P. laterale 26.4 17.2‒36.5 P. borinquenum 19.2 12.1‒27.1 P. laevicolle from Cuba 4.7 2.2‒7.9 P. exstriatum crown 2.4 1.3‒3.7 Age (Mya) 95% confidence interval (Mya) Phaenonotum stem 101.7 81.7‒121.9 Phaenonotum crown 82.5 65.4‒100.2 Caribbean Phaenonotum stem 46.6 35.6‒58.1 P. delgadoi 35.7 25.5‒47.2 P. ondreji and P. laterale 26.4 17.2‒36.5 P. borinquenum 19.2 12.1‒27.1 P. laevicolle from Cuba 4.7 2.2‒7.9 P. exstriatum crown 2.4 1.3‒3.7 View Large Table 1. Divergence ages and their confidence intervals for major Phaenonotum clades including all Caribbean ones Age (Mya) 95% confidence interval (Mya) Phaenonotum stem 101.7 81.7‒121.9 Phaenonotum crown 82.5 65.4‒100.2 Caribbean Phaenonotum stem 46.6 35.6‒58.1 P. delgadoi 35.7 25.5‒47.2 P. ondreji and P. laterale 26.4 17.2‒36.5 P. borinquenum 19.2 12.1‒27.1 P. laevicolle from Cuba 4.7 2.2‒7.9 P. exstriatum crown 2.4 1.3‒3.7 Age (Mya) 95% confidence interval (Mya) Phaenonotum stem 101.7 81.7‒121.9 Phaenonotum crown 82.5 65.4‒100.2 Caribbean Phaenonotum stem 46.6 35.6‒58.1 P. delgadoi 35.7 25.5‒47.2 P. ondreji and P. laterale 26.4 17.2‒36.5 P. borinquenum 19.2 12.1‒27.1 P. laevicolle from Cuba 4.7 2.2‒7.9 P. exstriatum crown 2.4 1.3‒3.7 View Large Biogeography analyses Of the six models included in BioGeoBEARS, those implementing a founder event (i.e. perapatric speciation, j parameter in the analyses) fit the data better both in time-stratified and non-time-stratified analyses than those not allowing for jump dispersal (Table 2). Analyses allowing the founder event resulted in identical ancestral area estimates in all nodes and corners in non-time-stratified analyses under all three basic models (DEC + J, DIVALIKE + J and BAYAREALIKE + J) and in identical estimates in time-stratified analyses with DEC + J and DIVALIKE + J models (time-stratified analysis with BAYAREALIKE + J model is nearly identical, only with slightly different scenario for P. exstriatum clade and P. borinquenum, see Supporting Information, Supplement S2 for details). Models not implementing the founder event resulted in very different ancestral area estimates for each model in non-time-stratified analyses, but in identical estimates under all three models in time-stratified ones. The number of continent-to-Caribbean and Caribbean-to-continent dispersal events (and source/sink regions on continents) for all three models in non-time-stratified and time-stratified analyses are shown in Table 3. Based on the AICc metrics, the best-performing models were DIVALIKE + J for both non-time-stratified and time-stratified analyses (Fig. 3). Table 2. Comparison of the models used for non-time-stratified and time-stratified analyses of the data set with multiple populations of P. exstriatum LnL npar d e j AIC AICc Non-time constrained  DEC −91.91 2 0.0013 < 0.0001 0 187.8 188.3  DEC + J −70.21 3 0.0004 < 0.0001 0.046 146.4 147.3  DIVALIKE −84.36 2 0.0016 < 0.0001 0 172.7 173.2  DIVALIKE + J* −69.8 3 0.0005 < 0.0001 0.060 145.6 146.5  BAYAREALIKE −125.4 2 0.0023 0.026 0 254.8 255.2  BAYAREALIKE + J −70.56 3 0.0004 < 0.0001 0.049 147.1 148 Time constrained  DEC −104.9 2 0.016 0.0049 0 213.9 214.3  DEC + J −98.94 3 0.011 0.0039 0.17 203.9 204.8  DIVALIKE −100.2 2 0.017 0.0042 0 204.4 204.8  DIVALIKE + J* −95.44 3 0.012 0.0034 0.099 196.9 197.8  BAYAREALIKE −134.2 2 0.033 0.025 0 272.5 272.9  BAYAREALIKE + J −101.8 3 0.0077 0.0066 0.24 209.5 210.5 LnL npar d e j AIC AICc Non-time constrained  DEC −91.91 2 0.0013 < 0.0001 0 187.8 188.3  DEC + J −70.21 3 0.0004 < 0.0001 0.046 146.4 147.3  DIVALIKE −84.36 2 0.0016 < 0.0001 0 172.7 173.2  DIVALIKE + J* −69.8 3 0.0005 < 0.0001 0.060 145.6 146.5  BAYAREALIKE −125.4 2 0.0023 0.026 0 254.8 255.2  BAYAREALIKE + J −70.56 3 0.0004 < 0.0001 0.049 147.1 148 Time constrained  DEC −104.9 2 0.016 0.0049 0 213.9 214.3  DEC + J −98.94 3 0.011 0.0039 0.17 203.9 204.8  DIVALIKE −100.2 2 0.017 0.0042 0 204.4 204.8  DIVALIKE + J* −95.44 3 0.012 0.0034 0.099 196.9 197.8  BAYAREALIKE −134.2 2 0.033 0.025 0 272.5 272.9  BAYAREALIKE + J −101.8 3 0.0077 0.0066 0.24 209.5 210.5 AIC, Akaike information criterion; AICc, size-corrected AIC; LnL, log likelihood; n par, number of parameters in the analysis; d, e, j, parameres of the model (d, dispersal; e, extinction, j, founder event). *Best-performing model for each groups of analyses. View Large Table 2. Comparison of the models used for non-time-stratified and time-stratified analyses of the data set with multiple populations of P. exstriatum LnL npar d e j AIC AICc Non-time constrained  DEC −91.91 2 0.0013 < 0.0001 0 187.8 188.3  DEC + J −70.21 3 0.0004 < 0.0001 0.046 146.4 147.3  DIVALIKE −84.36 2 0.0016 < 0.0001 0 172.7 173.2  DIVALIKE + J* −69.8 3 0.0005 < 0.0001 0.060 145.6 146.5  BAYAREALIKE −125.4 2 0.0023 0.026 0 254.8 255.2  BAYAREALIKE + J −70.56 3 0.0004 < 0.0001 0.049 147.1 148 Time constrained  DEC −104.9 2 0.016 0.0049 0 213.9 214.3  DEC + J −98.94 3 0.011 0.0039 0.17 203.9 204.8  DIVALIKE −100.2 2 0.017 0.0042 0 204.4 204.8  DIVALIKE + J* −95.44 3 0.012 0.0034 0.099 196.9 197.8  BAYAREALIKE −134.2 2 0.033 0.025 0 272.5 272.9  BAYAREALIKE + J −101.8 3 0.0077 0.0066 0.24 209.5 210.5 LnL npar d e j AIC AICc Non-time constrained  DEC −91.91 2 0.0013 < 0.0001 0 187.8 188.3  DEC + J −70.21 3 0.0004 < 0.0001 0.046 146.4 147.3  DIVALIKE −84.36 2 0.0016 < 0.0001 0 172.7 173.2  DIVALIKE + J* −69.8 3 0.0005 < 0.0001 0.060 145.6 146.5  BAYAREALIKE −125.4 2 0.0023 0.026 0 254.8 255.2  BAYAREALIKE + J −70.56 3 0.0004 < 0.0001 0.049 147.1 148 Time constrained  DEC −104.9 2 0.016 0.0049 0 213.9 214.3  DEC + J −98.94 3 0.011 0.0039 0.17 203.9 204.8  DIVALIKE −100.2 2 0.017 0.0042 0 204.4 204.8  DIVALIKE + J* −95.44 3 0.012 0.0034 0.099 196.9 197.8  BAYAREALIKE −134.2 2 0.033 0.025 0 272.5 272.9  BAYAREALIKE + J −101.8 3 0.0077 0.0066 0.24 209.5 210.5 AIC, Akaike information criterion; AICc, size-corrected AIC; LnL, log likelihood; n par, number of parameters in the analysis; d, e, j, parameres of the model (d, dispersal; e, extinction, j, founder event). *Best-performing model for each groups of analyses. View Large Table 3. Number of continent-to-Caribbean and Caribbean-to-continent colonization events (total number and number of colonizations from/to each continental area) inferred for the genus Phaenonotum in analyses including four populations of P. exstriatum Model Time constraints Continent-to-Caribbean events Caribbean-to-Continent events Total NA CA nSA CA + SA NA + CA + SA Total NA CA DEC Non-strat. 3× 1× 2× ‒ ‒ 1× 1× ‒ DIVALIKE Non-strat. 4× ‒ 4× ‒ ‒ 0× ‒ ‒ BAYAREALIKE Non-strat. 5× 2× ‒ 3× ‒ 1× 1× ‒ DEC + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DIVALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× BAYAREALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DEC Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DIVALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× BAYAREALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DEC + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× DIVALIKE + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× BAYAREALIKE + J Time-strat. 5× 1× ‒ 4× ‒ ‒ 1× ‒ 1× Model Time constraints Continent-to-Caribbean events Caribbean-to-Continent events Total NA CA nSA CA + SA NA + CA + SA Total NA CA DEC Non-strat. 3× 1× 2× ‒ ‒ 1× 1× ‒ DIVALIKE Non-strat. 4× ‒ 4× ‒ ‒ 0× ‒ ‒ BAYAREALIKE Non-strat. 5× 2× ‒ 3× ‒ 1× 1× ‒ DEC + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DIVALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× BAYAREALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DEC Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DIVALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× BAYAREALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DEC + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× DIVALIKE + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× BAYAREALIKE + J Time-strat. 5× 1× ‒ 4× ‒ ‒ 1× ‒ 1× Models best fitting the data are marked in grey. Abbreviations of source/sink areas: CA, Central America; NA, North America; nSA, northern South America. Time constraint, non-strat., non-time-stratified analysis; time-strat., time-stratified analysis. View Large Table 3. Number of continent-to-Caribbean and Caribbean-to-continent colonization events (total number and number of colonizations from/to each continental area) inferred for the genus Phaenonotum in analyses including four populations of P. exstriatum Model Time constraints Continent-to-Caribbean events Caribbean-to-Continent events Total NA CA nSA CA + SA NA + CA + SA Total NA CA DEC Non-strat. 3× 1× 2× ‒ ‒ 1× 1× ‒ DIVALIKE Non-strat. 4× ‒ 4× ‒ ‒ 0× ‒ ‒ BAYAREALIKE Non-strat. 5× 2× ‒ 3× ‒ 1× 1× ‒ DEC + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DIVALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× BAYAREALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DEC Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DIVALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× BAYAREALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DEC + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× DIVALIKE + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× BAYAREALIKE + J Time-strat. 5× 1× ‒ 4× ‒ ‒ 1× ‒ 1× Model Time constraints Continent-to-Caribbean events Caribbean-to-Continent events Total NA CA nSA CA + SA NA + CA + SA Total NA CA DEC Non-strat. 3× 1× 2× ‒ ‒ 1× 1× ‒ DIVALIKE Non-strat. 4× ‒ 4× ‒ ‒ 0× ‒ ‒ BAYAREALIKE Non-strat. 5× 2× ‒ 3× ‒ 1× 1× ‒ DEC + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DIVALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× BAYAREALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DEC Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DIVALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× BAYAREALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DEC + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× DIVALIKE + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× BAYAREALIKE + J Time-strat. 5× 1× ‒ 4× ‒ ‒ 1× ‒ 1× Models best fitting the data are marked in grey. Abbreviations of source/sink areas: CA, Central America; NA, North America; nSA, northern South America. Time constraint, non-strat., non-time-stratified analysis; time-strat., time-stratified analysis. View Large Figure 3. View largeDownload slide Results of the ancestral area estimation using time-stratified and non-time-stratified DIVALIKE + J model of BioGeoBEARS. Maps (A–G) show simplified continent and island positions in the respective time window used for the time-stratified analysis. Pie charts show two or three most probable ancestral areas for the respective node. Figure 3. View largeDownload slide Results of the ancestral area estimation using time-stratified and non-time-stratified DIVALIKE + J model of BioGeoBEARS. Maps (A–G) show simplified continent and island positions in the respective time window used for the time-stratified analysis. Pie charts show two or three most probable ancestral areas for the respective node. Most analyses infer South America as the ancestral range of the MRCA of Phaenonotum (Fig. 3), with several independent colonizations of northern South America and Central America (through northern South America in time-constrained analyses). Most analyses (non-time-stratified allowing for founder event and all time-stratified except BAYAREALIKE + J model) infer that ancestor(s) of P. delgadoi + P. laterale + P. ondreji colonized the Caribbean from northern South America: non-time-stratified models and time-stratified models not allowing for founder event infer a single colonization followed by series of founder events (non-time-stratified analyses) or combination of range expansions and vicariance events (time-stratified analyses); time-stratified analyses allowing for founder event infer two independent colonizations followed by vicariance event between Jamaica and Hispaniola. Phaenonotum borinquenum and Caribbean P. laevicolle were estimated to colonize the Caribbean independently by the non-time-stratified analysis. In contrast, time-stratified analysis favours earlier colonization of the Caribbean by the ancestor of these taxa and back-colonization from the Caribbean to the continent in P. laevicolle. Phaenonotum exstriatum was revealed to be of South American origin in non-time-constrained analyses allowing for founder event and all time-constrained analyses; the ancestral range of its MRCA was either Caribbean or Caribbean + Central America, depending on the model implemented. The results of all analyses are available in the Supporting Information, Supplement S2. Alternative analyses with P. exstriatum as a single terminal resulted in estimates nearly identical to those treating the four populations of P. exstriatum separately, with DEC + J as the best-performing model for time-stratified and non-time-stratified analyses. Both analyses revealed northern South America as an ancestral range of the MRCA of P. exstriatum and its sister species (undescribed species from Suriname, voucher MF1062). Results of these analyses are available in Supporting Information, Supplement S2. Systematics of the Caribbean Phaenonotum Key to the Caribbean Phaenonotum 1. Eyes small, interocular distance 5.0–6.0× the width of one eye in dorsal view (Fig. 6A–C). Metaventrite very short, its total width 5.0–6.0× the length behind mesocoxae (Fig. 7A–C). Species without hind wings (apterous). 2 – Eyes moderately large, interocular distance 3.5–4.0× the width of one eye in dorsal view (Fig. 6D–F). Metaventrite moderately long, its total width 3.0–4.0× the length behind mesocoxae (Fig. 7D–E). Species with fully developed hind wings (macropterous). 4 2. Metaventrite with pubescent cavities at sides of meso-metaventral process (Fig. 7A). Elytral suture strongly elevated posteriorly; elytral punctation very coarse (Fig. 5A). Pronotum with moderately coarse punctation, without microsculpture. Eastern Cuba (Fig. 4). P. delgadoi Deler-Hernández et al., 2013. – Metaventrite without pubescent cavities at sides of meso-metaventral process (Fig. 7B–F). Elytral suture weakly elevated posteriorly; elytral punctation never extremely coarse. Pronotum either with distinct microsculpture or with extremely fine (nearly invisible) punctation. 3 3. Body length 3.1–3.4 mm. Head and pronotum with mesh-like microsculpture and distinct punctation. Elytral punctation rather coarse, yellowish margins of elytra present in apical half only (Fig. 5B). Mesoventral process moderately wide, median portion of metaventrite with narrow highly elevated median keel (Fig. 7C). Aedeagus 0.6 mm long (Fig. 6H). Jamaica (Fig. 4). P. ondreji sp. nov. – Body length 2.5–2.7 mm. Head and pronotum without microsculpture, pronotal punctation extremely fine, almost invisible. Elytral punctation fine, whole lateral margins of elytra with yellowish stripe (Fig. 5C). Mesoventral process very narrow, metaventrite with slightly elevated median portion, without median keel. Aedeagus 0.35 mm long (Fig. 6I). Hispaniola (Fig. 4). P. laterale sp. nov. 4. Median lobe of the aedeagus wide; gonopore large, situated in apical third of the median lobe. Bases of lateral struts of the median lobe weakly expanded laterally (Fig. 6L). Widespread in the Greater Antilles (Fig. 4). .. P. laevicolle complex – Median lobe of the aedeagus narrow; gonopore small, situated subapically. Bases of lateral struts of the median lobe largely expanded laterally (Fig. 6J, K). 5 5. Body length 3.3–4.0 mm. Bases of parameres meeting in single point (Fig. 6K). Widespread in Greater and Lesser Antilles (Fig. 4). P. exstriatum (Say, 1835) – Body length 3.2–3.3 mm. Bases of parameres widely joined together (Fig. 6J). Puerto Rico (Fig. 4). P. borinquenum sp. nov. Figure 4. View largeDownload slide Current distribution of the genus Phaenonotum in the Caribbean. A, single island endemics; B, P. exstriatum; C, P. laevicolle complex. The localities of sequenced specimens are shown and marked by the code of the respective DNA extract. Figure 4. View largeDownload slide Current distribution of the genus Phaenonotum in the Caribbean. A, single island endemics; B, P. exstriatum; C, P. laevicolle complex. The localities of sequenced specimens are shown and marked by the code of the respective DNA extract. Figure 5. View largeDownload slide Habitus of the Caribbean Phaenonotum (dorsal and lateral views). A–D, species endemic for the Greater Antilles: A, P. delgadoi Deler-Hernández et al., 2013 from eastern Cuba; B, P. ondreji sp. nov. from Jamaica; C, P. laterale sp. nov. from southern Hispaniola; D, P. borinquenum sp. nov. from Puerto Rico. E, F, non-endemic species: E, P. exstriatum (Say, 1835) from Cuba; F, Phaenonotum sp. from the P. laevicolle complex from Cuba. Figure 5. View largeDownload slide Habitus of the Caribbean Phaenonotum (dorsal and lateral views). A–D, species endemic for the Greater Antilles: A, P. delgadoi Deler-Hernández et al., 2013 from eastern Cuba; B, P. ondreji sp. nov. from Jamaica; C, P. laterale sp. nov. from southern Hispaniola; D, P. borinquenum sp. nov. from Puerto Rico. E, F, non-endemic species: E, P. exstriatum (Say, 1835) from Cuba; F, Phaenonotum sp. from the P. laevicolle complex from Cuba. Figure 6. View largeDownload slide Head in frontal view (A–F) and aedeagus (G–L) of the Caribbean Phaenonotum. A, G, P. delgadoi Deler-Hernández et al., 2013; B, H, P. ondreji sp. nov.; C, I, P. laterale sp. nov.; D, J, P. borinquenum sp. nov.; E, K, P. exstriatum (Say, 1835); F, L, Phaenonotum sp. from the P. laevicolle complex from Cuba. A–F, not to scale. Figure 6. View largeDownload slide Head in frontal view (A–F) and aedeagus (G–L) of the Caribbean Phaenonotum. A, G, P. delgadoi Deler-Hernández et al., 2013; B, H, P. ondreji sp. nov.; C, I, P. laterale sp. nov.; D, J, P. borinquenum sp. nov.; E, K, P. exstriatum (Say, 1835); F, L, Phaenonotum sp. from the P. laevicolle complex from Cuba. A–F, not to scale. Figure 7. View largeDownload slide Meso-metaventral morphology of the Caribbean Phaenonotum (left: middle and right portions of complete meso-metaventrite, right: detail of mesoventral elevation and anterior metaventral process). A, P. delgadoi Deler-Hernández et al., 2013; B, P. laterale sp. nov.; C, P. ondreji sp. nov.; D, P. borinquenum sp. nov.; E, P. exstriatum (Say, 1835); F, Phaenonotum sp. from the P. laevicolle complex from Cuba. Not to scale. Figure 7. View largeDownload slide Meso-metaventral morphology of the Caribbean Phaenonotum (left: middle and right portions of complete meso-metaventrite, right: detail of mesoventral elevation and anterior metaventral process). A, P. delgadoi Deler-Hernández et al., 2013; B, P. laterale sp. nov.; C, P. ondreji sp. nov.; D, P. borinquenum sp. nov.; E, P. exstriatum (Say, 1835); F, Phaenonotum sp. from the P. laevicolle complex from Cuba. Not to scale. Phaenonotum borinquenum sp. nov. (Figs 4A, 5D, 6D, J, 7D) Type locality: Puerto Rico, Naguabo, El Yunque National Forest, 18°16.1′N 65°48.1′W, 575 m. Type material: Holotype: male (NMPC): ‘PUERTO RICO: Naguabo: El Yunque Nat. Forest, S part La Sabana recr area 5.25 km N of Río Blanco above rd. PR191, 18°16.1′N 65°48.1′W, elevation 575 m, 21.vi.2016, A. Deler-Hernández lgt., PR05/DNA isolation: MF1730, isolated by A. Deler-Hernández 2016, isolate deposited at Department of Entomology, National Museum in Prague (green printed)’ (molecular voucher MF1730). Paratype: one female (NMPC; associated with the holotype by cox1 sequence): Puerto Rico: Naguabo, El Yunque Nat. Forest, S part La Sabana recr area 5.6 km N of Río Blanco at rd. PR191, 18°16.1′N 65°47.6′W, 510 m a.s.l., 21.vi.–2.vii.2016, Deler-Hernández, Fikáček, Seidel lgt., PR03/DNA isolation: MF1729, isolated by A. Deler-Hernández 2016, isolate deposited at Department of Entomology, National Museum in Prague (green printed)’ (Molecular voucher MF1729). Description: Habitus as in Figure 5D. Body size 3.2–3.3 mm (holotype: 3.3 mm). Body elongate oval, moderately convex, elytral suture not elevated. Dorsum black (dark brown in the teneral paratype), lateral margins of pronotum and elytra without distinct paler stripe; ventral surface dark brown to black; femora and tibiae black; antennae, maxillary palpi and tarsi yellowish. Head with sparse and moderately coarse punctation, without microsculpture (except posteriorly on frons); eyes moderately large, separated by 4.0× dorsal width of one eye (Fig. 6D). Pronotum with sparse punctures slightly finer than on head, interstices without microsculpture. Elytral punctation strongly impressed, coarser than on pronotum and head; elytral interstices without microsculpture. Wings present, fully developed. Mesoventral elevation as wide as metaventral process posteriorly, narrowing anteriorly, with distinct anterior hood; metaventrite without pubescent pits on sides of metaventral process; metaventrite c. 4.0× wider than its length posterior of mesocoxae; median moderately elevated part of metaventrite narrow throughout (Fig. 7D). Aedeagus 0.5 mm long (Fig. 6J). Median lobe rather widely triangular, c. 1.8× longer along midline than wide; apex not reaching apices of parameres; gonopore moderately large, subapical; lateral struts projecting laterad. Parameres distinctly sinuate on lateral margin, slightly expanded subapically; widely meeting each other basally. Phallobase longer than wide. Etymology: The species name is a Latinized adjective derived from the Spanish version of the indigenous Taíno name of the islands of Puerto Rico. Diagnosis: Phaenonotum borinquenum may be distinguished from other Caribbean species by the combination of moderately large eyes, moderately long metaventrite, presence of wings and morphology of the aedeagus. It is extremely similar to P. exstriatum in its external morphology and genital morphology, and may be distinguished from it only by its smaller body size and widely meeting bases of parameres only. Despite the strong morphological similarity, it is not closely related to P. exstriatum. Distribution: Phaenonotum borinquenum is only known from the eastern part of Puerto Rico (Fig. 4A). Phaenonotum delgadoi Deler-Hernández, Cala-Riquelme & Fikáček, 2013 (Figs 4A, 5A, 6A, G, 7A) Type material examined: See Deler-Hernández et al. (2013), including one sequenced paratype from the type locality (molecular voucher MF455). Additional material examined:Cuba: Holguín Prov.: 5 spec. (NMPC): La Melba, Parque Nacional Alexander von Humboldt, 20.43352°N 74.82507°W, 336 m, rainforest litter, 21.ix.2014, R. Anderson, F. Cala-Riquelme, A. Deler-Hernández lgt. (2014-002); 4 spec. (NMPC): 20.45396°N 74.82342°W, 510 m, pluviselva litter, 22.ix.2014, R. Anderson, F. Cala-Riquelme, A. Deler-Hernández lgt. (2014-011); 19 spec. (NMPC): road out La Melba, 20.51524°N 74.81844°W, 407 m, elfin forest litter, 24.ix.2014, R. Anderson, F. Cala-Riquelme, A. Deler-Hernández lgt. (2014-015); 8 spec. (NMPC): road out La Melba, road out of La Melba, 20.59086°N 74.83627°W, 130 m, scrub forest litter, 24.ix.2014, R. Anderson, F. Cala-Riquelme, A. Deler-Hernández lgt. (2014-016). Published records: Cuba: Guantánamo Prov.: El Yunque de Baracoa (Deler-Hernández et al., 2013a). Holguín Prov.: La Melba (Deler-Hernández et al., 2013a). Redescription: Habitus as in Figure 5A. Body length 2.2–2.5 mm (holotype: 2.3 mm). Body oval, strongly convex, elytral suture distinctly elevated posteriorly. Dorsum black, lateral margins of pronotum and elytra without distinct paler stripe; ventral surface brown to dark brown; femora and tibiae reddish; antennae, maxillary palpi and tarsi yellowish. Head with sparse fine punctation, without microsculpture (except posteriorly on frons); eyes small, separated by 5.2× dorsal width of one eye (Fig. 6A). Pronotum with very sparse minute punctures much smaller than on head, interstices without microsculpture. Elytral punctation very strongly impressed, much coarser than on pronotum and head; elytral interstices without microsculpture. Wings completely absent. Mesoventral elevation as wide as metaventral process throughout, not narrowing anteriorly, with distinct anterior hood; metaventrite with a deep pubescent pit on each side of metaventral process; metaventrite c. 5.8× wider than its length posterior of mesocoxae; median moderately elevated part of metaventrite narrow anteriorly, widening posteriorly (Fig. 7A). Aedeagus 0.4 mm long (Fig. 6G). Median lobe rather narrowly triangular, c. 1.9× longer along midline than wide; apex not reaching apices of parameres; gonopore small, subapical; lateral struts simple, not expanded. Parameres indistinctly sinuate, nearly evenly arcuate on lateral margin, not expanded subapically, broadly meeting each other basally. Phallobase slightly wider than long. Diagnosis: Phaenonotum delgadoi may be distinguished from all other Phaenonotum known to us by the combination of extremely coarse elytral punctation, elevated elytral suture and deep pubescent pits on each side of metaventral process. See Identification Key for additional characters. Distribution: Phaenonotum delgadoi is only known from eastern Cuba, all known localities are situated in the Nipe-Sagua-Baracoa mountain system (Fig. 4A). Phaenonotum exstriatum (Say, 1835) (Figs 4B, 5E, 6E, K, 7E) Type material: Neotype of Hydrophilus exstriatus Say, 1835 designated by Smetana (1978) from southeastern states of USA (deposited in Museum of Comparative Zoology, Harvard University, Boston, USA): not examined in our study. Types of Phaenonotum dubium Sharp, 1882 were examined by Deler-Hernández & Fikáček (2016) and confirmed as being conspecific to North American specimens treated as P. exstriatum (synonymy proposed by Smetana, 1978). Material examined:Barbados: 1 spec. (SBPC): Jack-in-Box Gully, forest, 13°11’N 59°34.3’W, 230 m, UV light, 23.vi.2007, S & J. Peck lgt. (07-25). Cuba: Cienfuegos prov.: 1 spec. (NMPC): Soledad, 22.12682°N 80.33289°W, 71 m, MV light, 21.v.2013, A.B.T. Smith lgt. Guantánamo prov.: 2 spec. (NMPC): Baracoa, Jobo Dulce, 20°18′18.00″N 74°27′21.60″W, 94 m, 29.x.2010, R. Correa lgt; El Yunque, 3.2 km SW of campismo, 20°19′N 74°34′W, 150 m, 13.vi.2012, A. Deler-Hernández lgt. Holguín Prov.: 1 spec. (NMPC): Mayarí, Guatemala, Guarina Nuñez, 20°42′4.06″N 75°40′24.45″W, c. 27 m, 26.iii.2013, A. Deler-Hernández lgt. Granma Prov.: 1 spec. (NMPC): Cauto Cristo, Rio Cauto, El Sitio, 20°16′22.80″N 76°29′2.40″W, 135 m, 1.v.2005, L. Chaves. Santiago de Cuba prov.: 9 spec. (NMPC): El Vivero, 1.3 km NEE of Dos Caminos, 20°10′49.36″N 75°46′41.43″W, c. 170 m, at light, 23.v.2013, Deler-Hernández & A. Smith lgt. (1 spec: molecular voucher MF654); 1 spec. (NMPC): Daiquirí, 1–4.vi.1985, S. Bílý lgt. Dominica: 8 spec. (BMNH): La Plaine, 97-67, 22.i.1889, G. A. Ramage lgt. Dominican Republic: 1 spec. (NMPC): Samaná, 6.4 km N of Samaná, road to El Valle, 19°15.78′N 69°20.23′W, 21 m, 5.ix.2014, Deler, Fikáček, Gimmel (DR36). Grenada: 1 spec. (SBPC): Grand Etang Forest Reserve, 12°04.952′N 61°42.162′W, 434 m, nursery edge forest, UV trap, 14.viii.2010, S. Peck lgt. (10–69); 25 spec. (BMNH): Grand Etang (Windward side), 1900 ft, H. H. Smith.; 1 spec. (SBPC): Par. St. Andrew, Mirabeau, Agric. Lab, UV trap, 23.ii.1990, R. E. Woodruff lgt.; 34 spec. (BMNH): Mount Gay Est. (Leeward side), H. H. Smith lgt.; 3 spec. (BMNH): Chantilly Est. (Windward side), H. H. Smith lgt.; 8 spec. (BMNH): Balthazar (Windward side), H. H. Smith lgt.; 1 spec. (BMNH): Vendome Est. (Leeward side), H. H. Smith lgt. Cayman Islands: 1 spec. (BMNH): Grand Cayman, by freshwater lake near George Town., UV light trap, 4.viii.1970, Joy Farradane lgt. Haiti: 1 spec. (BMNH): Port au Prince., 1.iii.1908, M. Cameron lgt. Jamaica: 1 spec. (BMNH): Kingston., 16.ii.1908, M. Cameron lgt.; 9 spec. (SBPC): Ewarton, St. Cath. Par., St. Clair Cave, 27.xii.1972, S & J. Peck. Monserrat: 1 spec. (BMNH): Salem 8.ix.1975, J. Cooter lgt. Saint Lucia: 1 spec. (SBPC): Mon Repos, 6.5 km W of Fox Grovelnn, 13°52.5′N 60°56.4′W, 300 m, submontane forest litter, 22.vii.2007, S & J. Peck lgt. (07-76). Saint Vincent and the Grenadines: 1 spec. (SBPC): Saint Vincent, Brighton Bay Village, 13°07.97′N 61°10.06′W, 1 m, streamside UV trap 8.vi.2007, S & J. Peck lgt. (07-10); 15 spec. (BMNH): Saint Vincent, without additional data, H. H. Smith lgt. Puerto Rico: 7 spec. (NMPC): Naguabo, El Yunque Nat. Forest, S part 3.45 km N of Río Blanco at rd. PR191, 18°14.8′N 65°47.7′W, 170 m, 24.vi.2016, Deler, Fikáček & Seidel lgt. (1 spec.: molecular voucher MF1728). Trinidad & Tobago: 2 spec. (BMNH): Trinidad, St. Augustine, 15.iv.1926, C. L. Withycombe lgt.; 1 spec. (BMNH): Trinidad, St. Margarita, 11.vi.1942. E. C. Humphries lgt. USA: Florida: 1 spec. (NMPC): Highlands Co., Venus, 4 miles W of Fish Eating Creek, 15.viii.1965, W. Suter lgt., det. A. Smetana; 1 spec. (NMPC): Alachua Co., Gainesville, black light, 19.vii.1978, F. N. Young lgt., det. A. Smetana; 1 spec. (NMPC): Kansas: Douglas Co., Bridenthal Ecological Reserve, 3.2 miles N of Baldwin City, 38.81043°N 95.18669°W, 270 m, ix.2009, Eldgedge & Fikáček lgt.; 1 spec. (NMPC): Douglas Co., Lawrence Baker Wetlands, 27.viii.2009, Gustafson, Eldgedge & Fikáček lgt.; 1 spec. (NMPC): Delaware: New Castle Co., Frenchtown Woods Natural Area, 23.v.2004, at light, A. E. Z. Short lgt. (AS-04-065) (molecular voucher MF1063). Costa Rica: 1 spec. (NMPC): Guanacaste: 6.6 km from main road, roadside gravel stream running through culvert, 10°09′26.7″N 85°22′47.5″W, 50 m, 13.i.2004, Short & Lebbin lgt. (AS-04-037) (molecular voucher MF1738). Redescription: Habitus as in Figure 5E. Body length 3.3–4.0 mm. Body elongate oval, moderately convex, elytral suture not elevated posteriorly. Dorsum black, lateral margins of elytra with very narrow indistinct paler stripe, pronotum paler in posterolateral corners; ventral surface dark brown to black; femora and tibiae black; antennae, maxillary palpi and tarsi yellowish. Head with sparse moderately coarse punctation, without microsculpture (except posteriorly on frons); eyes moderately large, separated by 4.0× dorsal width of one eye (Fig. 6E). Pronotum with sparse fine punctures, slightly smaller than on head, interstices without microsculpture. Elytral punctation moderately impressed, coarser than on pronotum and head; elytral interstices without microsculpture. Wings present, well developed. Mesoventral elevation as wide as metaventral process posteriorly, slightly narrowing anteriorly, with distinct anterior hood; metaventrite without pubescent pits on sides of metaventral process; metaventrite c. 3.9× wider than its length posterior of mesocoxae; median moderately elevated part of metaventrite moderately wide throughout (Fig. 7E). Aedeagus 0.4 mm long (Fig. 6K). Median lobe narrowly triangular, c. 2.1× longer along midline than wide; apex reaching apices of parameres; gonopore small, subapical; lateral struts expanded laterally. Parameres distinctly sinuate on lateral margin, slightly expanded subapically. Phallobase longer than wide. Diagnosis: Phaenonotum exstriatum may be distinguished from Caribbean species except P. borinquenum by the combination of moderately large eyes, moderately long metaventrite, presence of wings and morphology of the aedeagus. It may be distinguished from P. borinquenum by larger body size and bases of parameres meeting in single point only. Despite the strong morphological similarity, it is not closely related to P. borinquenum (see Fig. 2). Distribution: Phaenonotum exstriatum is widespread in the eastern USA and in Central America (Smetana, 1978; Deler-Hernández & Fikáček, 2016) as well as in Greater and Lesser Antilles (Fig. 4B). The only record from South America is from eastern Colombia Gonzáles-Rodríguez, García-Hernández & Clarkson (2017). but the species is probably widespread in northern South America as it is common in southern islands of Lesser Antilles and it also occurs in Trinidad. Phaenonotum laevicolle Sharp, 1882 complex (Figs 4C, 5F, 6F, L, 7F) Type material examined: Types of P. laevicolle Sharp, 1882, see Deler-Hernández & Fikáček (2016). Material examined:Cuba: Cienfuegos Prov.: 1 male (NMPC): Río Cabagan, Gruta Mengoa, 21.93123°N 80.08461°W, 651 m, 20.v.2014, A. Deler-Hernández lgt. (molecular voucher MF 1115); 3 spec. (ZMHB): Sierra del Escambrai, 5 km N Topes de Collantes, c. 600 m, sifting of leafs, 17.xii.2007, M. Schülke lgt. (CU7-4). Sancti Spíritus Prov.: 9 spec. (ZMHB): Sierra del Escambrai, Topes de Collantes, c. 700 m, Streu, Totholz, 17.xii.2007, M. Schülke lgt. (CU7-3). Granma Prov.: 7 spec. (ZMHB): Sierra Maestra, La Habanita, 35 km NE Pilón, 900–1000 m, sifted hay, 20.xii.2007, M. Schülke lgt. (CU7-8). Santiago de Cuba Prov.: 1 spec. (NMPC): El Vivero, 1.6 km E of Dos Caminos, 20°10.8′N 75°46.4′W, 150 m, 20–21.vi.2012, A. Deler-Hernández & Fikáček (MF18). Haiti: 3 spec. (BMNH): Port au Prince, 1.iii.1908, M. Cameron lgt. Venezuela: 1 male (NMPC): Monagas small pond between Morichal Largo & Temblador, 9°05′47.9″N 62°43′37.1″W, 29 m, 2.ii.2010, Short, García & Joly lgt. (VZ10-0202-03A) (molecular voucher MF1740). Redescription: (refers to the Caribbean specimens examined): Habitus as in Figure 5F. Body length 2.7–3.1 mm. Body elongate oval, moderately convex, elytral suture not elevated posteriorly. Dorsum black, lateral margins of elytra with paler stripe reaching subapically, pronotum paler in posterolateral corners; ventral brown to dark brown; femora dark brown to reddish, tibiae reddish; antennae, maxillary palpi and tarsi yellowish. Head with sparse fine punctation, without microsculpture (except posteriorly on frons); eyes moderately large, separated by 3.6× dorsal width of one eye (Fig. 6F). Pronotum with sparse fine punctures similar to that on head, interstices without microsculpture. Elytral punctation moderately impressed, coarser than on pronotum and head; elytral interstices without microsculpture. Wings present, well developed. Mesoventral elevation slightly narrower than metaventral process posteriorly, slightly narrowing anteriorly, with distinct anterior hood; metaventrite without pubescent pits on sides of metaventral process; metaventrite c. 3.5× wider than its length posterior of mesocoxae; median moderately elevated part of metaventrite moderately wide throughout (Fig. 7F). Aedeagus 0.5 mm long (Fig. 6L). Median lobe rather triangular, c. 1.7× longer along midline than wide; apex reaching apices of parameres; gonopore large, wide, situated in apical third of median lobe; lateral struts very shortly expanded laterally. Parameres distinctly sinuate on lateral margin, strongly expanded subapically, broadly meeting each other basally. Phallobase longer than wide. Comments: Both sequenced specimens (MF1115 from Cuba and MF1740 from Venezuela) form a strongly supported clade in the molecular analysis and are evidently closely related (pairwise distance of cox1 sequences is 4.8%). Both specimens correspond with each other in external morphology and the characteristic shape of male genitalia, and only differ in body size: the Cuban specimen is smaller (2.8 mm), the Venezuelan specimen larger (3.4 mm). In this aspect, the sequenced Cuban specimen corresponds to all additional Greater Antillean specimens examined, which are also rather small (2.7–3.1 mm). Hence, it seems probable that Venezuelan specimen is not conspecific with the Greater Antillean ones, but additional material is necessary to evaluate the intraspecific genetic and morphological variability properly to decide whether the sequenced specimens represent one or two species. The external morphology, body size and the morphology of genitalia of the Caribbean specimens correspond to the types of P. laevicolle Sharp, 1882 described from Guatemala and examined by Deler-Hernández & Fikáček (2016). However, due to the absence of DNA-grade specimens of P. laevicolle from Central America, we are unable to determine whether the Greater Antillean specimens are conspecific. For that reason, we consider all above specimens as members of the Phaenonotum laevicolle species complex whose taxonomy needs to be addressed once more material will be available. Distribution: The Caribbean specimens of the Phaenonotum laevicolle complex are known from central and eastern Cuba and western Hispaniola (Haiti). Outside the Caribbean, the species complex clearly occurs in southernmost Northern and Central Americas (types of P. laevicolle) and in northern South America [as P. globulosum (Mulsant, 1844) from Colombia (Hansen, 1999; not examined by us) and the Venezuelan specimen sequenced by us]. The records from Argentina (types of P. spegazziinii Bruch, 1915 not examined by us) seem doubtful (Oliva et al., 2002) and are not considered here. Based on these sources, we estimate the distribution of the species complex to be as shown in Figure 4C. Phaenonotum laterale sp. nov. (Figs 4A, 5C, 6C, I, 7B) Type locality: Dominican Republic, Barahona, Monumento Natural Miguel Domingo Fuerte ‘Cachote’ 18°5.91′N 71°11.35′W, 1188 m. Type material: Holotype: male (NMPC): ‘Dominican Republic: Barahona, MN Domingo Fuerte “Cachote,” 18°5.91′N 71°11.35′W, 1188 m, 14.viii.2014, Deler, Fikáček, Gimmel DR03/montane broadleaf cloud forest with numerous Cyathea: sifting of thin layer of wet leaf litter and mosses’. Paratypes: Dominican Republic: 10 spec. (NMPC): same data as holotype (including molecular voucher MF1013); 25 spec. (CMN, NMPC, MNHNSD, NHMW): MN Domingo Fuerte ‘Cachote’, 18°5.21′N 71°11.46′W, 1205 m, sparse montane cloud forest with ferns and mosses here and there, sifting of thin layer of leaf litter, 14.viii.2014, M. Fikáček lgt. (DR03a); 49 spec. (BMNH, DZRJ, MCZ, MNHNSD, NMPC, SBMN): MN Domingo Fuerte ‘Cachote’, 18°4.48–5.37′N 71°11.03–11.54′W, 1188 m, 14.viii.2014, secondary montane broadleaf forest with sparse ferns and moss, sifting of leaf litter, Deler-Hernández, Fikáček & Gimmel lgt. (DR04). Description: Habitus as in Figure 5C. Body length 2.5–2.7 mm (holotype: 2.7 mm). Body oval, moderately convex, elytral suture slightly elevated posteriorly. Dorsum brown to dark brown, lateral margins of pronotum and lateral and sutural margins of elytra with distinct yellowish stripe; ventral surface brown to reddish brown; femora and tibiae reddish; antennae, maxillary palpi and tarsi yellowish. Head with sparse fine punctation, without microsculpture (except posteriorly on frons); eyes small, separated by 5.1× dorsal width of one eye (Fig. 6C). Pronotum with very sparse and very fine punctures much smaller than on head, interstices with not very distinct mesh-like microsculpture. Elytral punctation sparse and moderately impressed, similar to that on head, much coarser than on pronotum; elytral interstices without microsculpture. Wings completely absent. Mesoventral elevation extremely narrow, as wide as metaventral process throughout, not narrowing anteriorly, without distinct anterior hood; metaventrite without deep pubescent pits at sides of metaventral process; metaventrite c. 5.0× wider than its length posterior of mesocoxae; median weakly elevated part of metaventrite narrow throughout (Fig. 7B). Aedeagus 0.35 mm long (Fig. 6I). Median lobe rather narrowly triangular, c. 2.1× longer along midline than wide; apex reaching apices of parameres; gonopore small, subapical; lateral struts simple, not expanded. Parameres indistinctly sinuate on lateral margin, not expanded subapically, widely meeting each other basally. Phallobase as long as wide. Etymology: The species name refers to the conspicuous yellow stripe along lateral margins of pronotum and elytra characteristic of this species. Diagnosis: Phaenonotum laterale may be distinguished from other Caribbean species, based on its brown-yellow coloration, obsolete pronotal punctation and extremely narrow meso-metaventral keel. See Identification Key for further diagnostic characters. Distribution: Phaenonotum laterale sp. nov. is only known from a high altitude cloud forest region in the southwestern Dominican Republic, geologically situated on the southern paleoisland of Hispaniola (Fig. 4A). Phaenonotum ondreji sp. nov. (Figs 4A, 5B, 6B, H, 7C) Type locality: Jamaica, Blue Mountains National Park, 18°5′14.13″N 76°43′37.55″W, 1279 m. Type material: Holotype: male (NMPC): Jamaica: Blue Mountain, 18°5′14.13″N 76°43′37.55″W, 1279 m, 14.xi.2013, F. Cala-Riquelme leg.’. Paratypes: Jamaica: 19 spec. (CMN, MCZ, NMPC, NHMW) same data as holotype (including molecular voucher MF980); 2 spec. (BMNH): Newcastle, 19–22.viii.1908, M. Cameron lgt. Description: Habitus as in Figure 5B. Body length 3.2–3.4 mm (holotype: 3.1 mm). Body oval, highly convex, elytral suture slightly elevated posteriorly. Dorsum dark brown to black, posterolateral corners of pronotum and apical part of lateral margins of elytra distinctly yellowish; ventral surface brown to reddish brown; femora and tibiae yellowish to reddish; antennae, maxillary palpi and tarsi yellowish. Head with sparse moderately coarse punctation, with mesh-like microsculpture; eyes small, separated by 6.9× dorsal width of one eye (Fig. 6B). Pronotum with sparse and fine punctures, smaller than on head, interstices with distinct mesh-like microsculpture. Elytral punctation sparse, moderately impressed, coarser than on head and pronotum; elytral interstices without microsculpture. Wings completely absent. Mesoventral elevation very narrow, as wide as metaventral process throughout, not narrowing anteriorly, without distinct anterior hood; metaventrite without deep pubescent pits at sides of metaventral process; metaventrite c. 5.1× wider than its length posterior of mesocoxae; median highly elevated portion narrow throughout (Fig. 7C). Aedeagus 0.6 mm long (Fig. 6H). Median lobe very narrowly triangular, c. 2.3× longer along midline than wide; apex slightly overlapping apices of parameres; gonopore small, subapical; lateral struts weakly expanded. Parameres sinuate on lateral margin, weakly expanded subapically, narrowly meeting each other basally. Phallobase as long as wide. Etymology: This species is named in honour of Ondřej Jelínek, a good friend of the senior author, in appreciation of his friendship. Diagnosis: Phaenonotum ondreji may be distinguished from other Caribbean species by the combination of small eyes, head and pronotum with distinct mesh-like microsculpture, and narrow highly elevated meso-metaventral keel. See Identification Key for additional diagnostic characters. Distribution: Phaenonotum ondreji sp. nov. is only known from the eastern part of the island (Blue Mountains range, Fig. 4A). DISCUSSION Biogeography of Phaenonotum and origin of Caribbean species Caribbean lineages of Phaenonotum diverged from their sister taxa during the middle Eocene to early Pleistocene (c. between 47 and 2.4 Mya), that is long after the bolide impact to the Caribbean region at the Cretaceous-Paleogene (K-Pg) boundary c. 65.5 Mya (Schulte et al., 2010). This is incongruent with the vicariance hypothesis of Rosen (1975, 1985), which assumes that Caribbean lineages originated by the colonization of the proto-Antillean volcanic arc in the Late Cretaceous and survived the K-Pg boundary. The Cenozoic origin of Phaenonotum lineages however corresponds to other insect and spider groups studied so far, of which Caribbean clades were dated to originate between the Eocene and the Miocene (Wahlberg, 2006; Wahlberg & Freitas, 2007; Ceccarelli & Zaldívar-Riverón, 2013; Zhang & Maddison, 2013; Matos-Maraví et al., 2014; McHugh et al., 2014; Dziki et al., 2015; Lewis et al., 2015; Rodriguez et al., 2015; Zhang et al., 2017). The Caribbean Phaenonotum fauna originated from three to four independent colonization events from northern South America, the earliest of which correspond(s) in timing to the timespan of the GAARlandia land bridge connecting northern South America and the Greater Antilles through the emerged Aves Ridge at the Eocene–Oligocene boundary (Iturralde-Vinent & MacPhee, 1999; Iturralde-Vinent, 2006). Subsequent colonizations happened during the late Oligocene to early Miocene after the Antillean archipelago lost direct connection to the continent (Iturralde-Vinent & MacPhee, 1999; Iturralde-Vinent, 2006). Hence, the evolutionary history of Phaenonotum may be best explained by the combination of colonization via a land bridge in some lineages and the over-water dispersal events (i.e. founder events) in others. The oldest colonization of Phaenonotum to the Caribbean gave rise to a small clade today containing three single-island endemics: P. delgadoi from eastern Cuba, P. ondreji from eastern Jamaica and P. laterale from southern Hispaniola. The divergence of this clade from modern continental clades slightly predated the timespan of the GAARlandia land bridge, which allows for two possible scenarios of colonizations: single colonization of the Greater Antilles via GAARlandia and subsequent split into multiple species in the archipelago, or parallel colonizations of the Greater Antilles via GAARlandia by several species of the clade and subsequent speciation in the archipelago (our BioGeoBEARS analyses propose both scenarios as possible, depending on the model and time-stratification used and whether a founder event is allowed or not). Naturally, we cannot totally exclude the over-water dispersal in case it coincided by age or slightly predated the timespan of the land bridge (Poux et al., 2006; de Queiroz, 2016). However, unlike all other Phaenonotum examined, all three species of this clade are wingless, indicating that their MRCA probably had no metathoracic wings either. The expected limited dispersal abilities of the members of the clade together with the time overlap of their arrival to the Greater Antilles with the timespan of the GAARlandia land bridge make the scenario of the over-land colonization of the Greater Antilles more probable than the over-water dispersal alternative. The divergence of the Cuban P. delgadoi from the remaining two species of the clade was dated to the Eocene–Oligocene boundary (c. 36 Mya), coinciding with the maximum landspan of the GAARlandia and pre-dating the subdivision of the Greater Antilles into particular palaeoislands (separation of Puerto Rico from Hispaniola + Cuba by Mona Passage c. 30–20 Mya and separation of eastern Cuba from central Hispaniola by Windward Passage c. 17–14 Mya; Iturralde-Vinent & MacPhee, 1999; Matos-Maraví et al., 2014). Both clades may have split while colonizing separate mountain ranges of the GAARlandia peninsula (Iturralde-Vinent & MacPhee, 1999; Iturralde-Vinent, 2006) or before colonizing the peninsula. In contrast, the divergence of the Hispaniolan P. laterale and Jamaican P. ondreji was dated to the Late Oligocene (c. 26 Mya) when only the Blue Mountain range of Jamaica was emergent and widely isolated from other islands (Iturralde-Vinent & MacPhee, 1999; Iturralde-Vinent, 2006). This implies a long-distance over-water founder event, improbable for wingless mountain species. Interestingly, Jamaica-Hispaniola divergences of similar age are also found in some other animal groups (c. 22 Mya in Osteopilus tree frogs: Moen & Wiens, 2009; c. 20 Mya in Exophthalmus weevils: Zhang et al., 2017; c. 15 Mya in Calisto butterflies: Matos-Maraví et al., 2014). This may indicate either much closer geological connection between Hispaniola and Jamaica in the Late Oligocene to Early Miocene, or a constellation largely facilitating Hispaniola to Jamaica dispersal (e.g. by strong sea currents or hurricanes; Hedges, 2006). Puerto Rico was a part of the GAARlandia land bridge, but no species of the above Phaenonotum clade has been recorded from the island, in contrast to the predictions of the over-land GAARlandia scenario. Similarly, the endemic occurrence of the Hispaniolan P. laterale in the southern part of the island that emerged in the Late Miocene, largely postdating the origin of the species (26.4 Mya, Late Oligocene) also disagrees with the current model of historical geography of the Greater Antilles. This indicates that range shifts and extinctions probably played a significant role in the evolutionary history of Phaenonotum beetles in the Caribbean. On the other hand, the absence of the members of this Phaenonotum clade from the Lesser Antilles is congruent with the GAARlandia hypothesis: older Lesser Antilles islands are of c. Oligocene to Miocene origin, the younger ones arose during the Pliocene only and none of them were part of the GAARlandia land bridge, which was situated slightly more to the west in the place of the nowadays submerged Aves Ridge (Iturralde-Vinent & MacPhee, 1999; Thorpe et al., 2004; Iturralde-Vinent, 2006). The remaining Phaenonotum clades probably colonized the Caribbean by two or three independent events during the Late Oligocene to Pliocene by over-water dispersal. One of these events (dated to Early Miocene, c. 19 Mya) gave rise to the Puerto Rican P. borinquenum, that is the only single-island endemic, which is not a member of the Caribbean endemic lineage discussed above. Both widespread continental-Caribbean species (P. laevicolle and P. exstriatum) were estimated to originate in the Caribbean and back-colonized the continent during the Pliocene and Pleistocene in the time-stratified BioGeoBEARS analyses. Our ancestral range estimates may be, however, strongly influenced by incomplete sampling of continental Phaenonotum species and limited knowledge on their distribution, and additional studies are hence necessary to corroborate our results. Nevertheless, our data suggest that the Caribbean Region may act not only as a sink of colonizations from the continent, but also a source of taxa colonizing North and Central America. This corresponds to the situation recently revealed for the Exophthalmus weevils and for Heracleidos butterflies, both of which colonized Central America from the Caribbean during the Miocene to Pleistocene (Lewis et al., 2015; Zhang et al., 2017); multiple examples are also known for vertebrates (Bellemain & Ricklefs, 2008; Ricklefs & Bermingham, 2008). Moreover, Central America was an archipelago similar to todayʼs Greater Antilles during the Miocene to Pliocene, and Caribbean species hence probably colonized islands with unbalanced fauna rather than a continent with diverse well-established fauna characterized by strong interspecific competition (Bellemain & Ricklefs, 2008). All four single-island endemic Phaenonotum species inhabit leaf litter in lowland (P. delgadoi) or montane cloud forests (P. ondreji, P. laterale and P. borinquenum), the primary forest habitats in the island interior. In contrast, the remaining two species are more widespread (P. exstriatum is recorded from all Greater Antillean islands including the Cayman Islands, and from most islands of the Lesser Antilles, while P. laevicolle is known from Cuba and Hispaniola; Fig. 4), and inhabit disturbed lowland habitats (P. exstriatum aquatic and semiaquatic habitats and decaying plant material, P. laevicolle decaying plant material including hay piles and leaf litter in farmland). When biology and distribution data are correlated to the age of the respective species, they are in a good agreement with the taxon cycle hypothesis proposed originally by Wilson (1959, 1961) for Melanesian ants and subsequently confirmed, for example, for Lesser Antillean birds (Ricklefs & Cox, 1972) (see Ricklefs & Bermingham, 2002 for review and multiple examples, and Economo & Sarnat, 2012 for detailed re-analysis of taxon cycle in Melanesian ant fauna). Following the predictions of the taxon cycle, the recently colonizing species (P. laevicolle and P. exstriatum) inhabit wider spectrum of lowland disturbed habitats and are good dispersers (range expansion phase), whereas the locally endemic species are remnants of ancient colonizations inhabiting exclusively inland primary habitats and having poor dispersal abilities (they switched from the expansion phase at the time of colonization into the range contraction phase; this may be followed by another range expansion phase or by extinction). Species-level systematics of Phaenonotum Despite the beetles being frequently collected, the systematics of Phaenonotum beetles has not been studied properly until now, and very little is known about the continental fauna of the genus. Sixteen species are known from the continental Neotropics and two from the Nearctic (with the widespread P. exstriatum inhabiting both regions). The real species diversity seems to be much higher, which is illustrated, for example, by the fact that of the c. 13 continental species included in our analysis, we were only able to identify two of them (P. exstriatum and P. laevicolle), whereas remaining ones probably represent undescribed species. The extreme morphological similarity of the species and limited number of characters useful for species-level identification are clearly the main obstacles for species-level studies and identification of Phaenonotum. Moreover, extreme similarity in external morphology and male genitalia does not necessarily indicate a close phylogenetic relationship between species. This is illustrated by P. exstriatum and P. borinquenum sp. nov., which are not closely related, yet are nearly identical in morphological characters. In contrast, the P. laevicolle complex seems closely related to P. borinquenum but can be very easily distinguished from it by genital morphology (compare Fig. 6L, J). The combination of morphology and molecular data seems hence necessary for proper understanding of species limits even in distantly related species in Phaenonotum. The same is naturally true for closely related similar species. This is the reason why we refrain from describing the Cuban and Hispaniolan species of the P. laevicolle complex and do not decide whether the sequenced specimens from Cuba and Venezuela belonging to this species complex represent one or two species. Sequences of freshly collected specimens from the type locality of P. laevicolle or its surroundings and from multiple specimens across the range of the species complex would be necessary to analyse the species limits in this case. Neotropical Coelostomatini and the position of Phaenonotum Despite limited taxon sampling, our analyses indicate that the Neotropical endemic genera of Coelostomatini (Cyclotypus, Phaenonotum, Phaenostoma, Lachnodacnum) form a monophyletic group. This result corresponds with topology revealed in the phylogenetic analysis containing a wider spectrum of coelostomatine genera but fewer Neotropical taxa (V. Sýkora, unpubl. data) and suggests that the early evolution of the Coelostomatini was strongly influenced by paleogeography: the origin of this clade was revealed as Early-Middle Cretaceous in our analysis when South America was largely isolated from other continents. The internal topology of the Neotropical clade disagrees with current generic concepts, which are based on the combination of a few easy-to-observe morphological characters (Hansen, 1991; Clarkson et al., 2014): Phaenonotum is revealed polyphyletic when comprising the morphologically aberrant P. caribense, and the bromeliad inhabiting genus Lachnodacnum is nested within the polyphyletic Phaenostoma. The fact that the aberrant morphology corresponds to a separate phylogenetic position in one taxon (Phaenonotum caribense) but to biology rather than phylogenetic position in another (Lachnodacnum) clearly indicates that solving the systematics of the Neotropical clade of Coelostomatini (and in fact of the tribe as a whole) is a complex task. Additional studies combining morphological and molecular characters are necessary to understand the higher-level systematics of the group and to recognize ecology-based morphological characters from phylogeny-informative ones. Additional taxa also need to be included in the analyses – this especially concerns Hydroglobus puncticollis (Bruch, 1915) from Argentina, which probably belongs to the Neotropical clade. The clade of the core Phaenonotum (i.e. Phaenonotum after excluding P. caribense) corresponds well with the current morphology-based concept of the genus (Hansen, 1991), with African P. africanum, whose assignment to the genus is doubtful and needs to be tested, being the only exception. Based on our analyses, the core Phaenonotum are either sister group to Phaenostoma + Lachnodacnum (Fig. 1A) or to just the Phaenostoma posticatum clade (Fig. 1B). The latter topology seems to be better supported by morphology, as both Phaenostoma posticatum and the core Phaenonotum species share the unique shape of the metanepisternum (widened in posterior third, narrowing anteriorly, Fig. 7); all remaining coelostomatine species examined including Phaenostoma kontax, Lachnodacnum, Phaenonotum caribense and Cyclotypus have the metanepisternum with lateral margins parallel-sided (and hence the same width) throughout. However, the undescribed Phaenonotum species from Peru (MF845) bears the parallel-sided rather than anteriorly narrowing metanepisternum despite being strongly supported as a member of the core Phaenonotum clade in our analyses. Hence, the phylogenetic significance of this character needs to be further tested. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Supplement S1. Molecular data and phylogenetic and divergence dating analyses. Supplement S2. Imput data and results of the BioGeoBEARS analyses. Supplement S3. Complete data of all examined specimens. [Version of Record, published online 15 December 2017; http://zoobank.org/urn:lsid:zoobank.org:pub:9FBFC95E-DB5D-42C8-BD41-C3921F1016E3] ACKNOWLEDGEMENTS We are grateful to the authorities and personnel of the Cuban Ministry of Science, Technology and Environment (CICA-CITMA) and the Empresa Nacional para la Protección de la Flora y la Fauna (ENPFF) for providing access to protected areas under their control; to Carlos Suriel, Edda Martínez Caléz and Marilyn Colon Negron for their help with arranging the collecting permits in the Dominican Republic and Puerto Rico; to Robert Anderson (Canadian Museum of Nature, Ottawa), Matthew L. Gimmel (Santa Barbara Museum of Natural History) and people of the Cuba and Jamaica CarBio teams for their help during our fieldwork; to S. Peck (Carleton University) and curators of the institutional collections (see Material and methods) for the possibility to examine their collections; and to Andrew Short (University of Kansas, Lawrence, USA), Joseph Parker (Columbia University, New York) and Bruno Clarkson (Universidade Federal do Rio de Janeiro) for providing continental specimens used in our analyses. We specially acknowledge the comments and critique of Andrew Short and two anonymous reviewers on the earlier version of this draft, and Robert Angus (BMNH) for a careful correction of our English text. The study was supported by the grant SVV 260 434/2017 at the Department of Zoology, Charles University and the Rufford Foundation (15144-1) to A.D.H. and by the Ministry of Culture of the Czech Republic (DKRVO 2017/14, National Museum, 00023272) to M.F. REFERENCES Agnarsson I , LeQuier SM , Kuntner M , Cheng R-C , Coddington JA , Binford G . 2016 . Phylogeography of a good Caribbean disperser: Argiope argentata (Araneae, Araneidae) and a new ‘cryptic’ species from Cuba . ZooKeys 625 : 25 – 44 . Google Scholar CrossRef Search ADS Ali JR . 2012 . Colonizing the Caribbean: is the GAARlandia land-bridge hypothesis gaining a foothold ? Journal of Biogeography 39 : 431 – 433 . Google Scholar CrossRef Search ADS Alonso R , Crawford AJ , Bermingham E . 2012 . Molecular phylogeny of an endemic radiation of Cuban toads (Bufonidae: Peltophryne) based on mitochondrial and nuclear genes . Journal of Biogeography 39 : 434 – 451 . Google Scholar CrossRef Search ADS Archangelsky M . 1989 . A new species of the genus Phaenonotum Sharp, 1882 (Coleoptera, Hydrophilidae, Sphaeridiinae) from Venezuela . Aquatic Insects 4 : 217 – 220 . Arriaga-Varela E , Seidel M , Deler-Hernández A , Senderov V , Fikáček M . 2017 . A review of the Cercyon (Coleoptera: Hydrophilidae: Sphaeridiinae) of the Greater Antilles . ZooKeys 681 : 39 – 93 . Google Scholar CrossRef Search ADS Bartolini C , Lang H , Spell T . 2003 . Geo-chronology, geochemistry, and tectonic setting of the Mesozoic Nazas Arc in north-central Mexico, and its continuation to northern South America . In: Bartolini C , Buffler RT , Blickwede JF , eds. The Circum-Gulf of Mexico and the Caribbean: hydrocarbon habitats, basin formation, and plate tectonics . Tulsa: American Association of Petroleum Geologists Memoirs 79 : 427 – 461 . Bellemain E , Ricklefs RE . 2008 . Are islands the end of the colonization road ? Trends in Ecology and Evolution 23 : 461 – 468 . Google Scholar CrossRef Search ADS Bloom DD , Fikáček M , Short AE . 2014 . Clade age and diversification rate variation explain disparity in species richness among water scavenger beetle (Hydrophilidae) lineages . PLoS ONE 9 : e98430 . Google Scholar CrossRef Search ADS Bocák L , Barton C , Crampton-Platt A , Chesters D , Ahrens D , Vogler AP . 2014 . Building Coleoptera tree-of-life for >8000 species: composition of public DNA data and fit with Linnean classification . Systematic Entomology 39 : 97 – 110 . Google Scholar CrossRef Search ADS Bouckaert R , Heled J , Kühnert D , Vaughan T , Wu C-H , Xie D , Suchard MA , Rambaut A , Drummond AJ . 2014 . BEAST 2: a software platform for Bayesian evolutionary analysis . PLoS Computational Biology 10 : e1003537 . Google Scholar CrossRef Search ADS Ceccarelli FS , Zaldívar-Riverón A . 2013 . Broad polyphyly and historical biogeography of the Neotropical wasp genus Notiospathius (Braconidae: Doryctinae) . Molecular Phylogenetics and Evolution 69 : 142 – 152 . Google Scholar CrossRef Search ADS Claramount S , Cracraft J . 2015 . A new time tree reveals Earth history’s imprint on the evolution of modern birds . Science Advances 1 : e1501005 . Google Scholar CrossRef Search ADS Crews SC , Gillespie RG . 2010 . Molecular systematics of Selenops spiders (Araneae: Selenopidae) from North and Central America: implications for Caribbean biogeography . Biological Journal of the Linnean Society 101 : 288 – 322 . Google Scholar CrossRef Search ADS Deler-Hernández A , Cala-Riquelme F , Fikáček M . 2013a . Description of a new species of Phaenonotum from eastern Cuba (Coleoptera: Hydrophilidae: Sphaeridiinae) . Acta Entomologica Musei Nationalis Pragae 53 : 615 – 622 . Deler-Hernández A , Cala-Riquelme F , Fikáček M . 2014 . A review of the genus Oosternum Sharp of the West Indies (Coleoptera: Hydrophilidae: Sphaeridiinae) . Deutsche Entomologische Zeitschrift 61 : 43 – 63 . Google Scholar CrossRef Search ADS Deler-Hernández A , Fikáček M . 2016 . Redescriptions and lectotype designations of Central American species of Phaenonotum Sharp (Coleoptera, Hydrophilidae) based on the type material from the David Sharp collection . ZooKeys 579 : 83 – 98 . Google Scholar CrossRef Search ADS Deler-Hernández A , Fikáček M , Cala-Riquelme F . 2013b . A review of the genus Berosus Leach of Cuba (Coleoptera: Hydrophilidae) . ZooKeys 273 : 73 – 106 . Google Scholar CrossRef Search ADS Dziki A , Binford GJ , Coddington JA , Agnarsson I . 2015 . Spintharus flavidus in the Caribbean – a 30 million year biogeographical history and radiation of a ‘widespread species’ . PeerJ 3 : e1422 . Google Scholar CrossRef Search ADS Economo EP , Sarnat EM . 2012 . Revisiting the ants of Melanesia taxon cycle: historical and human-mediated invasions of a tropical archipelago . American Naturalist 180 : E1 – E16 . Google Scholar CrossRef Search ADS Esposito LA , Bloom T , Caicedo L , Alicia-Serrano A , Sanchez-Ruiz J , May-Collado LJ , Binford G , Agnarsson I . 2015 . Islands within islands: diversification of tailless whip spiders (Amblypygi, Phrynus) in Caribbean caves . Molecular Phylogenetics and Evolution 93 : 107 – 117 . Google Scholar CrossRef Search ADS Foley NM , Springer MS , Teeling EC . 2016 . Mammal madness: is the mammal tree of life not yet resolved ? Philosophical Transactions of the Royal Society B: Biological Sciences 371 : 20150140 . Google Scholar CrossRef Search ADS González-Rodríguez LM , García-Hernández AL , Clarkson B. 2017 . First records of water scavenger beetle species (Coleoptera, Hydrophilidae) from Quindío Department, Colombia . Check List 13 : 605 - 620 . Google Scholar CrossRef Search ADS Hansen M . 1991 . The hydrophiloid beetles. phylogeny, classification and a revision of the genera (Coleoptera, Hydrophiloidea) . Biologiske Skrifter 40 : 1 – 368 . Hansen M . 1999 . World catalogue of insects. Volume 2. Hydrophiloidea (s. str.) (Coleoptera) . Stenstrup : Apollo Books , 1– 416 pp. Hedges SB . 2006 . Paleogeography of the Antilles and the origin of West Indian terrestrial vertebrates . Annals of the Missouri Botanical Garden 93 : 231 – 244 . Google Scholar CrossRef Search ADS Hedges SB , Hass CA , Maxson LR . 1992 . Caribbean biogeography: molecular evidence for dispersal in West Indian terrestrial vertebrates . Proceedings of the National Academy of Sciences of the United States of America 89 : 1909 – 1913 . Google Scholar CrossRef Search ADS Huelsenbeck JP , Larget B , Alfaro ME . 2004 . Bayesian phylogenetic model selection using reversible jump Markov chain Monte Carlo . Molecular Biology and Evolution 21 : 1123 – 1133 . Google Scholar CrossRef Search ADS Iturralde-Vinent MA . 1982 . Aspectos geológicos de la biogeografía de Cuba . Ciencias de la Tierra y el Espacio 5 : 85 – 100 . Iturralde-Vinent MA . 2006 . Meso-Cenozoic Caribbean paleogeography: implications for the historical biogeography of the region . International Geology Review 48 : 791 – 827 . Google Scholar CrossRef Search ADS Iturralde-Vinent MA , MacPhee RDE . 1999 . Paleogeography of the Caribbean region: implications for Cenozoic biogeography . American Museum of Natural History Bulletin 238 : 1 – 95 . Landis MJ , Matzke NJ , Moore BR , Huelsenbeck JP . 2013 . Bayesian analysis of biogeography when the number of areas is large . Systematic Biology 62 : 789 – 804 . Google Scholar CrossRef Search ADS Lanfear R , Calcott B , Ho SYW , Guindon S . 2012 . PartitionFinder: combined selection of partitioning schemes and substitution models for phylogenetic analyses . Molecular Biology and Evolution 29 : 1695 – 1701 . Google Scholar CrossRef Search ADS Lawrence JF , Slipinski A . 2014 . Australian beetles volume 1: morphology, classification, and keys . Collingwood : CSIRO Publishing , 1– 576 pp. Lewis DS , Sperling FAH , Nakahara S , Cotton AM , Kawahara AY , Condamine FL . 2015 . Role of Caribbean Islands in the diversification and biogeography of Neotropical Heraclides swallowtails . Cladistics 31 : 291 – 314 . Google Scholar CrossRef Search ADS Matos-Maraví P , Núñez Águila R , Peña C , Miller JY , Sourakov A , Wahlberg N . 2014 . Causes of endemic radiation in the Caribbean: evidence from the historical biogeography and diversification of the butterfly genus Calisto (Nymphalidae: Satyrinae: Satyrini) . BMC Evolutionary Biology 14 : 199 . Google Scholar CrossRef Search ADS Matzke NJ . 2013 . Probabilistic historical biogeography: new models for founder-event speciation, imperfect detection, and fossils allow improved accuracy and model-testing . Frontiers of Biogeography 5 : 242 – 248 . Matzke NJ . 2014 . Model selection in historical biogeography reveals that founder-event speciation is a crucial process in Island Clades . Systematic Biology 63 : 951 – 970 . Google Scholar CrossRef Search ADS McHugh A , Yablonsky C , Binford G , Agnarsson I . 2014 . Molecular phylogenetics of Caribbean Micrathena (Araneae: Araneidae) suggests multiple colonisation events and single island endemism . Invertebrate Systematics 28 : 337 – 349 . Mittermeier RA , Robles-Gil P , Hoffman M , Pilgrim J , Brooks T , Mittermeier CG , Lamoreux J , da Fonseca GAB . 2005 . Hotspots revisited: earth’s biologically richest and most endangered terrestrial ecoregions . México : Agrupación Sierra Madre , 1– 392 pp. Moen DS , Wiens JJ . 2009 . Phylogenetic evidence for competitively driven divergence: body-size evolution in Caribbean treefrogs (Hylidae: Osteopilus) . Evolution 63 : 195 – 214 . Google Scholar CrossRef Search ADS Noonan BP , Pramuk JB , Bezy RL , Sinclair EA , de Queiroz K , Sites JW Jr . 2013 . Phylogenetic relationships within the lizard clade Xantusiidae: using trees and divergence times to address evolutionary questions at multiple levels . Molecular Phylogenetics and Evolution 69 : 109 – 122 . Google Scholar CrossRef Search ADS O’Dea A. Lessios HA , Coates AG , Eytan RI , Restrepo-Moreno SA , Cione AL , Collins LS , de Queiroz A , Farris DW , Norris RD , Stallard RF , Woodburne MO , Aguilera O , Aubry MPP , Berggren WA , Budd AF , Cozzuol MA , Coppard SE , Duque-Caro H , Finnegan S , Gasparini GMM , Grossman EL , Johnson KG , Keigwin LD , Knowlton N , Leigh EG , Leonard-Pingel JS , Marko PB , Pyenson ND , Rachello-Dolmen PG , Soibelzon E , Soibelzon L , Todd JA , Vermeij GJ , Jackson JB . 2016 . Formation of the Isthmus of Panama . Science Advances 2 : e1600883 . Google Scholar CrossRef Search ADS Oliva A , Fernández LA , Bachmann AO . 2002 . Sinopsis de los Hydrophiloidea acuáticos de la Argentina . Monografias del Museo Argentino de Ciencias Naturales 2 : 1 – 67 . Papadopoulou A , Anastasiou I , Vogler AP . 2010 . Revisiting the insect mitochondrial molecular clock: the mid-Aegean trench calibration . Molecular Biology and Evolution 27 : 1659 – 1672 . Google Scholar CrossRef Search ADS Poux C , Chevret P , Huchon D , de Jong WW , Douzery EJP . 2006 . Arrival and diversification of caviomorph rodents and platyrrhine primates in South America . Systematic Biology 52 : 228 – 244 . Google Scholar CrossRef Search ADS de Queiroz A . 2016 . Jurassic primates, immobile ducks and other oddities: a reply to Headsʼ review of the Monkeyʼs Voyage . Australian Systematic Botany 29 : 403 – 423 . Google Scholar CrossRef Search ADS Rambaut A . 2012 . FigTree v1.4 . Available at: http://tree.bio.ed.ac.uk/software/figtree Rambaut A , Suchard MA , Xie D , Drummond AJ . 2014 . Tracer v1.6 . Available at: http://beast.bio.ed.ac.uk/Tracer Ree RH , Smith SA . 2008 . Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis . Systematic Biology 57 : 4 – 14 . Google Scholar CrossRef Search ADS Říčan O , Piálek L , Zardoya R , Doadrio I , Zrzavý J . 2013 . Biogeography of the Mesoamerican Cichlidae (Teleostei: Heroini): colonization through the GAARlandia land bridge and early diversification . Journal of Biogeography 40 : 579 – 593 . Google Scholar CrossRef Search ADS Ricklefs RE , Bermingham E . 2002 . The concept of the taxon cycle in biogeography . Global Ecology & Biogeography 11 : 353 – 361 . Google Scholar CrossRef Search ADS Ricklefs RE , Bermingham E . 2008 . The West Indies as a laboratory of biogeography and evolution . Philosophical Transactions of the Royal Society B: Biological Sciences 363 : 2393 – 2413 . Google Scholar CrossRef Search ADS Ricklefs RE , Cox GW . 1972 . Taxon cycles in the West Indian avifauna . American Naturalist 106 : 195 – 219 . Google Scholar CrossRef Search ADS Rodriguez J , Pitts JP , von Dohlen CD . 2015 . Historical biogeography of the widespread spider wasp tribe Aporini (Hymenoptera: Pompilidae) . Journal of Biogeography 42 : 495 – 506 . Google Scholar CrossRef Search ADS Roelants K , Gower DJ , Wilkinson M , Loader SP , Biju SD , Guillaume K , Moriau L , Bossuyt F . 2007 . Global patterns of diversification in the history of modern amphibians . Proceedings of the National Academy of Sciences of the United States of America 104 : 887 – 892 . Google Scholar CrossRef Search ADS Ronquist F . 1997 . Dispersal-vicariance analysis: a new approach to the quantification of historical biogeography . Systematic Biology 46 : 195 – 203 . Google Scholar CrossRef Search ADS Ronquist F , Teslenko M , van der Mark P , Ayres DL , Darling A , Höhna S , Larget B , Liu L , Suchard MA , Huelsenbeck JP . 2012 . MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space . Systematic Biology 61 : 539 – 542 . Google Scholar CrossRef Search ADS Rosen DE . 1975 . A vicariance model of Caribbean biogeography . Systematic Biology 24 : 431 – 464 . Google Scholar CrossRef Search ADS Rosen DE . 1985 . Geological hierarchies and biogeographic congruence in the Caribbean . Annals of the Missouri Botanical Garden 72 : 636 – 659 . Google Scholar CrossRef Search ADS Sanmartín I , Ronquist F . 2004 . Southern hemisphere biogeography inferred by event-based models: plant versus animal patterns . Systematic Biology 53 : 216 – 243 . Google Scholar CrossRef Search ADS Sato JJ , Ohdachi SD , Echenique-Diaz L , Borroto-Páez L , Begué-Quiala G , Delgado-Labañino JL , Gámez-Díez J , Alvarez-Lemus J , Nguyen ST , Yamaguchi N , Kita M . 2016 . Molecular phylogenetic analysis of nuclear genes suggests a Cenozoic over-water dispersal origin for the Cuban Solenodon . Scientific Reports 6 : 31173 . Google Scholar CrossRef Search ADS Schulte P , Alegret L , Arenillas I , Arz JA , Barton PJ , Bown PR , Bralower TJ , Christeson GL , Claeys P , Cockell CS , Collins GS , Deutsch A , Goldin TJ , Goto K , Grajales-Nishimura JM , Grieve RAF , Gulick SPS , Johnson KR , Kiessling W , Koeberl C , Kring DA , MacLeod KG , Matsui T , Melosh J , Montanari A , Morgan JV , Neal CV , Nichols DJ , Norris RD , Pierazzo E , Ravizza G , Rebolledo-Vieyra M , Reimold WU , Robin E , Salge T , Speijer RP , Sweet AG , Urrutia-Fucugauchi J , Vajda V , Whalen MT , Willumsen PS . 2010 . The Chicxulub Asteroid Impact and Mass Extinction at the Cretaceous-Paleogene Boundary . Science 327 : 1215 – 1218 . Google Scholar CrossRef Search ADS Seidel M , Arriaga-Varela E , Fikáček M . 2016 . Establishment of Cylominae Zaitzev, 1908 as a valid name for the subfamily Rygmodinae Orchymont, 1916 with an updated list of genera (Coleoptera: Hydrophilidae) . Acta Entomologica Musei Nationalis Pragae 56 : 159 – 165 . Short AEZ . 2004 . Review of the Enochrus Thomson of the West Indies (Coleoptera: Hydrophilidae) . Koleopterologische Rundschau 74 : 351 – 361 . Short AEZ , Fikáček M . 2013 . Molecular phylogeny, evolution, and classification of the Hydrophilidae (Coleoptera) . Systematic Entomology 38 : 723 – 752 . Google Scholar CrossRef Search ADS Smetana A . 1978 . Revision of the subfamily Sphaeridiinae of America North of Mexico (Coleoptera: Hydrophilidae) . Memoirs of the Entomological Society of Canada 105 : 1 – 292 . Google Scholar CrossRef Search ADS Spangler PJ , Short AEZ . 2008 . Three new species of Neotropical Tropisternus Solier (Coleoptera: Hydrophilidae) . Zootaxa 1917 : 65 – 68 . Stadler T . 2009 . On incomplete sampling under birth–death models and connections to the sampling-based coalescent . Journal of Theoretical Biology 261 : 58 – 66 . Google Scholar CrossRef Search ADS Stamatakis A . 2014 . RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies . Bioinformatics 30 : 1312 – 1313 . Google Scholar CrossRef Search ADS Tada R , Iturralde-Vinent MA , Matsui T , Tajika E , Oji T , Goto K , Nakano Y , Takayama H , Yamamoto S , Kiyokawa S , Toyoda K , García Delgado D , Díaz-Otero C , Rojas R . 2003 . K/T boundary deposits in the paleo–western Caribbean basin . In: Bartolini C , Buffler R , Blickwede J , eds. The Circum–Gulf of Mexico and the Caribbean: hydrocarbon habitats, basin formation, and plate tectonics . Tulsa: American Association of Petroleum Geologists Memoir 79 , 582 – 604 . Thorpe RS , Malhotra A , Stenson A , Reardon JT . 2004 . Adaptation and speciation in Lesser Antillean anoles . In: Dieckmann U , Doebeli M , Metz JAJ , Tautz D , eds. Adaptive speciation . Cambridge : Cambridge University Press , 324 – 335 . Uit de Weerd DR , Robinson DG , Rosenberg G . 2016 . Evolutionary and biogeographical history of the land snail family Urocoptidae (Gastropoda: Pulmonata) across the Caribbean region . Journal of Biogeography 43 : 763 – 777 . Google Scholar CrossRef Search ADS Wahlberg N . 2006 . That awkward age for butterflies: insights from the age of the butterfly subfamily Nymphalinae (Lepidoptera: Nymphalidae) . Systematic Biology 55 : 703 – 714 . Google Scholar CrossRef Search ADS Wahlberg N , Freitas AV . 2007 . Colonization of and radiation in South America by butterflies in the subtribe Phyciodina (Lepidoptera: Nymphalidae) . Molecular Phylogenetics and Evolution 44 : 1257 – 1272 . Google Scholar CrossRef Search ADS Weaver PF , Cruz A , Johnson S , Dupin J , Weaver KF . 2016 . Colonizing the Caribbean: biogeography and evolution of livebearing fishes of the genus Limia (Poeciliidae) . Journal of Biogeography 43 : 1808 – 1819 . Google Scholar CrossRef Search ADS Wilson EO . 1959 . Adaptive shift and dispersal in a tropical ant fauna . Evolution 13 : 122 – 144 . Google Scholar CrossRef Search ADS Wilson EO . 1961 . The nature of the taxon cycle in the Melanesian ant fauna . American Naturalist 95 : 169 – 193 . Google Scholar CrossRef Search ADS Xie W , Lewis PO , Fan Y , Kuo L , Chen MH . 2011 . Improving marginal likelihood estimation for Bayesian phylogenetic model selection . Systematic Biology 60 : 150 – 160 . Google Scholar CrossRef Search ADS Zhang G , Basharat U , Matzke N , Franz NM . 2017 . Model selection in statistical historical biogeography of Neotropical insects – the Exophthalmus genus complex (Curculionidae: Entiminae) . Molecular Phylogenetics and Evolution 109 : 226 – 239 . Google Scholar CrossRef Search ADS Zhang JX , Maddison WP . 2013 . Molecular phylogeny, divergence times and biogeography of spiders of the subfamily Euophryinae (Araneae: Salticidae) . Molecular Phylogenetics and Evolution 68 : 81 – 92 . Google Scholar CrossRef Search ADS © 2017 The Linnean Society of London, Zoological Journal of the Linnean Society This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Zoological Journal of the Linnean Society Oxford University Press

Multiple origins of the Phaenonotum beetles in the Greater Antilles (Coleoptera: Hydrophilidae): phylogeny, biogeography and systematics

Loading next page...
 
/lp/ou_press/multiple-origins-of-the-phaenonotum-beetles-in-the-greater-antilles-wH79izYAeN
Publisher
The Linnean Society of London
Copyright
© 2017 The Linnean Society of London, Zoological Journal of the Linnean Society
ISSN
0024-4082
eISSN
1096-3642
D.O.I.
10.1093/zoolinnean/zlx071
Publisher site
See Article on Publisher Site

Abstract

Abstract The systematics and the phylogenetic position of the Caribbean representatives of Phaenonotum Sharp (Coleoptera: Hydrophilidae) are investigated to understand the composition of the Caribbean fauna and its origin. Phylogenetic analysis based on mitochondrial and nuclear genes has revealed the Caribbean species to be situated in three deeply nested clades, inferring multiple colonization of Caribbean islands from the continent. Time-tree analysis and BioGeoBEARS analyses of ancestral ranges estimated the oldest clade, consisting of wingless single-island endemics of Cuba (P. delgadoi), Jamaica (P. ondreji sp. nov.) and Hispaniola (P. laterale sp. nov.), to have diverged c. 46.6 Mya from the South American ancestor and subsequently colonizing the Caribbean most likely via the GAARlandia land bridge connecting South America with the Greater Antilles. The remaining three Caribbean species, including the Puerto Rican endemic, P. borinquenum sp. nov., are of more recent (Miocene to Pliocene) origin and colonized the Greater Antilles by over-water dispersal. All the Caribbean species are illustrated and diagnosed, and three new species are described. The genus Phaenonotum, excluding P. caribense Archangelsky, is confirmed as a monophylum. We demonstrate that species-level taxonomy of Phaenonotum is difficult to solve by morphology alone and ideally requires the combination of morphology and molecular markers. ancestral range analysis, Caribbean biogeography, endemics, GAARlandia, molecular dating, new species, phylogeny, systematics, taxon cycle INTRODUCTION The Greater Antilles (i.e. islands of Cuba, Hispaniola, Jamaica and Puerto Rico in the Caribbean Region) are known for their high species diversity and endemism and are considered one of the world biodiversity hotspots (Mittermeier et al., 2005). The complex geological history of the region, characterized by emergence and sinking of particular islands and changes in their interconnections, has played a crucial role in generating these diverse and endemic faunas. The Antilles first emerged as the Antillean volcanic arc system in the Cretaceous in the gap between North and South America and repeatedly emerged and submerged while moving eastward into the Proto-Caribbean Sea later during the Cenozoic. The uplift of the core Greater Antilles started during the Middle Eocene and reached its maximum at the Eocene–Oligocene boundary. The area of the islands was reduced during the higher sea-level period in the Late Oligocene to Middle Miocene, but the islands remained emerged, possibly with the exception of parts of today’s Jamaica and Hispaniola, which are believed to have emerged permanently only during the Neogene (e.g. Iturralde-Vinent & MacPhee, 1999; Bartolini, Lang & Spell, 2003; Iturralde-Vinent, 2006). The origin of the present day terrestrial and freshwater biota of the archipelago has been explained by three alternative scenarios. The first scenario assumes that ancestors of extant lineages colonized the Antillean volcanic arc system from southern North America (via land bridges or by short-distance over-water dispersal) during the Late Cretaceous and Early Paleogene and survived there as relicts to the present (e.g. Rosen, 1975, 1985). This scenario was originally proposed for solenodons and Cricosaura lizards (e.g. Hedges, 2006), but only the Cricosaura clade has been confirmed to be old enough (Noonan et al., 2013; Sato et al., 2016). Under this scenario, the ancient Antillean fauna would have had to survive the collision of the large bolide with the Earth at the Yucatan Peninsula 65 Mya and its consequences that exterminated most of the Cretaceous terrestrial life forms in the Caribbean region (Iturralde-Vinent, 1982; Iturralde-Vinent & MacPhee, 1999; Tada et al., 2003). The second scenario suggests long-distance over-water dispersal from the surrounding landmasses during the Cenozoic (e.g. Hedges, Hass & Maxson, 1992; Hedges, 2006), which was recently documented, for example, for solenodons (Sato et al., 2016) and urocoptid snails (Uit de Weerd, Robinson & Rosenberg, 2016). In contrast to over-water dispersal, Iturralde-Vinent & MacPhee (1999) introduced the GAARlandia theory (GAAR = Greater Antilles + Aves Ridge) assuming a land bridge connection of the Greater Antilles with northern South America resulting from a tectonic uplift and sea-level fall at the Eocene–Oligocene boundary. This third scenario assumes that numerous terrestrial and freshwater organisms colonized the archipelago during a relatively short period (c. 35–33 Mya), as was suggested, for example, for Peltophryne toads (Alonso, Crawford & Bermingham, 2012), poecilid fishes (Weaver et al., 2016) or heroine cichlid fishes (Říčan et al., 2013). The GAARlandia hypothesis is still a subject of debate: Hedges (2006) and Ali (2012) point to the absence of geological evidence and lack of any clear signal in vertebrate phylogenetic analyses and hypothesize GAARlandia as no more than a chain of widely spaced islands situated between northern South America and the Greater Antilles. The above scenarios have been tested by numerous phylogenetic studies, which make the Caribbean one of the model regions for understanding historical island biogeography. Paradoxically, the vast majority of these studies are focused on vertebrates (i.e. relatively species-poor clades with rather young modern crown-groups; e.g. Roelants et al., 2007; Claramount & Cracraft, 2015; Foley, Springer & Teeling, 2016) and plants (i.e. groups with resistant seeds facilitating long-distance dispersal; e.g. Sanmartín & Ronquist, 2004). Little is known about the origin and biogeography of Caribbean insects and other terrestrial or freshwater arthropods, that is species-rich old groups that may provide numerous independent examples of evolutionary histories corroborating or contradicting the above biogeographic scenarios. Dated molecular phylogenetic studies have been performed only for selected groups of Caribbean butterflies (Wahlberg, 2006; Wahlberg & Freitas, 2007; Matos-Maraví et al., 2014; Lewis et al., 2015), wasps (Ceccarelli & Zaldívar-Riverón, 2013; Rodriguez, Pitts & von Dohlen, 2015) and arachnids (e.g. Crews & Gillespie, 2010; Zhang & Maddison, 2013; McHugh et al., 2014; Dziki et al., 2015; Esposito et al., 2015; Agnarsson et al., 2016). Only a single very recent study is available for Caribbean beetles (Coleoptera), which is the most speciose and most intensively studied insect group (Zhang et al., 2017). The beetle family Hydrophilidae comprises c. 3350 species inhabiting both aquatic and terrestrial habitats and is distributed worldwide (Hansen, 1999; Short & Fikáček, 2013; Lawrence & Slipinski, 2014; Seidel, Arriaga-Varela & Fikáček, 2016). In the Caribbean, the family is represented by 16 genera, 11 of which contain at least one species endemic to the Greater Antilles (Hansen, 1999; Short, 2004; Spangler & Short, 2008; Deler-Hernández, Cala-Riquelme & Fikáček, 2013a; Deler-Hernández, Fikáček & Cala-Riquelme, 2013b; Deler-Hernández, Cala-Riquelme & Fikáček, 2014; Arriaga-Varela et al., 2017). Among them, the genus Phaenonotum Sharp, 1882 contains 18 described and many undescribed species in the Neotropical and southern Nearctic Regions, and one East-African species (Hansen, 1999), all inhabiting semiaquatic or terrestrial habitats. A single widespread species, Phaenonotum exstriatum (Say, 1835), was originally recorded from the Caribbean, until the surprising discovery of P. delgadoi Deler-Hernández et al., 2013 endemic to eastern Cuba (Deler-Hernández et al., 2013a). Our subsequent fieldwork revealed additional candidates for endemic species in Jamaica, Hispaniola and Puerto Rico, indicating that Phaenonotum possibly underwent an island radiation in the Greater Antilles. In this paper, we analyse the systematics and evolutionary history of the Caribbean Phaenonotum in order to understand to what extent it corroborates the aforementioned scenarios explaining the origin of Caribbean faunas. MATERIAL AND METHODS DNA sequences The molecular study is based on Phaenonotum specimens collected in all main islands of the Greater Antilles during our fieldwork in 2010–2016. Specimens were collected either manually from aquatic habitats, rotten plant debris and at the light traps, or by sifting of forest leaf litter and extracting the beetles from sifting samples using Berlese and/or Winkler funnels. All specimens were collected in 96% ethanol and stored at −20 °C in the lab. To understand the origin of the Caribbean fauna, we included Phaenonotum species from Costa Rica, Guatemala, Venezuela, Suriname, Guyana, Ecuador, Peru and the USA, that is all continental DNA-grade material available to us. To test the monophyly of Phaenonotum, we also included all available DNA-grade specimens of the genera Phaenostoma Orchymont, 1937 and Lachnodacnum Orchymont, 1937, which form a strongly supported clade with Phaenonotum (V. Sýkora et al., unpubl. data). Outgroup taxa consist of selected genera representing the main clades of the tribe Coelostomatini (Cyclotypus Sharp, 1882, Dactylosternum Wollaston, 1854 and Coelostoma Brullé, 1835) and the sphaeridiine tribes Protosternini (Sphaerocetum Fikáček, 2010) and Sphaeridiini (Sphaeridium Fabricius, 1775). Most specimens used for this study were newly extracted and sequenced, but in a few cases, we adopted the sequences published by Short & Fikáček (2013). DNA was extracted from complete specimens cut into two parts between prothorax and mesothorax, using the commercial DNeasy Tissue Kit (Qiagen GmbH, Hilden, Germany) following the manufacturer’s instructions. Voucher specimens and DNA extractions are kept in the collection of the Department of Entomology of the National Museum, Prague (NMPC). Our gene selection corresponds with those used in previous phylogenetic studies of the family Hydrophilidae (e.g. Short & Fikáček, 2013) and contains three mitochondrial genes (3′ end cytochrome c oxidase I, cytochrome c oxidase II and 16S rRNA), and two nuclear genes (18S rRNA and 28S rRNA); sequences of 18S rRNA were amplified in two parts (3′ end and 5′ end) and subsequently combined; for primers and PCR conditions, see Supporting Information, Supplement S1. Ten microlitres of each PCR product were purified by adding 1.0 µL Exonuclease 1 [Exo1 (20 U/µL)] (ThermoFisherScientific) and 2.0 µL Thermosensitive Alkaline Phosphatase [FastAP (1 U/µL)] (ThermoFisherScientific) and incubating the mixture for 15 min at 37 °C followed by a 15 min inactivation step at 80 °C. Sanger sequencing was performed by BIOCEV (Vestec, Czech Republic). DNA sequences were assembled and edited using Geneious 7.1.9 and submitted to GenBank under the accession numbers shown in Supporting Information, Supplement S1. In total, we obtained sequences for 36 terminal taxa (25 specimens of Phaenonotum, four of Phaenostoma, one of Lachnodacnum and six outgroup taxa). Phylogenetic analyses Our molecular data set includes sequences of three ribosomal genes, which are non-trivial for alignment especially in case of distantly related taxa (e.g. Bocák et al., 2014), and a rather large proportion of missing data for cox2 and 16S genes, which we failed to amplify despite repeated effort. To test the impact of these problems, we compared the results of analyses of three different data sets: (1) 35 taxa data set: contains all Phaenonotum + Phaenostoma + Lachnodacnum specimens and all outgroups; (2) 30 taxa data set: contains all Phaenonotum + Phaenostoma + Lachnodacnum specimens, but is only rooted with closely related Cyclotypus (eliminates the problem with alignment of distantly related taxa); (3) 22 taxa data set: contains only Phaenonotum + Phaenostoma + Lachnodacnum with three and more genes, and is rooted by Cyclotypus (eliminates the problem with the alignment of non-related taxa and the effect of missing cox2 and 16S data). Specimens MF1736 and MF1741 from Venezuela were genetically very close (2.1% in cox1), corresponding to intraspecific variation in cox1 found in Phaenonotum exstriatum (0.0‒3.1%), and were hence considered as conspecific; they were combined for our analyses and included in all three data sets. Sequences were aligned with MAFFT algorithm as implemented in Geneious 7.1.9 software, using the default settings (score matrix = 200PAM/k = 2, gap open penalty = 1.53, offset value = 0.123). The final alignment has the total length of 4827 bp, consisting of the following gene fragments: cox1 (778 bp), cox2 (694 bp), 16S (530 bp), 18S (1798 bp) and 28S (1027 bp). The data set was divided into nine partitions (by genes, plus cox1 and cox2 were both divided by codon positions). Each data set was analysed using Bayesian inference and maximum likelihood. Bayesian analyses were performed in MrBayes 3.2.6 (Ronquist et al., 2012) using four chains of 25000000 generations and sampling the chain every 1000 generations. We sampled across the substitution model space in the Bayesian MCMC analysis itself (Huelsenbeck, Larget & Alfaro, 2004; Ronquist et al., 2012) instead of prior testing for an appropriate model for each of nine partitions. Results were examined in Tracer v.1.6 (Rambaut et al., 2014) to check for the proper effective sample size, proper mixing of chains and reaching the stationary phase; 25% burn-in was used for the construction of the final consensus tree. Maximum likelihood analyses were performed using RAxML 8.0 (Stamatakis, 2014) with a GTR substitution model and 1000 bootstrap replicates. Resulting trees were visualized in FigTree 1.4.3 (Rambaut, 2012). Divergence dating We performed a divergence dating analysis using the full data set of 35 taxa in BEAST 2.4.5 (Bouckaert et al., 2014) with fixed tree topology as revealed by the Bayesian analysis of the 35 taxa data set; the internal topology of P. exstriatum was manipulated to be bifurcate in agreement with results of the maximum likelihood analyses of 35 and 30 taxa. We divided the data set into five partitions corresponding to each gene (cox1, cox2, 16S rRNA, 18S rRNA and 28S rRNA) and used PartitionFinder v1.1.1 (Lanfear et al., 2012) to estimate the evolutionary model that best fit the data for each partition separately. The following substitution models were selected using Bayesian information criterion and used for particular genes in the divergence dating analysis: GTR + I + G for cox1, cox2 and 28S; GTR + G for 16S; and TrNef + I + G for 18S. Due to the absence of fossils belonging to the tribe Coelostomatini, we used the combination of a rate dating and node dating constraining the age of the most recent common ancestor (MRCA) of the Coelostomatini. Molecular clock models were linked into two partitions (mtDNA: cox1, cox2, 16S rRNA; nDNA: 18S rRNA, 28S rRNA) and substitution rates were set to 0.0133 (mtDNA) and 0.0017 (nDNA) substitutions per million years following Papadopoulou, Anastasiou & Vogler (2010). The age of the MRCA of the Coelostomatini was constrained to 152.5 Mya [95% highest posterior density (confidence interval): 134–170 Mya], following the results of the time tree analysis of the whole family Hydrophilidae performed by Bloom, Fikáček & Short (2014) based on a wide spectrum of fossil calibrations (for list of fossils used see supplementary material in Bloom et al., 2014). We performed a Bayes Factor comparison to test alternative clock models (non-clock, strict clock, relaxed clock) using a stepping-stone method (Xie et al., 2011) as implemented in MrBayes 3.2.6 (Ronquist et al., 2012) (see Supporting Information, Supplement S1). The birth–death model was used for the tree prior as this model is commonly used to model speciations and extinctions in inferring phylogenies using Bayesian methods; thus, at any point in time, every lineage can undergo speciation at rate λ or go extinct at rate μ (Stadler, 2009). Due to the problems with convergence of parameters, we performed two separate runs each with MCMC chain length set to 500 million generations and sampling frequency of every 25000 generations. We combined both runs using LogCombiner 2.4.0 and applied a 10% burn-in fraction after checking the convergence of parameters in TRACER 1.6 (Rambaut et al., 2014). TreeAnnotator 2.4.1 was used to summarize the information of all trees in the sample onto the maximum clade credibility tree. The resulting tree was visualized in FigTree 1.4.3 (Rambaut, 2012). Historical biogeography reconstruction For historical biogeography estimation, we used the time tree resulting from the divergence dating analysis as the input tree, from which we excluded all outgroups except Cyclotypus. Four specimens of P. exstriatum were included into the analysis, each representing a geographically distinct population. An alternative set of analyses with a single terminal for P. exstriatum was also performed. The distribution ranges were divided into the following eight areas: A ‒ North America incl. Mexico; B ‒ Central America; C ‒ northern South America (corresponding to Guiana Shield + northwestern Venezuela and northern Colombia, i.e. regions on the Caribbean coast and in direct contact with Central America); D ‒ South America (remaining continent south of the former region); E ‒ Cuba; F ‒ Jamaica; G ‒ Hispaniola; H ‒ Puerto Rico. As no inter-island divergences are present in our phylogeny, we coded the Caribbean islands in their current shape, without considering their historical composition of multiple paleoislands. Islands of the Lesser Antilles were not considered as separate areas, as only the youngest and widespread P. exstriatum is known to occur in some of them; instead, the presence/absence of the Lesser Antilles was considered when setting dispersal multipliers for the time-stratified analysis (see below). The distribution of terminal taxa is based on examined specimens in all undescribed species; distribution of P. exstriatum and P. laevicolle complex on the continent follows Smetana (1978); Hansen (1999); Oliva, Fernández & Bachmann (2002) and Gonzáles-Rodríguez, García-Hernández & Clarkson (2017). We carried out the historical biogeography analyses in the R package BioGeoBEARS (Matzke, 2014) to estimate the timing of Phaenonotum colonization of the Greater Antilles archipelago. This package contains three models implemented in a maximum likelihood framework: DEC model (Ree & Smith, 2008), DIVALIKE model (likelihood version of the DIVA model: Ronquist, 1997) and BAYAREALIKE model (likelihood version of BayArea model: Landis et al., 2013). Moreover, each model is available in its original version and with an additional parameter j representing the founder event, that is speciation following long-distance dispersal; six different models are hence available in total. We conducted non-time-stratified and time-stratified analyses to estimate ancestral area distribution on given nodes, each with all six models. Unconstrained non-stratified analyses were done using default parameter values. For time-stratified analyses, time periods were defined as follows, to reflect different paleogeography of the area in each period (Iturralde-Vinent, 2006; O’Dea et al., 2016): (1) 0–2.8 Mya: from present to the closing of Isthmus of Panama; (2) 2.8–9 Mya: the Greater Antilles emerged, the chain of islands (Lesser Antilles) present between the Greater Antilles and South America; (3) 9–32 Mya: the Greater Antilles significantly reduced in area and widely separated from South America by deep sea; (4) 32–38 Mya: the Greater Antilles connected to northern South America by GAARlandia land bridge; (5) 38–55 Mya: the Greater Antilles started to form and were well separated from surrounding continents; (6) 55–110 Mya: prior to the formation of the Greater Antilles. Dispersal probabilities were set to reflect the paleogeography as follows: 0.8 for adjacent continental areas; 0.5 for non-adjacent continental areas; 0.2 for adjacent islands (or island-continent) separated by less than 200 km from sea; 0.05 for connection by island chain (e.g. Lesser Antilles) or intermediate island (e.g. Hispaniola between Cuba and Puerto Rico); 0.001 for long-distance dispersal (areas separated by more than 200 km from sea); 0.000001 when dispersal was not possible (i.e. when the area was submerged; we followed the BioGeoBEARS manual in setting extremely low rather than zero probabilities in such cases). Areas not present during particular time slice (e.g. Cuba in the Late Cretaceous) were not allowed for the reconstruction using the Areas Allowed matrix. Models of both non-time-stratified and time-stratified analyses were compared using likelihood values and Akaike information criterion corrected for small sample sizes (AICc) (Matzke, 2013). All input files for the biogeography analyses are available in Supporting Information, Supplement S2. Morphology and taxonomy Examined specimens are deposited in the following collections: BMNH Natural History Museum, London, United Kingdom (M.V.L. Barclay); CMN Canadian Museum of Nature, Ottawa, Canada (R. Anderson); DZRJ Coleção Entomológica Prof. José Alfredo Pinheiro Dutra, Instituto de Biologia, Universidade Federal de Rio de Janeiro, Rio de Janeiro, Brazil (B. Clarkson); MCZ Museum of Comparative Zoology, Cambridge, USA (P. Perkins); MNHNSD Museo Nacional de Historia Natural, Santo Domingo, Dominican Republic (C. Suriel); NHMW Naturhistorisches Museum, Wien, Austria (M. Jäch, A. Komarek); NMPC National Museum, Prague, Czech Republic (M. Fikáček); SBNM Santa Barbara Museum of Natural History, California, USA (M. L. Gimmel); SBPC Stewart Peck Personal Collection, Ottawa, Canada; UPRM University of Puerto Rico, Mayagüez, Puerto Rico (A. Segarra); ZMHB Museum für Naturkunde der Humboldt-Universität, Berlin, Germany (J. Frisch, B. Jäger). Habitus photographs were taken using Canon EOS 550D digital camera with attached Canon MP-E65mm f/2.8 1–5× macro lens and subsequently adapted in Adobe Photoshop CS5 and CorelDRAW Home & Student X8. Photographs of genitalia were taken using Canon EOS 1100D digital camera attached to Olympus BX41 compound microscope and subsequently combined with Helicon Focus software. Scanning electron micrographs of the holotype of the new species were taken using a Hitachi S-3700N environmental electron microscope at the Department of Paleontology, National Museum in Prague. General morphological terminology follows Smetana (1978), Archangelsky (1989) and Hansen (1991). All Caribbean species were compared with type specimens of the Central and South American species deposited in the Natural History Museum, London (Sharp collection; see Deler-Hernández & Fikáček, 2016) and in the Institut Royal des Sciences Naturelles de Belgique, Brussels (Orchymont collection), and with unidentified material from Mexico, Costa Rica, Ecuador and Peru deposited in National Museum, Prague, in order to reveal whether or not they might be conspecific with continental species. Complete data of all examined specimens in DarwinCore-formatted Excel spreadsheet are available in Supporting Information, Supplement S3 and in the data set submitted to Zenodo. Depository of primary data All primary data and results of all analyses were uploaded as a.zip file into the Zenodo depository (https://zenodo.org/) under doi: 10.5281/zenodo.850595. Parts of the data were also uploaded to specialized archives as specified below. RESULTS Phylogenetic analyses All analyses performed revealed a strongly supported clade consisting of Neotropical genera Phaenostoma, Phaenonotum and Lachnodacnum, but do not support the current status of the genera (Fig. 1). Phaenonotum caribenseArchangelsky, 1989 is revealed as not related to the remaining Phaenonotum species and is revealed as a sister group to the rest of the Neotropical clade in most analyses (strongly supported by Bayesian analysis with 30 and 22 taxa, weakly supported by maximum likelihood analyses); in the Bayesian analysis with 35 taxa, it is revealed as sister to Phaenostoma kontax Gustafson & Short, 2010 and Lachnodacnum, with moderate support. When P. caribense is excluded, the genus Phaenonotum (= core Phaenonotum hereafter) is revealed as strongly supported monophylum in all analyses. The genus Phaenostoma is revealed as sister to the core Phaenonotum in most analyses (moderately supported by Bayesian analysis except that with 35 taxa, weakly supported by all maximum likelihood analyses), with Lachnodacnum always nested within Phaenostoma. In the Bayesian analysis with 35 taxa, Phaenonotum caribense + Phaenostoma kontax + Lachnodacnum clade is revealed as sister to the Phaenostoma posticatum + core Phaenonotum. Figure 1. View largeDownload slide The phylogenetic position and polyphyly of Phaenonotum as revealed by our maximum likelihood (A) and Bayesian analyses (B). Branch support is indicated for analyses with 35, 30 and 22 terminal taxa, respectively. Figure 1. View largeDownload slide The phylogenetic position and polyphyly of Phaenonotum as revealed by our maximum likelihood (A) and Bayesian analyses (B). Branch support is indicated for analyses with 35, 30 and 22 terminal taxa, respectively. Within the core Phaenonotum clade, all six analyses revealed the same five strongly to moderately supported clades and two undescribed species (MF1739 from Guatemala and MF861 from Peru) not closely related to other species and forming deeply divergent separate clades (Fig. 2). The relationships between these seven principal clades are the same in all analyses performed, with the exception of MF1739, which is revealed as early branching in Bayesian analyses and more deeply nested in maximum likelihood analyses. All Caribbean species are part of the strongly supported deeply nested monophylum containing three of the principal clades: the clade of the endemic Cuban, Jamaican and Hispaniolan species; the clade containing Puerto Rican endemic P. borinquenum sp. nov., P. laevicolle complex and an undescribed species from Guyana (MF1061); and the clade consisting of P. exstriatum and undescribed species from Venezuela and Suriname. Within the Caribbean endemic clade, the internal topology is the same in all analyses, with the Cuban P. delgadoi diverging first, and the Jamaican P. ondreji sp. nov. and Hispaniolan P. laterale sp. nov. as sister taxa. The internal topology of the P. borinquenum + laevicolle clade varies the most among analyses, with either P. borinquenum or the undescribed species from Guyana (MF1061) revealed as the earliest diverging taxon. The internal topology of P. exstriatum clade differs between Bayesian and maximum likelihood analyses, with Puerto Rican (MF1728) and Costa Rican (MF1738) revealed as sister taxa in maximum likelihood ones; the Cuban specimen (MF654) is always revealed as sister to the one from USA: Delaware (MF1063). Figure 2. View largeDownload slide Summary of the internal topology of the core Phaenonotum clade revealed by Bayesian and maximum likelihood analyses of all three data sets (35 taxa, 30 taxa and 22 taxa, see Material and methods for details). A, topology resulting from Bayesian inference of 35 taxa data set, with the presence and support of particular clades in all analyses mapped by colour squares; B, alternative topology of the borinquenum‒laevicolle clade obtained from maximum likelihood analysis of the 30 taxa data set; C, alternative internal topology of P. exstriatum clade obtained from maximum likelihood analyses of 35 and 30 taxa data sets. Figure 2. View largeDownload slide Summary of the internal topology of the core Phaenonotum clade revealed by Bayesian and maximum likelihood analyses of all three data sets (35 taxa, 30 taxa and 22 taxa, see Material and methods for details). A, topology resulting from Bayesian inference of 35 taxa data set, with the presence and support of particular clades in all analyses mapped by colour squares; B, alternative topology of the borinquenum‒laevicolle clade obtained from maximum likelihood analysis of the 30 taxa data set; C, alternative internal topology of P. exstriatum clade obtained from maximum likelihood analyses of 35 and 30 taxa data sets. Divergence dating The relaxed clock model was identified as the best-fitting clock model based on the Bayes factor comparison (see Supporting Information, Supplement S1), and it was hence implemented in BEAST time tree analysis. The core Phaenonotum clade was estimated to originate during the middle Cretaceous (c. 102 Mya), with the diversification of modern clades starting in Late Cretaceous (c. 83 Mya). The clade composed of the endemic Cuban, Jamaican and Hispaniolan Phaenonotum diverged during the Eocene (c. 47 Mya), with P. delgadoi from Cuba diverging at about Eocene–Oligocene boundary (c. 36 Mya), and P. ondreji (Jamaica) from P. laterale (Hispaniola) during the Oligocene (c. 26.4 Mya). The Puerto Rican endemic P. borinquenum diverged in the early Miocene (c. 19 Mya). The divergence of the Cuban and Venezuelan specimens of P. laevicolle complex was dated to the Pliocene–Pleistocene (c. 4.7 Mya), and the modern populations of the widespread P. exstriatum diverged at about Pliocene–Pleistocene boundary (c. 2.4 Mya). Precise ages and 95% confidence intervals are listed in Table 1 for the principal Phaenonotum clades and illustrated in Supporting Information, Supplement S1 for all clades. Table 1. Divergence ages and their confidence intervals for major Phaenonotum clades including all Caribbean ones Age (Mya) 95% confidence interval (Mya) Phaenonotum stem 101.7 81.7‒121.9 Phaenonotum crown 82.5 65.4‒100.2 Caribbean Phaenonotum stem 46.6 35.6‒58.1 P. delgadoi 35.7 25.5‒47.2 P. ondreji and P. laterale 26.4 17.2‒36.5 P. borinquenum 19.2 12.1‒27.1 P. laevicolle from Cuba 4.7 2.2‒7.9 P. exstriatum crown 2.4 1.3‒3.7 Age (Mya) 95% confidence interval (Mya) Phaenonotum stem 101.7 81.7‒121.9 Phaenonotum crown 82.5 65.4‒100.2 Caribbean Phaenonotum stem 46.6 35.6‒58.1 P. delgadoi 35.7 25.5‒47.2 P. ondreji and P. laterale 26.4 17.2‒36.5 P. borinquenum 19.2 12.1‒27.1 P. laevicolle from Cuba 4.7 2.2‒7.9 P. exstriatum crown 2.4 1.3‒3.7 View Large Table 1. Divergence ages and their confidence intervals for major Phaenonotum clades including all Caribbean ones Age (Mya) 95% confidence interval (Mya) Phaenonotum stem 101.7 81.7‒121.9 Phaenonotum crown 82.5 65.4‒100.2 Caribbean Phaenonotum stem 46.6 35.6‒58.1 P. delgadoi 35.7 25.5‒47.2 P. ondreji and P. laterale 26.4 17.2‒36.5 P. borinquenum 19.2 12.1‒27.1 P. laevicolle from Cuba 4.7 2.2‒7.9 P. exstriatum crown 2.4 1.3‒3.7 Age (Mya) 95% confidence interval (Mya) Phaenonotum stem 101.7 81.7‒121.9 Phaenonotum crown 82.5 65.4‒100.2 Caribbean Phaenonotum stem 46.6 35.6‒58.1 P. delgadoi 35.7 25.5‒47.2 P. ondreji and P. laterale 26.4 17.2‒36.5 P. borinquenum 19.2 12.1‒27.1 P. laevicolle from Cuba 4.7 2.2‒7.9 P. exstriatum crown 2.4 1.3‒3.7 View Large Biogeography analyses Of the six models included in BioGeoBEARS, those implementing a founder event (i.e. perapatric speciation, j parameter in the analyses) fit the data better both in time-stratified and non-time-stratified analyses than those not allowing for jump dispersal (Table 2). Analyses allowing the founder event resulted in identical ancestral area estimates in all nodes and corners in non-time-stratified analyses under all three basic models (DEC + J, DIVALIKE + J and BAYAREALIKE + J) and in identical estimates in time-stratified analyses with DEC + J and DIVALIKE + J models (time-stratified analysis with BAYAREALIKE + J model is nearly identical, only with slightly different scenario for P. exstriatum clade and P. borinquenum, see Supporting Information, Supplement S2 for details). Models not implementing the founder event resulted in very different ancestral area estimates for each model in non-time-stratified analyses, but in identical estimates under all three models in time-stratified ones. The number of continent-to-Caribbean and Caribbean-to-continent dispersal events (and source/sink regions on continents) for all three models in non-time-stratified and time-stratified analyses are shown in Table 3. Based on the AICc metrics, the best-performing models were DIVALIKE + J for both non-time-stratified and time-stratified analyses (Fig. 3). Table 2. Comparison of the models used for non-time-stratified and time-stratified analyses of the data set with multiple populations of P. exstriatum LnL npar d e j AIC AICc Non-time constrained  DEC −91.91 2 0.0013 < 0.0001 0 187.8 188.3  DEC + J −70.21 3 0.0004 < 0.0001 0.046 146.4 147.3  DIVALIKE −84.36 2 0.0016 < 0.0001 0 172.7 173.2  DIVALIKE + J* −69.8 3 0.0005 < 0.0001 0.060 145.6 146.5  BAYAREALIKE −125.4 2 0.0023 0.026 0 254.8 255.2  BAYAREALIKE + J −70.56 3 0.0004 < 0.0001 0.049 147.1 148 Time constrained  DEC −104.9 2 0.016 0.0049 0 213.9 214.3  DEC + J −98.94 3 0.011 0.0039 0.17 203.9 204.8  DIVALIKE −100.2 2 0.017 0.0042 0 204.4 204.8  DIVALIKE + J* −95.44 3 0.012 0.0034 0.099 196.9 197.8  BAYAREALIKE −134.2 2 0.033 0.025 0 272.5 272.9  BAYAREALIKE + J −101.8 3 0.0077 0.0066 0.24 209.5 210.5 LnL npar d e j AIC AICc Non-time constrained  DEC −91.91 2 0.0013 < 0.0001 0 187.8 188.3  DEC + J −70.21 3 0.0004 < 0.0001 0.046 146.4 147.3  DIVALIKE −84.36 2 0.0016 < 0.0001 0 172.7 173.2  DIVALIKE + J* −69.8 3 0.0005 < 0.0001 0.060 145.6 146.5  BAYAREALIKE −125.4 2 0.0023 0.026 0 254.8 255.2  BAYAREALIKE + J −70.56 3 0.0004 < 0.0001 0.049 147.1 148 Time constrained  DEC −104.9 2 0.016 0.0049 0 213.9 214.3  DEC + J −98.94 3 0.011 0.0039 0.17 203.9 204.8  DIVALIKE −100.2 2 0.017 0.0042 0 204.4 204.8  DIVALIKE + J* −95.44 3 0.012 0.0034 0.099 196.9 197.8  BAYAREALIKE −134.2 2 0.033 0.025 0 272.5 272.9  BAYAREALIKE + J −101.8 3 0.0077 0.0066 0.24 209.5 210.5 AIC, Akaike information criterion; AICc, size-corrected AIC; LnL, log likelihood; n par, number of parameters in the analysis; d, e, j, parameres of the model (d, dispersal; e, extinction, j, founder event). *Best-performing model for each groups of analyses. View Large Table 2. Comparison of the models used for non-time-stratified and time-stratified analyses of the data set with multiple populations of P. exstriatum LnL npar d e j AIC AICc Non-time constrained  DEC −91.91 2 0.0013 < 0.0001 0 187.8 188.3  DEC + J −70.21 3 0.0004 < 0.0001 0.046 146.4 147.3  DIVALIKE −84.36 2 0.0016 < 0.0001 0 172.7 173.2  DIVALIKE + J* −69.8 3 0.0005 < 0.0001 0.060 145.6 146.5  BAYAREALIKE −125.4 2 0.0023 0.026 0 254.8 255.2  BAYAREALIKE + J −70.56 3 0.0004 < 0.0001 0.049 147.1 148 Time constrained  DEC −104.9 2 0.016 0.0049 0 213.9 214.3  DEC + J −98.94 3 0.011 0.0039 0.17 203.9 204.8  DIVALIKE −100.2 2 0.017 0.0042 0 204.4 204.8  DIVALIKE + J* −95.44 3 0.012 0.0034 0.099 196.9 197.8  BAYAREALIKE −134.2 2 0.033 0.025 0 272.5 272.9  BAYAREALIKE + J −101.8 3 0.0077 0.0066 0.24 209.5 210.5 LnL npar d e j AIC AICc Non-time constrained  DEC −91.91 2 0.0013 < 0.0001 0 187.8 188.3  DEC + J −70.21 3 0.0004 < 0.0001 0.046 146.4 147.3  DIVALIKE −84.36 2 0.0016 < 0.0001 0 172.7 173.2  DIVALIKE + J* −69.8 3 0.0005 < 0.0001 0.060 145.6 146.5  BAYAREALIKE −125.4 2 0.0023 0.026 0 254.8 255.2  BAYAREALIKE + J −70.56 3 0.0004 < 0.0001 0.049 147.1 148 Time constrained  DEC −104.9 2 0.016 0.0049 0 213.9 214.3  DEC + J −98.94 3 0.011 0.0039 0.17 203.9 204.8  DIVALIKE −100.2 2 0.017 0.0042 0 204.4 204.8  DIVALIKE + J* −95.44 3 0.012 0.0034 0.099 196.9 197.8  BAYAREALIKE −134.2 2 0.033 0.025 0 272.5 272.9  BAYAREALIKE + J −101.8 3 0.0077 0.0066 0.24 209.5 210.5 AIC, Akaike information criterion; AICc, size-corrected AIC; LnL, log likelihood; n par, number of parameters in the analysis; d, e, j, parameres of the model (d, dispersal; e, extinction, j, founder event). *Best-performing model for each groups of analyses. View Large Table 3. Number of continent-to-Caribbean and Caribbean-to-continent colonization events (total number and number of colonizations from/to each continental area) inferred for the genus Phaenonotum in analyses including four populations of P. exstriatum Model Time constraints Continent-to-Caribbean events Caribbean-to-Continent events Total NA CA nSA CA + SA NA + CA + SA Total NA CA DEC Non-strat. 3× 1× 2× ‒ ‒ 1× 1× ‒ DIVALIKE Non-strat. 4× ‒ 4× ‒ ‒ 0× ‒ ‒ BAYAREALIKE Non-strat. 5× 2× ‒ 3× ‒ 1× 1× ‒ DEC + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DIVALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× BAYAREALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DEC Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DIVALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× BAYAREALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DEC + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× DIVALIKE + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× BAYAREALIKE + J Time-strat. 5× 1× ‒ 4× ‒ ‒ 1× ‒ 1× Model Time constraints Continent-to-Caribbean events Caribbean-to-Continent events Total NA CA nSA CA + SA NA + CA + SA Total NA CA DEC Non-strat. 3× 1× 2× ‒ ‒ 1× 1× ‒ DIVALIKE Non-strat. 4× ‒ 4× ‒ ‒ 0× ‒ ‒ BAYAREALIKE Non-strat. 5× 2× ‒ 3× ‒ 1× 1× ‒ DEC + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DIVALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× BAYAREALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DEC Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DIVALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× BAYAREALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DEC + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× DIVALIKE + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× BAYAREALIKE + J Time-strat. 5× 1× ‒ 4× ‒ ‒ 1× ‒ 1× Models best fitting the data are marked in grey. Abbreviations of source/sink areas: CA, Central America; NA, North America; nSA, northern South America. Time constraint, non-strat., non-time-stratified analysis; time-strat., time-stratified analysis. View Large Table 3. Number of continent-to-Caribbean and Caribbean-to-continent colonization events (total number and number of colonizations from/to each continental area) inferred for the genus Phaenonotum in analyses including four populations of P. exstriatum Model Time constraints Continent-to-Caribbean events Caribbean-to-Continent events Total NA CA nSA CA + SA NA + CA + SA Total NA CA DEC Non-strat. 3× 1× 2× ‒ ‒ 1× 1× ‒ DIVALIKE Non-strat. 4× ‒ 4× ‒ ‒ 0× ‒ ‒ BAYAREALIKE Non-strat. 5× 2× ‒ 3× ‒ 1× 1× ‒ DEC + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DIVALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× BAYAREALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DEC Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DIVALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× BAYAREALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DEC + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× DIVALIKE + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× BAYAREALIKE + J Time-strat. 5× 1× ‒ 4× ‒ ‒ 1× ‒ 1× Model Time constraints Continent-to-Caribbean events Caribbean-to-Continent events Total NA CA nSA CA + SA NA + CA + SA Total NA CA DEC Non-strat. 3× 1× 2× ‒ ‒ 1× 1× ‒ DIVALIKE Non-strat. 4× ‒ 4× ‒ ‒ 0× ‒ ‒ BAYAREALIKE Non-strat. 5× 2× ‒ 3× ‒ 1× 1× ‒ DEC + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DIVALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× BAYAREALIKE + J Non-strat. 4× ‒ 3× ‒ 1× 2× 1× 1× DEC Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DIVALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× BAYAREALIKE Time-strat. 3× ‒ 3× ‒ ‒ 2× 1× 1× DEC + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× DIVALIKE + J Time-strat. 4× ‒ 4× ‒ ‒ 3× 1× 2× BAYAREALIKE + J Time-strat. 5× 1× ‒ 4× ‒ ‒ 1× ‒ 1× Models best fitting the data are marked in grey. Abbreviations of source/sink areas: CA, Central America; NA, North America; nSA, northern South America. Time constraint, non-strat., non-time-stratified analysis; time-strat., time-stratified analysis. View Large Figure 3. View largeDownload slide Results of the ancestral area estimation using time-stratified and non-time-stratified DIVALIKE + J model of BioGeoBEARS. Maps (A–G) show simplified continent and island positions in the respective time window used for the time-stratified analysis. Pie charts show two or three most probable ancestral areas for the respective node. Figure 3. View largeDownload slide Results of the ancestral area estimation using time-stratified and non-time-stratified DIVALIKE + J model of BioGeoBEARS. Maps (A–G) show simplified continent and island positions in the respective time window used for the time-stratified analysis. Pie charts show two or three most probable ancestral areas for the respective node. Most analyses infer South America as the ancestral range of the MRCA of Phaenonotum (Fig. 3), with several independent colonizations of northern South America and Central America (through northern South America in time-constrained analyses). Most analyses (non-time-stratified allowing for founder event and all time-stratified except BAYAREALIKE + J model) infer that ancestor(s) of P. delgadoi + P. laterale + P. ondreji colonized the Caribbean from northern South America: non-time-stratified models and time-stratified models not allowing for founder event infer a single colonization followed by series of founder events (non-time-stratified analyses) or combination of range expansions and vicariance events (time-stratified analyses); time-stratified analyses allowing for founder event infer two independent colonizations followed by vicariance event between Jamaica and Hispaniola. Phaenonotum borinquenum and Caribbean P. laevicolle were estimated to colonize the Caribbean independently by the non-time-stratified analysis. In contrast, time-stratified analysis favours earlier colonization of the Caribbean by the ancestor of these taxa and back-colonization from the Caribbean to the continent in P. laevicolle. Phaenonotum exstriatum was revealed to be of South American origin in non-time-constrained analyses allowing for founder event and all time-constrained analyses; the ancestral range of its MRCA was either Caribbean or Caribbean + Central America, depending on the model implemented. The results of all analyses are available in the Supporting Information, Supplement S2. Alternative analyses with P. exstriatum as a single terminal resulted in estimates nearly identical to those treating the four populations of P. exstriatum separately, with DEC + J as the best-performing model for time-stratified and non-time-stratified analyses. Both analyses revealed northern South America as an ancestral range of the MRCA of P. exstriatum and its sister species (undescribed species from Suriname, voucher MF1062). Results of these analyses are available in Supporting Information, Supplement S2. Systematics of the Caribbean Phaenonotum Key to the Caribbean Phaenonotum 1. Eyes small, interocular distance 5.0–6.0× the width of one eye in dorsal view (Fig. 6A–C). Metaventrite very short, its total width 5.0–6.0× the length behind mesocoxae (Fig. 7A–C). Species without hind wings (apterous). 2 – Eyes moderately large, interocular distance 3.5–4.0× the width of one eye in dorsal view (Fig. 6D–F). Metaventrite moderately long, its total width 3.0–4.0× the length behind mesocoxae (Fig. 7D–E). Species with fully developed hind wings (macropterous). 4 2. Metaventrite with pubescent cavities at sides of meso-metaventral process (Fig. 7A). Elytral suture strongly elevated posteriorly; elytral punctation very coarse (Fig. 5A). Pronotum with moderately coarse punctation, without microsculpture. Eastern Cuba (Fig. 4). P. delgadoi Deler-Hernández et al., 2013. – Metaventrite without pubescent cavities at sides of meso-metaventral process (Fig. 7B–F). Elytral suture weakly elevated posteriorly; elytral punctation never extremely coarse. Pronotum either with distinct microsculpture or with extremely fine (nearly invisible) punctation. 3 3. Body length 3.1–3.4 mm. Head and pronotum with mesh-like microsculpture and distinct punctation. Elytral punctation rather coarse, yellowish margins of elytra present in apical half only (Fig. 5B). Mesoventral process moderately wide, median portion of metaventrite with narrow highly elevated median keel (Fig. 7C). Aedeagus 0.6 mm long (Fig. 6H). Jamaica (Fig. 4). P. ondreji sp. nov. – Body length 2.5–2.7 mm. Head and pronotum without microsculpture, pronotal punctation extremely fine, almost invisible. Elytral punctation fine, whole lateral margins of elytra with yellowish stripe (Fig. 5C). Mesoventral process very narrow, metaventrite with slightly elevated median portion, without median keel. Aedeagus 0.35 mm long (Fig. 6I). Hispaniola (Fig. 4). P. laterale sp. nov. 4. Median lobe of the aedeagus wide; gonopore large, situated in apical third of the median lobe. Bases of lateral struts of the median lobe weakly expanded laterally (Fig. 6L). Widespread in the Greater Antilles (Fig. 4). .. P. laevicolle complex – Median lobe of the aedeagus narrow; gonopore small, situated subapically. Bases of lateral struts of the median lobe largely expanded laterally (Fig. 6J, K). 5 5. Body length 3.3–4.0 mm. Bases of parameres meeting in single point (Fig. 6K). Widespread in Greater and Lesser Antilles (Fig. 4). P. exstriatum (Say, 1835) – Body length 3.2–3.3 mm. Bases of parameres widely joined together (Fig. 6J). Puerto Rico (Fig. 4). P. borinquenum sp. nov. Figure 4. View largeDownload slide Current distribution of the genus Phaenonotum in the Caribbean. A, single island endemics; B, P. exstriatum; C, P. laevicolle complex. The localities of sequenced specimens are shown and marked by the code of the respective DNA extract. Figure 4. View largeDownload slide Current distribution of the genus Phaenonotum in the Caribbean. A, single island endemics; B, P. exstriatum; C, P. laevicolle complex. The localities of sequenced specimens are shown and marked by the code of the respective DNA extract. Figure 5. View largeDownload slide Habitus of the Caribbean Phaenonotum (dorsal and lateral views). A–D, species endemic for the Greater Antilles: A, P. delgadoi Deler-Hernández et al., 2013 from eastern Cuba; B, P. ondreji sp. nov. from Jamaica; C, P. laterale sp. nov. from southern Hispaniola; D, P. borinquenum sp. nov. from Puerto Rico. E, F, non-endemic species: E, P. exstriatum (Say, 1835) from Cuba; F, Phaenonotum sp. from the P. laevicolle complex from Cuba. Figure 5. View largeDownload slide Habitus of the Caribbean Phaenonotum (dorsal and lateral views). A–D, species endemic for the Greater Antilles: A, P. delgadoi Deler-Hernández et al., 2013 from eastern Cuba; B, P. ondreji sp. nov. from Jamaica; C, P. laterale sp. nov. from southern Hispaniola; D, P. borinquenum sp. nov. from Puerto Rico. E, F, non-endemic species: E, P. exstriatum (Say, 1835) from Cuba; F, Phaenonotum sp. from the P. laevicolle complex from Cuba. Figure 6. View largeDownload slide Head in frontal view (A–F) and aedeagus (G–L) of the Caribbean Phaenonotum. A, G, P. delgadoi Deler-Hernández et al., 2013; B, H, P. ondreji sp. nov.; C, I, P. laterale sp. nov.; D, J, P. borinquenum sp. nov.; E, K, P. exstriatum (Say, 1835); F, L, Phaenonotum sp. from the P. laevicolle complex from Cuba. A–F, not to scale. Figure 6. View largeDownload slide Head in frontal view (A–F) and aedeagus (G–L) of the Caribbean Phaenonotum. A, G, P. delgadoi Deler-Hernández et al., 2013; B, H, P. ondreji sp. nov.; C, I, P. laterale sp. nov.; D, J, P. borinquenum sp. nov.; E, K, P. exstriatum (Say, 1835); F, L, Phaenonotum sp. from the P. laevicolle complex from Cuba. A–F, not to scale. Figure 7. View largeDownload slide Meso-metaventral morphology of the Caribbean Phaenonotum (left: middle and right portions of complete meso-metaventrite, right: detail of mesoventral elevation and anterior metaventral process). A, P. delgadoi Deler-Hernández et al., 2013; B, P. laterale sp. nov.; C, P. ondreji sp. nov.; D, P. borinquenum sp. nov.; E, P. exstriatum (Say, 1835); F, Phaenonotum sp. from the P. laevicolle complex from Cuba. Not to scale. Figure 7. View largeDownload slide Meso-metaventral morphology of the Caribbean Phaenonotum (left: middle and right portions of complete meso-metaventrite, right: detail of mesoventral elevation and anterior metaventral process). A, P. delgadoi Deler-Hernández et al., 2013; B, P. laterale sp. nov.; C, P. ondreji sp. nov.; D, P. borinquenum sp. nov.; E, P. exstriatum (Say, 1835); F, Phaenonotum sp. from the P. laevicolle complex from Cuba. Not to scale. Phaenonotum borinquenum sp. nov. (Figs 4A, 5D, 6D, J, 7D) Type locality: Puerto Rico, Naguabo, El Yunque National Forest, 18°16.1′N 65°48.1′W, 575 m. Type material: Holotype: male (NMPC): ‘PUERTO RICO: Naguabo: El Yunque Nat. Forest, S part La Sabana recr area 5.25 km N of Río Blanco above rd. PR191, 18°16.1′N 65°48.1′W, elevation 575 m, 21.vi.2016, A. Deler-Hernández lgt., PR05/DNA isolation: MF1730, isolated by A. Deler-Hernández 2016, isolate deposited at Department of Entomology, National Museum in Prague (green printed)’ (molecular voucher MF1730). Paratype: one female (NMPC; associated with the holotype by cox1 sequence): Puerto Rico: Naguabo, El Yunque Nat. Forest, S part La Sabana recr area 5.6 km N of Río Blanco at rd. PR191, 18°16.1′N 65°47.6′W, 510 m a.s.l., 21.vi.–2.vii.2016, Deler-Hernández, Fikáček, Seidel lgt., PR03/DNA isolation: MF1729, isolated by A. Deler-Hernández 2016, isolate deposited at Department of Entomology, National Museum in Prague (green printed)’ (Molecular voucher MF1729). Description: Habitus as in Figure 5D. Body size 3.2–3.3 mm (holotype: 3.3 mm). Body elongate oval, moderately convex, elytral suture not elevated. Dorsum black (dark brown in the teneral paratype), lateral margins of pronotum and elytra without distinct paler stripe; ventral surface dark brown to black; femora and tibiae black; antennae, maxillary palpi and tarsi yellowish. Head with sparse and moderately coarse punctation, without microsculpture (except posteriorly on frons); eyes moderately large, separated by 4.0× dorsal width of one eye (Fig. 6D). Pronotum with sparse punctures slightly finer than on head, interstices without microsculpture. Elytral punctation strongly impressed, coarser than on pronotum and head; elytral interstices without microsculpture. Wings present, fully developed. Mesoventral elevation as wide as metaventral process posteriorly, narrowing anteriorly, with distinct anterior hood; metaventrite without pubescent pits on sides of metaventral process; metaventrite c. 4.0× wider than its length posterior of mesocoxae; median moderately elevated part of metaventrite narrow throughout (Fig. 7D). Aedeagus 0.5 mm long (Fig. 6J). Median lobe rather widely triangular, c. 1.8× longer along midline than wide; apex not reaching apices of parameres; gonopore moderately large, subapical; lateral struts projecting laterad. Parameres distinctly sinuate on lateral margin, slightly expanded subapically; widely meeting each other basally. Phallobase longer than wide. Etymology: The species name is a Latinized adjective derived from the Spanish version of the indigenous Taíno name of the islands of Puerto Rico. Diagnosis: Phaenonotum borinquenum may be distinguished from other Caribbean species by the combination of moderately large eyes, moderately long metaventrite, presence of wings and morphology of the aedeagus. It is extremely similar to P. exstriatum in its external morphology and genital morphology, and may be distinguished from it only by its smaller body size and widely meeting bases of parameres only. Despite the strong morphological similarity, it is not closely related to P. exstriatum. Distribution: Phaenonotum borinquenum is only known from the eastern part of Puerto Rico (Fig. 4A). Phaenonotum delgadoi Deler-Hernández, Cala-Riquelme & Fikáček, 2013 (Figs 4A, 5A, 6A, G, 7A) Type material examined: See Deler-Hernández et al. (2013), including one sequenced paratype from the type locality (molecular voucher MF455). Additional material examined:Cuba: Holguín Prov.: 5 spec. (NMPC): La Melba, Parque Nacional Alexander von Humboldt, 20.43352°N 74.82507°W, 336 m, rainforest litter, 21.ix.2014, R. Anderson, F. Cala-Riquelme, A. Deler-Hernández lgt. (2014-002); 4 spec. (NMPC): 20.45396°N 74.82342°W, 510 m, pluviselva litter, 22.ix.2014, R. Anderson, F. Cala-Riquelme, A. Deler-Hernández lgt. (2014-011); 19 spec. (NMPC): road out La Melba, 20.51524°N 74.81844°W, 407 m, elfin forest litter, 24.ix.2014, R. Anderson, F. Cala-Riquelme, A. Deler-Hernández lgt. (2014-015); 8 spec. (NMPC): road out La Melba, road out of La Melba, 20.59086°N 74.83627°W, 130 m, scrub forest litter, 24.ix.2014, R. Anderson, F. Cala-Riquelme, A. Deler-Hernández lgt. (2014-016). Published records: Cuba: Guantánamo Prov.: El Yunque de Baracoa (Deler-Hernández et al., 2013a). Holguín Prov.: La Melba (Deler-Hernández et al., 2013a). Redescription: Habitus as in Figure 5A. Body length 2.2–2.5 mm (holotype: 2.3 mm). Body oval, strongly convex, elytral suture distinctly elevated posteriorly. Dorsum black, lateral margins of pronotum and elytra without distinct paler stripe; ventral surface brown to dark brown; femora and tibiae reddish; antennae, maxillary palpi and tarsi yellowish. Head with sparse fine punctation, without microsculpture (except posteriorly on frons); eyes small, separated by 5.2× dorsal width of one eye (Fig. 6A). Pronotum with very sparse minute punctures much smaller than on head, interstices without microsculpture. Elytral punctation very strongly impressed, much coarser than on pronotum and head; elytral interstices without microsculpture. Wings completely absent. Mesoventral elevation as wide as metaventral process throughout, not narrowing anteriorly, with distinct anterior hood; metaventrite with a deep pubescent pit on each side of metaventral process; metaventrite c. 5.8× wider than its length posterior of mesocoxae; median moderately elevated part of metaventrite narrow anteriorly, widening posteriorly (Fig. 7A). Aedeagus 0.4 mm long (Fig. 6G). Median lobe rather narrowly triangular, c. 1.9× longer along midline than wide; apex not reaching apices of parameres; gonopore small, subapical; lateral struts simple, not expanded. Parameres indistinctly sinuate, nearly evenly arcuate on lateral margin, not expanded subapically, broadly meeting each other basally. Phallobase slightly wider than long. Diagnosis: Phaenonotum delgadoi may be distinguished from all other Phaenonotum known to us by the combination of extremely coarse elytral punctation, elevated elytral suture and deep pubescent pits on each side of metaventral process. See Identification Key for additional characters. Distribution: Phaenonotum delgadoi is only known from eastern Cuba, all known localities are situated in the Nipe-Sagua-Baracoa mountain system (Fig. 4A). Phaenonotum exstriatum (Say, 1835) (Figs 4B, 5E, 6E, K, 7E) Type material: Neotype of Hydrophilus exstriatus Say, 1835 designated by Smetana (1978) from southeastern states of USA (deposited in Museum of Comparative Zoology, Harvard University, Boston, USA): not examined in our study. Types of Phaenonotum dubium Sharp, 1882 were examined by Deler-Hernández & Fikáček (2016) and confirmed as being conspecific to North American specimens treated as P. exstriatum (synonymy proposed by Smetana, 1978). Material examined:Barbados: 1 spec. (SBPC): Jack-in-Box Gully, forest, 13°11’N 59°34.3’W, 230 m, UV light, 23.vi.2007, S & J. Peck lgt. (07-25). Cuba: Cienfuegos prov.: 1 spec. (NMPC): Soledad, 22.12682°N 80.33289°W, 71 m, MV light, 21.v.2013, A.B.T. Smith lgt. Guantánamo prov.: 2 spec. (NMPC): Baracoa, Jobo Dulce, 20°18′18.00″N 74°27′21.60″W, 94 m, 29.x.2010, R. Correa lgt; El Yunque, 3.2 km SW of campismo, 20°19′N 74°34′W, 150 m, 13.vi.2012, A. Deler-Hernández lgt. Holguín Prov.: 1 spec. (NMPC): Mayarí, Guatemala, Guarina Nuñez, 20°42′4.06″N 75°40′24.45″W, c. 27 m, 26.iii.2013, A. Deler-Hernández lgt. Granma Prov.: 1 spec. (NMPC): Cauto Cristo, Rio Cauto, El Sitio, 20°16′22.80″N 76°29′2.40″W, 135 m, 1.v.2005, L. Chaves. Santiago de Cuba prov.: 9 spec. (NMPC): El Vivero, 1.3 km NEE of Dos Caminos, 20°10′49.36″N 75°46′41.43″W, c. 170 m, at light, 23.v.2013, Deler-Hernández & A. Smith lgt. (1 spec: molecular voucher MF654); 1 spec. (NMPC): Daiquirí, 1–4.vi.1985, S. Bílý lgt. Dominica: 8 spec. (BMNH): La Plaine, 97-67, 22.i.1889, G. A. Ramage lgt. Dominican Republic: 1 spec. (NMPC): Samaná, 6.4 km N of Samaná, road to El Valle, 19°15.78′N 69°20.23′W, 21 m, 5.ix.2014, Deler, Fikáček, Gimmel (DR36). Grenada: 1 spec. (SBPC): Grand Etang Forest Reserve, 12°04.952′N 61°42.162′W, 434 m, nursery edge forest, UV trap, 14.viii.2010, S. Peck lgt. (10–69); 25 spec. (BMNH): Grand Etang (Windward side), 1900 ft, H. H. Smith.; 1 spec. (SBPC): Par. St. Andrew, Mirabeau, Agric. Lab, UV trap, 23.ii.1990, R. E. Woodruff lgt.; 34 spec. (BMNH): Mount Gay Est. (Leeward side), H. H. Smith lgt.; 3 spec. (BMNH): Chantilly Est. (Windward side), H. H. Smith lgt.; 8 spec. (BMNH): Balthazar (Windward side), H. H. Smith lgt.; 1 spec. (BMNH): Vendome Est. (Leeward side), H. H. Smith lgt. Cayman Islands: 1 spec. (BMNH): Grand Cayman, by freshwater lake near George Town., UV light trap, 4.viii.1970, Joy Farradane lgt. Haiti: 1 spec. (BMNH): Port au Prince., 1.iii.1908, M. Cameron lgt. Jamaica: 1 spec. (BMNH): Kingston., 16.ii.1908, M. Cameron lgt.; 9 spec. (SBPC): Ewarton, St. Cath. Par., St. Clair Cave, 27.xii.1972, S & J. Peck. Monserrat: 1 spec. (BMNH): Salem 8.ix.1975, J. Cooter lgt. Saint Lucia: 1 spec. (SBPC): Mon Repos, 6.5 km W of Fox Grovelnn, 13°52.5′N 60°56.4′W, 300 m, submontane forest litter, 22.vii.2007, S & J. Peck lgt. (07-76). Saint Vincent and the Grenadines: 1 spec. (SBPC): Saint Vincent, Brighton Bay Village, 13°07.97′N 61°10.06′W, 1 m, streamside UV trap 8.vi.2007, S & J. Peck lgt. (07-10); 15 spec. (BMNH): Saint Vincent, without additional data, H. H. Smith lgt. Puerto Rico: 7 spec. (NMPC): Naguabo, El Yunque Nat. Forest, S part 3.45 km N of Río Blanco at rd. PR191, 18°14.8′N 65°47.7′W, 170 m, 24.vi.2016, Deler, Fikáček & Seidel lgt. (1 spec.: molecular voucher MF1728). Trinidad & Tobago: 2 spec. (BMNH): Trinidad, St. Augustine, 15.iv.1926, C. L. Withycombe lgt.; 1 spec. (BMNH): Trinidad, St. Margarita, 11.vi.1942. E. C. Humphries lgt. USA: Florida: 1 spec. (NMPC): Highlands Co., Venus, 4 miles W of Fish Eating Creek, 15.viii.1965, W. Suter lgt., det. A. Smetana; 1 spec. (NMPC): Alachua Co., Gainesville, black light, 19.vii.1978, F. N. Young lgt., det. A. Smetana; 1 spec. (NMPC): Kansas: Douglas Co., Bridenthal Ecological Reserve, 3.2 miles N of Baldwin City, 38.81043°N 95.18669°W, 270 m, ix.2009, Eldgedge & Fikáček lgt.; 1 spec. (NMPC): Douglas Co., Lawrence Baker Wetlands, 27.viii.2009, Gustafson, Eldgedge & Fikáček lgt.; 1 spec. (NMPC): Delaware: New Castle Co., Frenchtown Woods Natural Area, 23.v.2004, at light, A. E. Z. Short lgt. (AS-04-065) (molecular voucher MF1063). Costa Rica: 1 spec. (NMPC): Guanacaste: 6.6 km from main road, roadside gravel stream running through culvert, 10°09′26.7″N 85°22′47.5″W, 50 m, 13.i.2004, Short & Lebbin lgt. (AS-04-037) (molecular voucher MF1738). Redescription: Habitus as in Figure 5E. Body length 3.3–4.0 mm. Body elongate oval, moderately convex, elytral suture not elevated posteriorly. Dorsum black, lateral margins of elytra with very narrow indistinct paler stripe, pronotum paler in posterolateral corners; ventral surface dark brown to black; femora and tibiae black; antennae, maxillary palpi and tarsi yellowish. Head with sparse moderately coarse punctation, without microsculpture (except posteriorly on frons); eyes moderately large, separated by 4.0× dorsal width of one eye (Fig. 6E). Pronotum with sparse fine punctures, slightly smaller than on head, interstices without microsculpture. Elytral punctation moderately impressed, coarser than on pronotum and head; elytral interstices without microsculpture. Wings present, well developed. Mesoventral elevation as wide as metaventral process posteriorly, slightly narrowing anteriorly, with distinct anterior hood; metaventrite without pubescent pits on sides of metaventral process; metaventrite c. 3.9× wider than its length posterior of mesocoxae; median moderately elevated part of metaventrite moderately wide throughout (Fig. 7E). Aedeagus 0.4 mm long (Fig. 6K). Median lobe narrowly triangular, c. 2.1× longer along midline than wide; apex reaching apices of parameres; gonopore small, subapical; lateral struts expanded laterally. Parameres distinctly sinuate on lateral margin, slightly expanded subapically. Phallobase longer than wide. Diagnosis: Phaenonotum exstriatum may be distinguished from Caribbean species except P. borinquenum by the combination of moderately large eyes, moderately long metaventrite, presence of wings and morphology of the aedeagus. It may be distinguished from P. borinquenum by larger body size and bases of parameres meeting in single point only. Despite the strong morphological similarity, it is not closely related to P. borinquenum (see Fig. 2). Distribution: Phaenonotum exstriatum is widespread in the eastern USA and in Central America (Smetana, 1978; Deler-Hernández & Fikáček, 2016) as well as in Greater and Lesser Antilles (Fig. 4B). The only record from South America is from eastern Colombia Gonzáles-Rodríguez, García-Hernández & Clarkson (2017). but the species is probably widespread in northern South America as it is common in southern islands of Lesser Antilles and it also occurs in Trinidad. Phaenonotum laevicolle Sharp, 1882 complex (Figs 4C, 5F, 6F, L, 7F) Type material examined: Types of P. laevicolle Sharp, 1882, see Deler-Hernández & Fikáček (2016). Material examined:Cuba: Cienfuegos Prov.: 1 male (NMPC): Río Cabagan, Gruta Mengoa, 21.93123°N 80.08461°W, 651 m, 20.v.2014, A. Deler-Hernández lgt. (molecular voucher MF 1115); 3 spec. (ZMHB): Sierra del Escambrai, 5 km N Topes de Collantes, c. 600 m, sifting of leafs, 17.xii.2007, M. Schülke lgt. (CU7-4). Sancti Spíritus Prov.: 9 spec. (ZMHB): Sierra del Escambrai, Topes de Collantes, c. 700 m, Streu, Totholz, 17.xii.2007, M. Schülke lgt. (CU7-3). Granma Prov.: 7 spec. (ZMHB): Sierra Maestra, La Habanita, 35 km NE Pilón, 900–1000 m, sifted hay, 20.xii.2007, M. Schülke lgt. (CU7-8). Santiago de Cuba Prov.: 1 spec. (NMPC): El Vivero, 1.6 km E of Dos Caminos, 20°10.8′N 75°46.4′W, 150 m, 20–21.vi.2012, A. Deler-Hernández & Fikáček (MF18). Haiti: 3 spec. (BMNH): Port au Prince, 1.iii.1908, M. Cameron lgt. Venezuela: 1 male (NMPC): Monagas small pond between Morichal Largo & Temblador, 9°05′47.9″N 62°43′37.1″W, 29 m, 2.ii.2010, Short, García & Joly lgt. (VZ10-0202-03A) (molecular voucher MF1740). Redescription: (refers to the Caribbean specimens examined): Habitus as in Figure 5F. Body length 2.7–3.1 mm. Body elongate oval, moderately convex, elytral suture not elevated posteriorly. Dorsum black, lateral margins of elytra with paler stripe reaching subapically, pronotum paler in posterolateral corners; ventral brown to dark brown; femora dark brown to reddish, tibiae reddish; antennae, maxillary palpi and tarsi yellowish. Head with sparse fine punctation, without microsculpture (except posteriorly on frons); eyes moderately large, separated by 3.6× dorsal width of one eye (Fig. 6F). Pronotum with sparse fine punctures similar to that on head, interstices without microsculpture. Elytral punctation moderately impressed, coarser than on pronotum and head; elytral interstices without microsculpture. Wings present, well developed. Mesoventral elevation slightly narrower than metaventral process posteriorly, slightly narrowing anteriorly, with distinct anterior hood; metaventrite without pubescent pits on sides of metaventral process; metaventrite c. 3.5× wider than its length posterior of mesocoxae; median moderately elevated part of metaventrite moderately wide throughout (Fig. 7F). Aedeagus 0.5 mm long (Fig. 6L). Median lobe rather triangular, c. 1.7× longer along midline than wide; apex reaching apices of parameres; gonopore large, wide, situated in apical third of median lobe; lateral struts very shortly expanded laterally. Parameres distinctly sinuate on lateral margin, strongly expanded subapically, broadly meeting each other basally. Phallobase longer than wide. Comments: Both sequenced specimens (MF1115 from Cuba and MF1740 from Venezuela) form a strongly supported clade in the molecular analysis and are evidently closely related (pairwise distance of cox1 sequences is 4.8%). Both specimens correspond with each other in external morphology and the characteristic shape of male genitalia, and only differ in body size: the Cuban specimen is smaller (2.8 mm), the Venezuelan specimen larger (3.4 mm). In this aspect, the sequenced Cuban specimen corresponds to all additional Greater Antillean specimens examined, which are also rather small (2.7–3.1 mm). Hence, it seems probable that Venezuelan specimen is not conspecific with the Greater Antillean ones, but additional material is necessary to evaluate the intraspecific genetic and morphological variability properly to decide whether the sequenced specimens represent one or two species. The external morphology, body size and the morphology of genitalia of the Caribbean specimens correspond to the types of P. laevicolle Sharp, 1882 described from Guatemala and examined by Deler-Hernández & Fikáček (2016). However, due to the absence of DNA-grade specimens of P. laevicolle from Central America, we are unable to determine whether the Greater Antillean specimens are conspecific. For that reason, we consider all above specimens as members of the Phaenonotum laevicolle species complex whose taxonomy needs to be addressed once more material will be available. Distribution: The Caribbean specimens of the Phaenonotum laevicolle complex are known from central and eastern Cuba and western Hispaniola (Haiti). Outside the Caribbean, the species complex clearly occurs in southernmost Northern and Central Americas (types of P. laevicolle) and in northern South America [as P. globulosum (Mulsant, 1844) from Colombia (Hansen, 1999; not examined by us) and the Venezuelan specimen sequenced by us]. The records from Argentina (types of P. spegazziinii Bruch, 1915 not examined by us) seem doubtful (Oliva et al., 2002) and are not considered here. Based on these sources, we estimate the distribution of the species complex to be as shown in Figure 4C. Phaenonotum laterale sp. nov. (Figs 4A, 5C, 6C, I, 7B) Type locality: Dominican Republic, Barahona, Monumento Natural Miguel Domingo Fuerte ‘Cachote’ 18°5.91′N 71°11.35′W, 1188 m. Type material: Holotype: male (NMPC): ‘Dominican Republic: Barahona, MN Domingo Fuerte “Cachote,” 18°5.91′N 71°11.35′W, 1188 m, 14.viii.2014, Deler, Fikáček, Gimmel DR03/montane broadleaf cloud forest with numerous Cyathea: sifting of thin layer of wet leaf litter and mosses’. Paratypes: Dominican Republic: 10 spec. (NMPC): same data as holotype (including molecular voucher MF1013); 25 spec. (CMN, NMPC, MNHNSD, NHMW): MN Domingo Fuerte ‘Cachote’, 18°5.21′N 71°11.46′W, 1205 m, sparse montane cloud forest with ferns and mosses here and there, sifting of thin layer of leaf litter, 14.viii.2014, M. Fikáček lgt. (DR03a); 49 spec. (BMNH, DZRJ, MCZ, MNHNSD, NMPC, SBMN): MN Domingo Fuerte ‘Cachote’, 18°4.48–5.37′N 71°11.03–11.54′W, 1188 m, 14.viii.2014, secondary montane broadleaf forest with sparse ferns and moss, sifting of leaf litter, Deler-Hernández, Fikáček & Gimmel lgt. (DR04). Description: Habitus as in Figure 5C. Body length 2.5–2.7 mm (holotype: 2.7 mm). Body oval, moderately convex, elytral suture slightly elevated posteriorly. Dorsum brown to dark brown, lateral margins of pronotum and lateral and sutural margins of elytra with distinct yellowish stripe; ventral surface brown to reddish brown; femora and tibiae reddish; antennae, maxillary palpi and tarsi yellowish. Head with sparse fine punctation, without microsculpture (except posteriorly on frons); eyes small, separated by 5.1× dorsal width of one eye (Fig. 6C). Pronotum with very sparse and very fine punctures much smaller than on head, interstices with not very distinct mesh-like microsculpture. Elytral punctation sparse and moderately impressed, similar to that on head, much coarser than on pronotum; elytral interstices without microsculpture. Wings completely absent. Mesoventral elevation extremely narrow, as wide as metaventral process throughout, not narrowing anteriorly, without distinct anterior hood; metaventrite without deep pubescent pits at sides of metaventral process; metaventrite c. 5.0× wider than its length posterior of mesocoxae; median weakly elevated part of metaventrite narrow throughout (Fig. 7B). Aedeagus 0.35 mm long (Fig. 6I). Median lobe rather narrowly triangular, c. 2.1× longer along midline than wide; apex reaching apices of parameres; gonopore small, subapical; lateral struts simple, not expanded. Parameres indistinctly sinuate on lateral margin, not expanded subapically, widely meeting each other basally. Phallobase as long as wide. Etymology: The species name refers to the conspicuous yellow stripe along lateral margins of pronotum and elytra characteristic of this species. Diagnosis: Phaenonotum laterale may be distinguished from other Caribbean species, based on its brown-yellow coloration, obsolete pronotal punctation and extremely narrow meso-metaventral keel. See Identification Key for further diagnostic characters. Distribution: Phaenonotum laterale sp. nov. is only known from a high altitude cloud forest region in the southwestern Dominican Republic, geologically situated on the southern paleoisland of Hispaniola (Fig. 4A). Phaenonotum ondreji sp. nov. (Figs 4A, 5B, 6B, H, 7C) Type locality: Jamaica, Blue Mountains National Park, 18°5′14.13″N 76°43′37.55″W, 1279 m. Type material: Holotype: male (NMPC): Jamaica: Blue Mountain, 18°5′14.13″N 76°43′37.55″W, 1279 m, 14.xi.2013, F. Cala-Riquelme leg.’. Paratypes: Jamaica: 19 spec. (CMN, MCZ, NMPC, NHMW) same data as holotype (including molecular voucher MF980); 2 spec. (BMNH): Newcastle, 19–22.viii.1908, M. Cameron lgt. Description: Habitus as in Figure 5B. Body length 3.2–3.4 mm (holotype: 3.1 mm). Body oval, highly convex, elytral suture slightly elevated posteriorly. Dorsum dark brown to black, posterolateral corners of pronotum and apical part of lateral margins of elytra distinctly yellowish; ventral surface brown to reddish brown; femora and tibiae yellowish to reddish; antennae, maxillary palpi and tarsi yellowish. Head with sparse moderately coarse punctation, with mesh-like microsculpture; eyes small, separated by 6.9× dorsal width of one eye (Fig. 6B). Pronotum with sparse and fine punctures, smaller than on head, interstices with distinct mesh-like microsculpture. Elytral punctation sparse, moderately impressed, coarser than on head and pronotum; elytral interstices without microsculpture. Wings completely absent. Mesoventral elevation very narrow, as wide as metaventral process throughout, not narrowing anteriorly, without distinct anterior hood; metaventrite without deep pubescent pits at sides of metaventral process; metaventrite c. 5.1× wider than its length posterior of mesocoxae; median highly elevated portion narrow throughout (Fig. 7C). Aedeagus 0.6 mm long (Fig. 6H). Median lobe very narrowly triangular, c. 2.3× longer along midline than wide; apex slightly overlapping apices of parameres; gonopore small, subapical; lateral struts weakly expanded. Parameres sinuate on lateral margin, weakly expanded subapically, narrowly meeting each other basally. Phallobase as long as wide. Etymology: This species is named in honour of Ondřej Jelínek, a good friend of the senior author, in appreciation of his friendship. Diagnosis: Phaenonotum ondreji may be distinguished from other Caribbean species by the combination of small eyes, head and pronotum with distinct mesh-like microsculpture, and narrow highly elevated meso-metaventral keel. See Identification Key for additional diagnostic characters. Distribution: Phaenonotum ondreji sp. nov. is only known from the eastern part of the island (Blue Mountains range, Fig. 4A). DISCUSSION Biogeography of Phaenonotum and origin of Caribbean species Caribbean lineages of Phaenonotum diverged from their sister taxa during the middle Eocene to early Pleistocene (c. between 47 and 2.4 Mya), that is long after the bolide impact to the Caribbean region at the Cretaceous-Paleogene (K-Pg) boundary c. 65.5 Mya (Schulte et al., 2010). This is incongruent with the vicariance hypothesis of Rosen (1975, 1985), which assumes that Caribbean lineages originated by the colonization of the proto-Antillean volcanic arc in the Late Cretaceous and survived the K-Pg boundary. The Cenozoic origin of Phaenonotum lineages however corresponds to other insect and spider groups studied so far, of which Caribbean clades were dated to originate between the Eocene and the Miocene (Wahlberg, 2006; Wahlberg & Freitas, 2007; Ceccarelli & Zaldívar-Riverón, 2013; Zhang & Maddison, 2013; Matos-Maraví et al., 2014; McHugh et al., 2014; Dziki et al., 2015; Lewis et al., 2015; Rodriguez et al., 2015; Zhang et al., 2017). The Caribbean Phaenonotum fauna originated from three to four independent colonization events from northern South America, the earliest of which correspond(s) in timing to the timespan of the GAARlandia land bridge connecting northern South America and the Greater Antilles through the emerged Aves Ridge at the Eocene–Oligocene boundary (Iturralde-Vinent & MacPhee, 1999; Iturralde-Vinent, 2006). Subsequent colonizations happened during the late Oligocene to early Miocene after the Antillean archipelago lost direct connection to the continent (Iturralde-Vinent & MacPhee, 1999; Iturralde-Vinent, 2006). Hence, the evolutionary history of Phaenonotum may be best explained by the combination of colonization via a land bridge in some lineages and the over-water dispersal events (i.e. founder events) in others. The oldest colonization of Phaenonotum to the Caribbean gave rise to a small clade today containing three single-island endemics: P. delgadoi from eastern Cuba, P. ondreji from eastern Jamaica and P. laterale from southern Hispaniola. The divergence of this clade from modern continental clades slightly predated the timespan of the GAARlandia land bridge, which allows for two possible scenarios of colonizations: single colonization of the Greater Antilles via GAARlandia and subsequent split into multiple species in the archipelago, or parallel colonizations of the Greater Antilles via GAARlandia by several species of the clade and subsequent speciation in the archipelago (our BioGeoBEARS analyses propose both scenarios as possible, depending on the model and time-stratification used and whether a founder event is allowed or not). Naturally, we cannot totally exclude the over-water dispersal in case it coincided by age or slightly predated the timespan of the land bridge (Poux et al., 2006; de Queiroz, 2016). However, unlike all other Phaenonotum examined, all three species of this clade are wingless, indicating that their MRCA probably had no metathoracic wings either. The expected limited dispersal abilities of the members of the clade together with the time overlap of their arrival to the Greater Antilles with the timespan of the GAARlandia land bridge make the scenario of the over-land colonization of the Greater Antilles more probable than the over-water dispersal alternative. The divergence of the Cuban P. delgadoi from the remaining two species of the clade was dated to the Eocene–Oligocene boundary (c. 36 Mya), coinciding with the maximum landspan of the GAARlandia and pre-dating the subdivision of the Greater Antilles into particular palaeoislands (separation of Puerto Rico from Hispaniola + Cuba by Mona Passage c. 30–20 Mya and separation of eastern Cuba from central Hispaniola by Windward Passage c. 17–14 Mya; Iturralde-Vinent & MacPhee, 1999; Matos-Maraví et al., 2014). Both clades may have split while colonizing separate mountain ranges of the GAARlandia peninsula (Iturralde-Vinent & MacPhee, 1999; Iturralde-Vinent, 2006) or before colonizing the peninsula. In contrast, the divergence of the Hispaniolan P. laterale and Jamaican P. ondreji was dated to the Late Oligocene (c. 26 Mya) when only the Blue Mountain range of Jamaica was emergent and widely isolated from other islands (Iturralde-Vinent & MacPhee, 1999; Iturralde-Vinent, 2006). This implies a long-distance over-water founder event, improbable for wingless mountain species. Interestingly, Jamaica-Hispaniola divergences of similar age are also found in some other animal groups (c. 22 Mya in Osteopilus tree frogs: Moen & Wiens, 2009; c. 20 Mya in Exophthalmus weevils: Zhang et al., 2017; c. 15 Mya in Calisto butterflies: Matos-Maraví et al., 2014). This may indicate either much closer geological connection between Hispaniola and Jamaica in the Late Oligocene to Early Miocene, or a constellation largely facilitating Hispaniola to Jamaica dispersal (e.g. by strong sea currents or hurricanes; Hedges, 2006). Puerto Rico was a part of the GAARlandia land bridge, but no species of the above Phaenonotum clade has been recorded from the island, in contrast to the predictions of the over-land GAARlandia scenario. Similarly, the endemic occurrence of the Hispaniolan P. laterale in the southern part of the island that emerged in the Late Miocene, largely postdating the origin of the species (26.4 Mya, Late Oligocene) also disagrees with the current model of historical geography of the Greater Antilles. This indicates that range shifts and extinctions probably played a significant role in the evolutionary history of Phaenonotum beetles in the Caribbean. On the other hand, the absence of the members of this Phaenonotum clade from the Lesser Antilles is congruent with the GAARlandia hypothesis: older Lesser Antilles islands are of c. Oligocene to Miocene origin, the younger ones arose during the Pliocene only and none of them were part of the GAARlandia land bridge, which was situated slightly more to the west in the place of the nowadays submerged Aves Ridge (Iturralde-Vinent & MacPhee, 1999; Thorpe et al., 2004; Iturralde-Vinent, 2006). The remaining Phaenonotum clades probably colonized the Caribbean by two or three independent events during the Late Oligocene to Pliocene by over-water dispersal. One of these events (dated to Early Miocene, c. 19 Mya) gave rise to the Puerto Rican P. borinquenum, that is the only single-island endemic, which is not a member of the Caribbean endemic lineage discussed above. Both widespread continental-Caribbean species (P. laevicolle and P. exstriatum) were estimated to originate in the Caribbean and back-colonized the continent during the Pliocene and Pleistocene in the time-stratified BioGeoBEARS analyses. Our ancestral range estimates may be, however, strongly influenced by incomplete sampling of continental Phaenonotum species and limited knowledge on their distribution, and additional studies are hence necessary to corroborate our results. Nevertheless, our data suggest that the Caribbean Region may act not only as a sink of colonizations from the continent, but also a source of taxa colonizing North and Central America. This corresponds to the situation recently revealed for the Exophthalmus weevils and for Heracleidos butterflies, both of which colonized Central America from the Caribbean during the Miocene to Pleistocene (Lewis et al., 2015; Zhang et al., 2017); multiple examples are also known for vertebrates (Bellemain & Ricklefs, 2008; Ricklefs & Bermingham, 2008). Moreover, Central America was an archipelago similar to todayʼs Greater Antilles during the Miocene to Pliocene, and Caribbean species hence probably colonized islands with unbalanced fauna rather than a continent with diverse well-established fauna characterized by strong interspecific competition (Bellemain & Ricklefs, 2008). All four single-island endemic Phaenonotum species inhabit leaf litter in lowland (P. delgadoi) or montane cloud forests (P. ondreji, P. laterale and P. borinquenum), the primary forest habitats in the island interior. In contrast, the remaining two species are more widespread (P. exstriatum is recorded from all Greater Antillean islands including the Cayman Islands, and from most islands of the Lesser Antilles, while P. laevicolle is known from Cuba and Hispaniola; Fig. 4), and inhabit disturbed lowland habitats (P. exstriatum aquatic and semiaquatic habitats and decaying plant material, P. laevicolle decaying plant material including hay piles and leaf litter in farmland). When biology and distribution data are correlated to the age of the respective species, they are in a good agreement with the taxon cycle hypothesis proposed originally by Wilson (1959, 1961) for Melanesian ants and subsequently confirmed, for example, for Lesser Antillean birds (Ricklefs & Cox, 1972) (see Ricklefs & Bermingham, 2002 for review and multiple examples, and Economo & Sarnat, 2012 for detailed re-analysis of taxon cycle in Melanesian ant fauna). Following the predictions of the taxon cycle, the recently colonizing species (P. laevicolle and P. exstriatum) inhabit wider spectrum of lowland disturbed habitats and are good dispersers (range expansion phase), whereas the locally endemic species are remnants of ancient colonizations inhabiting exclusively inland primary habitats and having poor dispersal abilities (they switched from the expansion phase at the time of colonization into the range contraction phase; this may be followed by another range expansion phase or by extinction). Species-level systematics of Phaenonotum Despite the beetles being frequently collected, the systematics of Phaenonotum beetles has not been studied properly until now, and very little is known about the continental fauna of the genus. Sixteen species are known from the continental Neotropics and two from the Nearctic (with the widespread P. exstriatum inhabiting both regions). The real species diversity seems to be much higher, which is illustrated, for example, by the fact that of the c. 13 continental species included in our analysis, we were only able to identify two of them (P. exstriatum and P. laevicolle), whereas remaining ones probably represent undescribed species. The extreme morphological similarity of the species and limited number of characters useful for species-level identification are clearly the main obstacles for species-level studies and identification of Phaenonotum. Moreover, extreme similarity in external morphology and male genitalia does not necessarily indicate a close phylogenetic relationship between species. This is illustrated by P. exstriatum and P. borinquenum sp. nov., which are not closely related, yet are nearly identical in morphological characters. In contrast, the P. laevicolle complex seems closely related to P. borinquenum but can be very easily distinguished from it by genital morphology (compare Fig. 6L, J). The combination of morphology and molecular data seems hence necessary for proper understanding of species limits even in distantly related species in Phaenonotum. The same is naturally true for closely related similar species. This is the reason why we refrain from describing the Cuban and Hispaniolan species of the P. laevicolle complex and do not decide whether the sequenced specimens from Cuba and Venezuela belonging to this species complex represent one or two species. Sequences of freshly collected specimens from the type locality of P. laevicolle or its surroundings and from multiple specimens across the range of the species complex would be necessary to analyse the species limits in this case. Neotropical Coelostomatini and the position of Phaenonotum Despite limited taxon sampling, our analyses indicate that the Neotropical endemic genera of Coelostomatini (Cyclotypus, Phaenonotum, Phaenostoma, Lachnodacnum) form a monophyletic group. This result corresponds with topology revealed in the phylogenetic analysis containing a wider spectrum of coelostomatine genera but fewer Neotropical taxa (V. Sýkora, unpubl. data) and suggests that the early evolution of the Coelostomatini was strongly influenced by paleogeography: the origin of this clade was revealed as Early-Middle Cretaceous in our analysis when South America was largely isolated from other continents. The internal topology of the Neotropical clade disagrees with current generic concepts, which are based on the combination of a few easy-to-observe morphological characters (Hansen, 1991; Clarkson et al., 2014): Phaenonotum is revealed polyphyletic when comprising the morphologically aberrant P. caribense, and the bromeliad inhabiting genus Lachnodacnum is nested within the polyphyletic Phaenostoma. The fact that the aberrant morphology corresponds to a separate phylogenetic position in one taxon (Phaenonotum caribense) but to biology rather than phylogenetic position in another (Lachnodacnum) clearly indicates that solving the systematics of the Neotropical clade of Coelostomatini (and in fact of the tribe as a whole) is a complex task. Additional studies combining morphological and molecular characters are necessary to understand the higher-level systematics of the group and to recognize ecology-based morphological characters from phylogeny-informative ones. Additional taxa also need to be included in the analyses – this especially concerns Hydroglobus puncticollis (Bruch, 1915) from Argentina, which probably belongs to the Neotropical clade. The clade of the core Phaenonotum (i.e. Phaenonotum after excluding P. caribense) corresponds well with the current morphology-based concept of the genus (Hansen, 1991), with African P. africanum, whose assignment to the genus is doubtful and needs to be tested, being the only exception. Based on our analyses, the core Phaenonotum are either sister group to Phaenostoma + Lachnodacnum (Fig. 1A) or to just the Phaenostoma posticatum clade (Fig. 1B). The latter topology seems to be better supported by morphology, as both Phaenostoma posticatum and the core Phaenonotum species share the unique shape of the metanepisternum (widened in posterior third, narrowing anteriorly, Fig. 7); all remaining coelostomatine species examined including Phaenostoma kontax, Lachnodacnum, Phaenonotum caribense and Cyclotypus have the metanepisternum with lateral margins parallel-sided (and hence the same width) throughout. However, the undescribed Phaenonotum species from Peru (MF845) bears the parallel-sided rather than anteriorly narrowing metanepisternum despite being strongly supported as a member of the core Phaenonotum clade in our analyses. Hence, the phylogenetic significance of this character needs to be further tested. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Supplement S1. Molecular data and phylogenetic and divergence dating analyses. Supplement S2. Imput data and results of the BioGeoBEARS analyses. Supplement S3. Complete data of all examined specimens. [Version of Record, published online 15 December 2017; http://zoobank.org/urn:lsid:zoobank.org:pub:9FBFC95E-DB5D-42C8-BD41-C3921F1016E3] ACKNOWLEDGEMENTS We are grateful to the authorities and personnel of the Cuban Ministry of Science, Technology and Environment (CICA-CITMA) and the Empresa Nacional para la Protección de la Flora y la Fauna (ENPFF) for providing access to protected areas under their control; to Carlos Suriel, Edda Martínez Caléz and Marilyn Colon Negron for their help with arranging the collecting permits in the Dominican Republic and Puerto Rico; to Robert Anderson (Canadian Museum of Nature, Ottawa), Matthew L. Gimmel (Santa Barbara Museum of Natural History) and people of the Cuba and Jamaica CarBio teams for their help during our fieldwork; to S. Peck (Carleton University) and curators of the institutional collections (see Material and methods) for the possibility to examine their collections; and to Andrew Short (University of Kansas, Lawrence, USA), Joseph Parker (Columbia University, New York) and Bruno Clarkson (Universidade Federal do Rio de Janeiro) for providing continental specimens used in our analyses. We specially acknowledge the comments and critique of Andrew Short and two anonymous reviewers on the earlier version of this draft, and Robert Angus (BMNH) for a careful correction of our English text. The study was supported by the grant SVV 260 434/2017 at the Department of Zoology, Charles University and the Rufford Foundation (15144-1) to A.D.H. and by the Ministry of Culture of the Czech Republic (DKRVO 2017/14, National Museum, 00023272) to M.F. REFERENCES Agnarsson I , LeQuier SM , Kuntner M , Cheng R-C , Coddington JA , Binford G . 2016 . Phylogeography of a good Caribbean disperser: Argiope argentata (Araneae, Araneidae) and a new ‘cryptic’ species from Cuba . ZooKeys 625 : 25 – 44 . Google Scholar CrossRef Search ADS Ali JR . 2012 . Colonizing the Caribbean: is the GAARlandia land-bridge hypothesis gaining a foothold ? Journal of Biogeography 39 : 431 – 433 . Google Scholar CrossRef Search ADS Alonso R , Crawford AJ , Bermingham E . 2012 . Molecular phylogeny of an endemic radiation of Cuban toads (Bufonidae: Peltophryne) based on mitochondrial and nuclear genes . Journal of Biogeography 39 : 434 – 451 . Google Scholar CrossRef Search ADS Archangelsky M . 1989 . A new species of the genus Phaenonotum Sharp, 1882 (Coleoptera, Hydrophilidae, Sphaeridiinae) from Venezuela . Aquatic Insects 4 : 217 – 220 . Arriaga-Varela E , Seidel M , Deler-Hernández A , Senderov V , Fikáček M . 2017 . A review of the Cercyon (Coleoptera: Hydrophilidae: Sphaeridiinae) of the Greater Antilles . ZooKeys 681 : 39 – 93 . Google Scholar CrossRef Search ADS Bartolini C , Lang H , Spell T . 2003 . Geo-chronology, geochemistry, and tectonic setting of the Mesozoic Nazas Arc in north-central Mexico, and its continuation to northern South America . In: Bartolini C , Buffler RT , Blickwede JF , eds. The Circum-Gulf of Mexico and the Caribbean: hydrocarbon habitats, basin formation, and plate tectonics . Tulsa: American Association of Petroleum Geologists Memoirs 79 : 427 – 461 . Bellemain E , Ricklefs RE . 2008 . Are islands the end of the colonization road ? Trends in Ecology and Evolution 23 : 461 – 468 . Google Scholar CrossRef Search ADS Bloom DD , Fikáček M , Short AE . 2014 . Clade age and diversification rate variation explain disparity in species richness among water scavenger beetle (Hydrophilidae) lineages . PLoS ONE 9 : e98430 . Google Scholar CrossRef Search ADS Bocák L , Barton C , Crampton-Platt A , Chesters D , Ahrens D , Vogler AP . 2014 . Building Coleoptera tree-of-life for >8000 species: composition of public DNA data and fit with Linnean classification . Systematic Entomology 39 : 97 – 110 . Google Scholar CrossRef Search ADS Bouckaert R , Heled J , Kühnert D , Vaughan T , Wu C-H , Xie D , Suchard MA , Rambaut A , Drummond AJ . 2014 . BEAST 2: a software platform for Bayesian evolutionary analysis . PLoS Computational Biology 10 : e1003537 . Google Scholar CrossRef Search ADS Ceccarelli FS , Zaldívar-Riverón A . 2013 . Broad polyphyly and historical biogeography of the Neotropical wasp genus Notiospathius (Braconidae: Doryctinae) . Molecular Phylogenetics and Evolution 69 : 142 – 152 . Google Scholar CrossRef Search ADS Claramount S , Cracraft J . 2015 . A new time tree reveals Earth history’s imprint on the evolution of modern birds . Science Advances 1 : e1501005 . Google Scholar CrossRef Search ADS Crews SC , Gillespie RG . 2010 . Molecular systematics of Selenops spiders (Araneae: Selenopidae) from North and Central America: implications for Caribbean biogeography . Biological Journal of the Linnean Society 101 : 288 – 322 . Google Scholar CrossRef Search ADS Deler-Hernández A , Cala-Riquelme F , Fikáček M . 2013a . Description of a new species of Phaenonotum from eastern Cuba (Coleoptera: Hydrophilidae: Sphaeridiinae) . Acta Entomologica Musei Nationalis Pragae 53 : 615 – 622 . Deler-Hernández A , Cala-Riquelme F , Fikáček M . 2014 . A review of the genus Oosternum Sharp of the West Indies (Coleoptera: Hydrophilidae: Sphaeridiinae) . Deutsche Entomologische Zeitschrift 61 : 43 – 63 . Google Scholar CrossRef Search ADS Deler-Hernández A , Fikáček M . 2016 . Redescriptions and lectotype designations of Central American species of Phaenonotum Sharp (Coleoptera, Hydrophilidae) based on the type material from the David Sharp collection . ZooKeys 579 : 83 – 98 . Google Scholar CrossRef Search ADS Deler-Hernández A , Fikáček M , Cala-Riquelme F . 2013b . A review of the genus Berosus Leach of Cuba (Coleoptera: Hydrophilidae) . ZooKeys 273 : 73 – 106 . Google Scholar CrossRef Search ADS Dziki A , Binford GJ , Coddington JA , Agnarsson I . 2015 . Spintharus flavidus in the Caribbean – a 30 million year biogeographical history and radiation of a ‘widespread species’ . PeerJ 3 : e1422 . Google Scholar CrossRef Search ADS Economo EP , Sarnat EM . 2012 . Revisiting the ants of Melanesia taxon cycle: historical and human-mediated invasions of a tropical archipelago . American Naturalist 180 : E1 – E16 . Google Scholar CrossRef Search ADS Esposito LA , Bloom T , Caicedo L , Alicia-Serrano A , Sanchez-Ruiz J , May-Collado LJ , Binford G , Agnarsson I . 2015 . Islands within islands: diversification of tailless whip spiders (Amblypygi, Phrynus) in Caribbean caves . Molecular Phylogenetics and Evolution 93 : 107 – 117 . Google Scholar CrossRef Search ADS Foley NM , Springer MS , Teeling EC . 2016 . Mammal madness: is the mammal tree of life not yet resolved ? Philosophical Transactions of the Royal Society B: Biological Sciences 371 : 20150140 . Google Scholar CrossRef Search ADS González-Rodríguez LM , García-Hernández AL , Clarkson B. 2017 . First records of water scavenger beetle species (Coleoptera, Hydrophilidae) from Quindío Department, Colombia . Check List 13 : 605 - 620 . Google Scholar CrossRef Search ADS Hansen M . 1991 . The hydrophiloid beetles. phylogeny, classification and a revision of the genera (Coleoptera, Hydrophiloidea) . Biologiske Skrifter 40 : 1 – 368 . Hansen M . 1999 . World catalogue of insects. Volume 2. Hydrophiloidea (s. str.) (Coleoptera) . Stenstrup : Apollo Books , 1– 416 pp. Hedges SB . 2006 . Paleogeography of the Antilles and the origin of West Indian terrestrial vertebrates . Annals of the Missouri Botanical Garden 93 : 231 – 244 . Google Scholar CrossRef Search ADS Hedges SB , Hass CA , Maxson LR . 1992 . Caribbean biogeography: molecular evidence for dispersal in West Indian terrestrial vertebrates . Proceedings of the National Academy of Sciences of the United States of America 89 : 1909 – 1913 . Google Scholar CrossRef Search ADS Huelsenbeck JP , Larget B , Alfaro ME . 2004 . Bayesian phylogenetic model selection using reversible jump Markov chain Monte Carlo . Molecular Biology and Evolution 21 : 1123 – 1133 . Google Scholar CrossRef Search ADS Iturralde-Vinent MA . 1982 . Aspectos geológicos de la biogeografía de Cuba . Ciencias de la Tierra y el Espacio 5 : 85 – 100 . Iturralde-Vinent MA . 2006 . Meso-Cenozoic Caribbean paleogeography: implications for the historical biogeography of the region . International Geology Review 48 : 791 – 827 . Google Scholar CrossRef Search ADS Iturralde-Vinent MA , MacPhee RDE . 1999 . Paleogeography of the Caribbean region: implications for Cenozoic biogeography . American Museum of Natural History Bulletin 238 : 1 – 95 . Landis MJ , Matzke NJ , Moore BR , Huelsenbeck JP . 2013 . Bayesian analysis of biogeography when the number of areas is large . Systematic Biology 62 : 789 – 804 . Google Scholar CrossRef Search ADS Lanfear R , Calcott B , Ho SYW , Guindon S . 2012 . PartitionFinder: combined selection of partitioning schemes and substitution models for phylogenetic analyses . Molecular Biology and Evolution 29 : 1695 – 1701 . Google Scholar CrossRef Search ADS Lawrence JF , Slipinski A . 2014 . Australian beetles volume 1: morphology, classification, and keys . Collingwood : CSIRO Publishing , 1– 576 pp. Lewis DS , Sperling FAH , Nakahara S , Cotton AM , Kawahara AY , Condamine FL . 2015 . Role of Caribbean Islands in the diversification and biogeography of Neotropical Heraclides swallowtails . Cladistics 31 : 291 – 314 . Google Scholar CrossRef Search ADS Matos-Maraví P , Núñez Águila R , Peña C , Miller JY , Sourakov A , Wahlberg N . 2014 . Causes of endemic radiation in the Caribbean: evidence from the historical biogeography and diversification of the butterfly genus Calisto (Nymphalidae: Satyrinae: Satyrini) . BMC Evolutionary Biology 14 : 199 . Google Scholar CrossRef Search ADS Matzke NJ . 2013 . Probabilistic historical biogeography: new models for founder-event speciation, imperfect detection, and fossils allow improved accuracy and model-testing . Frontiers of Biogeography 5 : 242 – 248 . Matzke NJ . 2014 . Model selection in historical biogeography reveals that founder-event speciation is a crucial process in Island Clades . Systematic Biology 63 : 951 – 970 . Google Scholar CrossRef Search ADS McHugh A , Yablonsky C , Binford G , Agnarsson I . 2014 . Molecular phylogenetics of Caribbean Micrathena (Araneae: Araneidae) suggests multiple colonisation events and single island endemism . Invertebrate Systematics 28 : 337 – 349 . Mittermeier RA , Robles-Gil P , Hoffman M , Pilgrim J , Brooks T , Mittermeier CG , Lamoreux J , da Fonseca GAB . 2005 . Hotspots revisited: earth’s biologically richest and most endangered terrestrial ecoregions . México : Agrupación Sierra Madre , 1– 392 pp. Moen DS , Wiens JJ . 2009 . Phylogenetic evidence for competitively driven divergence: body-size evolution in Caribbean treefrogs (Hylidae: Osteopilus) . Evolution 63 : 195 – 214 . Google Scholar CrossRef Search ADS Noonan BP , Pramuk JB , Bezy RL , Sinclair EA , de Queiroz K , Sites JW Jr . 2013 . Phylogenetic relationships within the lizard clade Xantusiidae: using trees and divergence times to address evolutionary questions at multiple levels . Molecular Phylogenetics and Evolution 69 : 109 – 122 . Google Scholar CrossRef Search ADS O’Dea A. Lessios HA , Coates AG , Eytan RI , Restrepo-Moreno SA , Cione AL , Collins LS , de Queiroz A , Farris DW , Norris RD , Stallard RF , Woodburne MO , Aguilera O , Aubry MPP , Berggren WA , Budd AF , Cozzuol MA , Coppard SE , Duque-Caro H , Finnegan S , Gasparini GMM , Grossman EL , Johnson KG , Keigwin LD , Knowlton N , Leigh EG , Leonard-Pingel JS , Marko PB , Pyenson ND , Rachello-Dolmen PG , Soibelzon E , Soibelzon L , Todd JA , Vermeij GJ , Jackson JB . 2016 . Formation of the Isthmus of Panama . Science Advances 2 : e1600883 . Google Scholar CrossRef Search ADS Oliva A , Fernández LA , Bachmann AO . 2002 . Sinopsis de los Hydrophiloidea acuáticos de la Argentina . Monografias del Museo Argentino de Ciencias Naturales 2 : 1 – 67 . Papadopoulou A , Anastasiou I , Vogler AP . 2010 . Revisiting the insect mitochondrial molecular clock: the mid-Aegean trench calibration . Molecular Biology and Evolution 27 : 1659 – 1672 . Google Scholar CrossRef Search ADS Poux C , Chevret P , Huchon D , de Jong WW , Douzery EJP . 2006 . Arrival and diversification of caviomorph rodents and platyrrhine primates in South America . Systematic Biology 52 : 228 – 244 . Google Scholar CrossRef Search ADS de Queiroz A . 2016 . Jurassic primates, immobile ducks and other oddities: a reply to Headsʼ review of the Monkeyʼs Voyage . Australian Systematic Botany 29 : 403 – 423 . Google Scholar CrossRef Search ADS Rambaut A . 2012 . FigTree v1.4 . Available at: http://tree.bio.ed.ac.uk/software/figtree Rambaut A , Suchard MA , Xie D , Drummond AJ . 2014 . Tracer v1.6 . Available at: http://beast.bio.ed.ac.uk/Tracer Ree RH , Smith SA . 2008 . Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis . Systematic Biology 57 : 4 – 14 . Google Scholar CrossRef Search ADS Říčan O , Piálek L , Zardoya R , Doadrio I , Zrzavý J . 2013 . Biogeography of the Mesoamerican Cichlidae (Teleostei: Heroini): colonization through the GAARlandia land bridge and early diversification . Journal of Biogeography 40 : 579 – 593 . Google Scholar CrossRef Search ADS Ricklefs RE , Bermingham E . 2002 . The concept of the taxon cycle in biogeography . Global Ecology & Biogeography 11 : 353 – 361 . Google Scholar CrossRef Search ADS Ricklefs RE , Bermingham E . 2008 . The West Indies as a laboratory of biogeography and evolution . Philosophical Transactions of the Royal Society B: Biological Sciences 363 : 2393 – 2413 . Google Scholar CrossRef Search ADS Ricklefs RE , Cox GW . 1972 . Taxon cycles in the West Indian avifauna . American Naturalist 106 : 195 – 219 . Google Scholar CrossRef Search ADS Rodriguez J , Pitts JP , von Dohlen CD . 2015 . Historical biogeography of the widespread spider wasp tribe Aporini (Hymenoptera: Pompilidae) . Journal of Biogeography 42 : 495 – 506 . Google Scholar CrossRef Search ADS Roelants K , Gower DJ , Wilkinson M , Loader SP , Biju SD , Guillaume K , Moriau L , Bossuyt F . 2007 . Global patterns of diversification in the history of modern amphibians . Proceedings of the National Academy of Sciences of the United States of America 104 : 887 – 892 . Google Scholar CrossRef Search ADS Ronquist F . 1997 . Dispersal-vicariance analysis: a new approach to the quantification of historical biogeography . Systematic Biology 46 : 195 – 203 . Google Scholar CrossRef Search ADS Ronquist F , Teslenko M , van der Mark P , Ayres DL , Darling A , Höhna S , Larget B , Liu L , Suchard MA , Huelsenbeck JP . 2012 . MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space . Systematic Biology 61 : 539 – 542 . Google Scholar CrossRef Search ADS Rosen DE . 1975 . A vicariance model of Caribbean biogeography . Systematic Biology 24 : 431 – 464 . Google Scholar CrossRef Search ADS Rosen DE . 1985 . Geological hierarchies and biogeographic congruence in the Caribbean . Annals of the Missouri Botanical Garden 72 : 636 – 659 . Google Scholar CrossRef Search ADS Sanmartín I , Ronquist F . 2004 . Southern hemisphere biogeography inferred by event-based models: plant versus animal patterns . Systematic Biology 53 : 216 – 243 . Google Scholar CrossRef Search ADS Sato JJ , Ohdachi SD , Echenique-Diaz L , Borroto-Páez L , Begué-Quiala G , Delgado-Labañino JL , Gámez-Díez J , Alvarez-Lemus J , Nguyen ST , Yamaguchi N , Kita M . 2016 . Molecular phylogenetic analysis of nuclear genes suggests a Cenozoic over-water dispersal origin for the Cuban Solenodon . Scientific Reports 6 : 31173 . Google Scholar CrossRef Search ADS Schulte P , Alegret L , Arenillas I , Arz JA , Barton PJ , Bown PR , Bralower TJ , Christeson GL , Claeys P , Cockell CS , Collins GS , Deutsch A , Goldin TJ , Goto K , Grajales-Nishimura JM , Grieve RAF , Gulick SPS , Johnson KR , Kiessling W , Koeberl C , Kring DA , MacLeod KG , Matsui T , Melosh J , Montanari A , Morgan JV , Neal CV , Nichols DJ , Norris RD , Pierazzo E , Ravizza G , Rebolledo-Vieyra M , Reimold WU , Robin E , Salge T , Speijer RP , Sweet AG , Urrutia-Fucugauchi J , Vajda V , Whalen MT , Willumsen PS . 2010 . The Chicxulub Asteroid Impact and Mass Extinction at the Cretaceous-Paleogene Boundary . Science 327 : 1215 – 1218 . Google Scholar CrossRef Search ADS Seidel M , Arriaga-Varela E , Fikáček M . 2016 . Establishment of Cylominae Zaitzev, 1908 as a valid name for the subfamily Rygmodinae Orchymont, 1916 with an updated list of genera (Coleoptera: Hydrophilidae) . Acta Entomologica Musei Nationalis Pragae 56 : 159 – 165 . Short AEZ . 2004 . Review of the Enochrus Thomson of the West Indies (Coleoptera: Hydrophilidae) . Koleopterologische Rundschau 74 : 351 – 361 . Short AEZ , Fikáček M . 2013 . Molecular phylogeny, evolution, and classification of the Hydrophilidae (Coleoptera) . Systematic Entomology 38 : 723 – 752 . Google Scholar CrossRef Search ADS Smetana A . 1978 . Revision of the subfamily Sphaeridiinae of America North of Mexico (Coleoptera: Hydrophilidae) . Memoirs of the Entomological Society of Canada 105 : 1 – 292 . Google Scholar CrossRef Search ADS Spangler PJ , Short AEZ . 2008 . Three new species of Neotropical Tropisternus Solier (Coleoptera: Hydrophilidae) . Zootaxa 1917 : 65 – 68 . Stadler T . 2009 . On incomplete sampling under birth–death models and connections to the sampling-based coalescent . Journal of Theoretical Biology 261 : 58 – 66 . Google Scholar CrossRef Search ADS Stamatakis A . 2014 . RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies . Bioinformatics 30 : 1312 – 1313 . Google Scholar CrossRef Search ADS Tada R , Iturralde-Vinent MA , Matsui T , Tajika E , Oji T , Goto K , Nakano Y , Takayama H , Yamamoto S , Kiyokawa S , Toyoda K , García Delgado D , Díaz-Otero C , Rojas R . 2003 . K/T boundary deposits in the paleo–western Caribbean basin . In: Bartolini C , Buffler R , Blickwede J , eds. The Circum–Gulf of Mexico and the Caribbean: hydrocarbon habitats, basin formation, and plate tectonics . Tulsa: American Association of Petroleum Geologists Memoir 79 , 582 – 604 . Thorpe RS , Malhotra A , Stenson A , Reardon JT . 2004 . Adaptation and speciation in Lesser Antillean anoles . In: Dieckmann U , Doebeli M , Metz JAJ , Tautz D , eds. Adaptive speciation . Cambridge : Cambridge University Press , 324 – 335 . Uit de Weerd DR , Robinson DG , Rosenberg G . 2016 . Evolutionary and biogeographical history of the land snail family Urocoptidae (Gastropoda: Pulmonata) across the Caribbean region . Journal of Biogeography 43 : 763 – 777 . Google Scholar CrossRef Search ADS Wahlberg N . 2006 . That awkward age for butterflies: insights from the age of the butterfly subfamily Nymphalinae (Lepidoptera: Nymphalidae) . Systematic Biology 55 : 703 – 714 . Google Scholar CrossRef Search ADS Wahlberg N , Freitas AV . 2007 . Colonization of and radiation in South America by butterflies in the subtribe Phyciodina (Lepidoptera: Nymphalidae) . Molecular Phylogenetics and Evolution 44 : 1257 – 1272 . Google Scholar CrossRef Search ADS Weaver PF , Cruz A , Johnson S , Dupin J , Weaver KF . 2016 . Colonizing the Caribbean: biogeography and evolution of livebearing fishes of the genus Limia (Poeciliidae) . Journal of Biogeography 43 : 1808 – 1819 . Google Scholar CrossRef Search ADS Wilson EO . 1959 . Adaptive shift and dispersal in a tropical ant fauna . Evolution 13 : 122 – 144 . Google Scholar CrossRef Search ADS Wilson EO . 1961 . The nature of the taxon cycle in the Melanesian ant fauna . American Naturalist 95 : 169 – 193 . Google Scholar CrossRef Search ADS Xie W , Lewis PO , Fan Y , Kuo L , Chen MH . 2011 . Improving marginal likelihood estimation for Bayesian phylogenetic model selection . Systematic Biology 60 : 150 – 160 . Google Scholar CrossRef Search ADS Zhang G , Basharat U , Matzke N , Franz NM . 2017 . Model selection in statistical historical biogeography of Neotropical insects – the Exophthalmus genus complex (Curculionidae: Entiminae) . Molecular Phylogenetics and Evolution 109 : 226 – 239 . Google Scholar CrossRef Search ADS Zhang JX , Maddison WP . 2013 . Molecular phylogeny, divergence times and biogeography of spiders of the subfamily Euophryinae (Araneae: Salticidae) . Molecular Phylogenetics and Evolution 68 : 81 – 92 . Google Scholar CrossRef Search ADS © 2017 The Linnean Society of London, Zoological Journal of the Linnean Society This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Zoological Journal of the Linnean SocietyOxford University Press

Published: Dec 15, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off