75 Phylogenetic placement of the Pacific Northwest subterranean endemic diving beetle Stygoporus oregonensis Larson & LaBonte (Dytiscidae, Hydroporinae) Kojun Kanda 1 , R. Antonio Gomez 1 , Richard Van Driesche 1 , Kelly B. Miller 2 , David R. Maddison 1 1 Department of Integrative Biology, Oregon State University, Corvallis, Oregon, United States of America 2 Department of Biology, University of New Mexico, Albuquerque, New Mexico, United States of America Corresponding author: Kojun Kanda ( firstname.lastname@example.org ) Academic editor: M. Michat |Received 14 July 2016|Accepted 12 October 2016|Published 16 November 2016 http://zoobank.org/2BF08A13-A6AD-448F-A9CC-B2D706745606 Citation: Kanda K, Gomez AR, Van Driesche R, Miller KB, Maddison DR (2016) Phylogenetic placement of the Pacic Northwest subterranean endemic diving beetle Stygoporus oregonensis Larson & LaBonte (Dytiscidae, Hydroporinae). ZooKeys 632: 75â€“91. doi: 10.3897/zookeys.632.9866 Abstract Stygoporus oregonensis Larson & LaBonte is a little-known subterranean diving beetle, which, until recently, had not been collected since the type series was taken from a shallow well in western Oregon, USA, in 1984. Here we report the discovery of additional specimens collected from a nearby well in the Willamette Valley. Sequence data from four mitochondrial genes, wingless , and histone III place Stygoporus Larson & LaBonte in the predominantly Mediterranean subtribe Siettitiina of the Hydroporini. Morphological support for these results is discussed, and details of the collecting circumstances of the new specimens are presented. We argue that the biogeographic patterns of Nearctic Siettitiina highlight the likelihood of additional undiscovered subterranean dytiscids in North America. Keywords Stygobiont, aquatic Coleoptera, Hydroporini, aquifer, Siettitiina, Nearctic, Oregon ZooKeys 632: 75 (2016) doi: 10.3897/zookeys.632.9866 http://zookeys.pensoft.net Copyright Kojun Kanda et al.. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. RESEARCH ARTICLE Launched to accelerate biodiversity research A peer-reviewed open-access journal
76 Introduction In the spring of 1984 an unusual, pale, blind diving beetle was found in a bathtub in a private residence near the town of Dallas, Oregon, USA. e bathtub received water directly from a shallow well that was drawing from the Willamette Lowland aquifer system in the central Willamette Valley. e residents sent the specimen to an ento mology extension specialist, Dr. J. Capizzi at Oregon State University, who recognized the beetle as distinct and suggested to the residents that they collect more specimens (Larson and LaBonte 1994). An additional eight specimens were found, and shortly thereafter, the residents treated the well with chlorine. No additional specimens were collected at the type locality following the wellâ€™s chlorine treatment (Larson and La Bonte 1994). e species was described and given the name Stygoporus oregonensis Larson & LaBonte in honor of its subterranean predilections and the state from which it was thus far known (Larson and LaBonte 1994). In the more than 30 years since the type series was collected, no additional specimens of S. oregonensis have been reported prior to the present study. Stygoporus oregonensis is a small-bodied diving beetle with pale, mostly yellow cu ticle, long elytral marginal setae, fused elytra, minute ight wings, and without eyes (Larson and LaBonte 1994; Fig. 1). ese morphological features are commonly ob served in various, often widely unrelated subterranean lineages and are considered to typify stygobitic Dytiscidae from around the world (Leys and Watts 2008; Leys et al. 2003; Miller et al. 2013; Spangler and Decu 1998; Watts and Humphreys 2009). An additional morphological feature common among stygobitic dytiscids is a discontinu ous body outline, contrasted with the more streamlined habitus of many diving beetles. Inferring the phylogenetic placement of stygobitic species is crucial for shedding light on their origins and developing a framework for studying adaptation and other responses to subterranean environments. Addressing the mechanisms responsible for the unusual though oft-repeated appearance of the stygobitic fauna and their often unexpected distributions is an active eld (Juan et al. 2010). Cave faunas are some of the most visually striking examples of convergence, and several recent studies of stygo bitic and troglobitic life have used the character-rich information present in molecular sequence data to help place these morphologically similar species into phylogenetic hypotheses (Faille et al. 2010; Gmez et al. 2016; Leys et al. 2003; Miller et al. 2013; Ribera et al. 2010; Toussaint et al. 2015; Wiens et al. 2003). In the United States, three aquatic beetle families are known to include stygobitic species: Dryopidae ( Stygoparnus comalensis Barr & Spangler, 1992), Elmidae ( Typhloe lmis Barr, 2015: 3 species), and Dytiscidae ( Ereboporus naturaconservatus Miller, Gibson & Alarie, 2009, Haideoporus texanus Young & Longley, 1976, Psychopomporus felipi Jean, Telles & Miller, 2012, Comaldessus stygius Spangler & Barr, 1995, and Stygoporus oregonensis Larson & LaBonte, 1994). Apart from S. oregonensis , all US stygobitic bee tles are only known to occur in the Edwards-Trinity aquifer system in central Texas. Whereas the relationships of Stygoparnus Barr and Spangler and Typhloelmis to other members of their respective families have yet to be explored with phylogenetic
77 Figure 1. Dorsal habitus of female Stygoporus oregonensis . Scale bar = 1 mm. methods, the placement of three of the four described Texas stygobitic dytiscids within the very diverse subfamily Hydroporinae was recently inferred using molecular se quences (Miller et al. 2013). Miller et al. (2013) did not include C. stygius in their analyses because it possesses several morphological synapomorphies that unambigu ously place it within Bidessini. e other Texas stygobites were placed in two clades, the Graptodytes group ( E. naturaconservatus and P. felipi ) and the Hydroporus group ( H. texanus ). Both of these generic groups are traditionally classied within the large, heterogeneous tribe Hydroporini sensu lato , which has been shown to be polyphyletic by several authors (Miller et al. 2006; Ribera et al. 2002; Ribera et al. 2008). Recently, Miller and Bergsten (2014) formalized the subgroups of Hydroporini s. l. establishing
78 the subtribe Siettitiina for Graptodytes group and Hydroporina for Hydroporus group; they also provisionally placed S. oregonensis in Hydroporina. In their paper describing S. oregonensis , Larson and LaBonte (1994) hypothesized that Stygoporus is related to the Nearctic genus Sanlippodytes Franciscolo (also placed in Hydroporina by Miller and Bergsten (2014)) based on similarly large metatrochant ers, apically produced metaventral processes, and Sanlippodytes exhibiting character states that â€œform a good base from which a truly subterranean beetle could evolveâ€ (Larson and LaBonte 1994). In addition, several Sanlippodytes species are known from a variety of habitats including acidic pools (Post 2010), interstitial spaces along margins of springs and creeks, within sand-clay or gravel substrate of cold springs, limnocrene pools, under beach debris or cover along the margins of alpine lakes, un der mosses in springs and seeps, and caves (Larson et al. 2000), which may be steps along the way to colonization of subterranean aquifers by the ancestor of S. oregonensis . However the relationship between these genera has yet to be tested. In this paper, we report additional specimens of S. oregonensis from a separate well, also in the central Willamette Valley, Oregon. ese specimens yielded DNA, from which we amplied six genes used in Miller et al.â€™s (2013) phylogeny of Hydropori nae. We incorporate our new sequences with data from Miller et al. (2013) to infer the phylogenetic placement of S. oregonensis and discuss morphological aspects of S. oregonensis in light of these results. Methods Discovery of Stygoporus oregonensis specimens Two mostly intact specimens of Stygoporus oregonensis and fragments of additional indi viduals were recovered from accumulated sand and detritus in the lter of a residential well system (USA: Oregon: Marion County, Talbot, south of Talbot Road South). e well sits near an old oxbow of the Willamette River and the wellhead is located roughly 14 m below the surface. is site is roughly 27km SSE of the type locality (Fig. 2). Be tween 2014 and 2016, the accumulated material in the well lter was checked six times (Suppl. material 3). e rst two surveys of the ltrate contained minute and pale beetle fragments assumed to be remnants of S. oregonensis . ese fragments did not appear to contain any soft tissue; they may have died long before the ltrate was examined. e mostly intact beetle specimens were both caught during the rainy winter months and contained soft tissue, which appeared to be suitable for DNA extraction and PCR sequencing. e specimens were found with the prothorax and head slightly separated from the rest of the body and the genitalia extruded as if they had expanded slightly. is damage may have occurred during depressurization: the removal of the lter causes a change in pressure from 8-10 psi to atmospheric pressure in approximately 2 seconds. In addition to S. oregonensis , we recovered crustaceans (ostracods, copepods, and Bathynellacea), numerous oribatid mites, and a few other insects (roscidae (Coleoptera),
79 Figure 2. e two known collection localities of Stygoporus oregonensis . Oregon/Washington State boundary in black. County boundaries in brown. Blue shaded region outlined with a dotted line cor responds to Willamette Lowland basin-ll aquifers. Type locality indicated by red star with black border. New collection locality indicated by black star. Chironomidae larvae (Diptera), and unattributed elytral fragments). While the roscidae appears to be an obvious terrestrial contaminant, we could not determine if the other taxa are associated with the aquifer or not. e pair of pale elytra recovered in one of the samples (OSAC Lot 20160620-03) was markedly smaller and stouter than that of S. oregonensis and while it may have come from a surface dwelling species, it raises the possibility of additional undiscovered species inhabiting the aquifer. DNA extraction and sequencing We extracted DNA from the two fairly intact specimens of S. oregonensis using DNeasy Blood and Tissue kits (Qiagen) following the manufacturer's protocols. Specimens
80 were disarticulated between the abdomen and thorax prior to extraction; we did not grind any tissue, and thus the exoskeleton was preserved. We successfully ampli ed and sequenced six of the seven gene fragments used in Miller et al. (2013): 12S rRNA (12S), 16S rRNA (16S), cytochrome c oxidase I (COI), cytochrome c oxidase II (COII), wingless ( wg ), and histone III (H3), but were unsuccessful at amplifying elongation factor 1-alpha. PCRs were performed in 25 microliter reactions on either an Eppendorf Mastercycler gradient or Mastercycler ProS using TaKaRa Ex Taq fol lowing manufacturerâ€™s protocols. We used primer pairs and amplication conditions described in Miller et al. (2013) for 12S, 16S, COI (Pat/Jerry), COII, and H3, and Kanda et al. (2015) for wg and the barcoding region of COI (Suppl. material 4). PCR cleanup, quantication, and sequencing were performed at the University of Arizonaâ€™s Genomic and Technology Core Facility (UAGC) using a 3730 XL Applied Biosystems automatic sequencer. Sequence processing and phylogenetic analyses Initial assembly of chromatograms was performed using Phred v. 0.020425.c (Green and Ewing 2002) and Phrap v. 0.990319 (Green 1999) as orchestrated by Mesquite v. 3.04 package Chromaseq v. 1.12 (Maddison and Maddison 2011, Maddison and Maddison 2015) with subsequent manual processing. S. oregonensis sequences were combined with single gene matrices from Miller et al. (2013). e taxon sampling used in Miller et al.â€™s (2013) study encompasses the morphological diversity of Hydropori nae, including numerous representatives of all currently recognized subtribes of Hy droporini (Suppl. material 5) and thus provides an excellent framework for inferring the phylogenetic placement of S. oregonensis . 12S and 16S matrices were aligned using MAFFT v. 7.130b (Katoh and Standley 2013) and the L-INS-i method. Alignment of protein-coding genes were performed manually since they either had no indels (COI, COII, and H3) or just a single inferred amino acid indel ( wg ). All nucleotide alignments were also combined into a single concatenated dataset. Optimal data partition schemes and model of molecular evolution for proteincoding genes were inferred using PartitionFinder v. 1.1.1 (Lanfear et al. 2012) starting from an initial partition scheme based on codon position. Examined models were re stricted to those available in RAxML, BIC was used to compare models, and the greedy algorithm was used for searches. Models for 12S and 16S were inferred using BIC implemented in jModelTest 2.0 (Darriba et al. 2012). PartitionFinder analysis was also conducted on the concatenated dataset starting with an initial partition scheme based on gene and codon. Optimal models and partitions for all datasets are presented in Table 1. We conducted Maximum Likelihood (ML) analyses on single gene and concat enated datasets using RAxML v. 8.0.3 (Stamatakis 2014) implemented through the
81 Table 1. Properties of phylogenetic datasets analyzed for this study. NTaxa : e number of taxa repre sented in the dataset. Partitions : Optimal partitioning scheme chosen by PartitionFinder . NChar (BP) : Number of characters (bases) in the aligned dataset/partition. Model : Optimal model of molecular evolu tion inferred by either jModelTest (12S, and 16S) or PartitionFinder (protein-coding genes). Dataset NTaxa Partitions NChar (BP) Model 12S 49 NA 362 GTR+I+G 16S 50 NA 533 HKY+I+G COI 44 (1) n1, n2 838 GTR+I+G (2) n3 418 GTR+G COII 43 (1) n1, n2 450 GTR+I+G (2) n3 224 GTR+G H3 50 (1) n1, n2, n3 328 GTR+I+G wg 20 (1) n1, n2 306 GTR+I+G (2) n3 154 GTR+G Concatenated 51 (1) 12S, 16S 895 GTR+I+G (2) n1 and n2 of all genes 1812 GTR+I+G (3) n3 of COI and COII 642 GTR+G Mesquite package Zephyr v. 1.1 (Maddison and Maddison 2015) with optimal parti tion schemes and models of molecular evolution. When dierent models were chosen for dierent partitions, we applied the most complex model to the entire dataset. Since the HKY substitution model that was selected for 16S is not available in RAxML, we instead used GTR. We conducted 500 independent searches for the maximum likeli hood tree and 1,000 bootstrap replicates on all datasets. Morphological methods Methods for gross morphological examination and use of terms follow Miller (2005, 2016). e two extracted specimens were also used for morphological study of female internal reproductive characters. Female genitalia were dissected following DNA ex traction, stained with 10% Chlorazol Black diluted in 75% ethanol, and examined on a slide in deionized water. During the course of study, the female genitalia of the recently acquired specimens were heavily damaged or lost accidentally after morpho logical features were recorded. Because of the extensive damage or loss, we chose not to image the genitalia. e female genital structures were mounted in Euparal on card stock and pinned beneath the specimen. e dorsal habitus image was taken with a Leica Z6 and JVC KY-F75U camera using Microvisionâ€™s Cartographer to take a stack of pictures at dierent focal planes. Stacking was performed using the PMax procedure implemented in Zerene Stacker (Zerene Systems). Removal of background and minor color adjustment was performed using Photoshop and Illustrator CS5 (Adobe).
82 Data availability All specimens examined in this study and the two DNA extractions are deposited in the Oregon State Arthropod Collection (OSAC), Oregon State University. Associated OSAC lot and voucher codes are given in Suppl. material 3. Final sequences for both specimens are available through GenBank (accession numbers KX882130 KX882141 ). Matrices used in the analyses are available as supplemental content (Suppl. material 6: MatricesForAnalyses.nex). Results Additional morphological characters for Stygoporus oregonensis Morphological characters discussed below are based on the original description of S. oregonensis (Larson and LaBonte 1994) and material examined for the present study; the latter allowed us to examine previously unstudied characters of the proventriculus and female genitalia. e proventriculus of S. oregonensis has a simple transverse tooth similar to that of Hydroporus Clairville with elds of papillae laterally. e female genitalia are of hydroporine-type (Miller 2001) with elongate ductwork. e external genitalia lack laterotergites, gonocoxosternites are broadly triangular and nely setose ventrally with an anteriorly rounded projection, gonocoxae are unfused, slender basally, broadening apically to a narrowly rounded apex, with numerous minute apical setae. Internally, the bursa is small and lacks a ring-like sclerite, the spermathecal duct is elongate and slender for most of its length, broadening before attaching to the small bulbous sper matheca, and the shorter fertilization duct is similarly slender and inserts ventrally on the vagina posterior to the common oviduct. Phylogenetic placement of Stygoporus oregonensis e maximum likelihood (ML) tree of the concatenated dataset is shown in Figure 3 and majority rules consensus tree from 1000 bootstrap replicates is shown in Figure 4. ML trees and bootstrap consensus trees for single-gene datasets are provided in Suppl. material 1 and 2. ML bootstrap support percentages (BSP) are summarized across phy logenetic reconstructions in Table 2 for hypotheses regarding the taxonomic placement of Stygoporus oregonensis . Maximum likelihood analysis of the concatenated dataset recovers S. oregonensis as sister to the Texas stygobite E. naturaconservatus (Figs 3, 4) with high bootstrap support (BSP=99.3). is clade is placed within the hydroporine subtribe Siettitiina, which is recovered with moderate support (BSP=75). Additional recovered genus or tribal-level groups largely correspond to the ML inference of phylogeny by Miller et al. (2013).
83 99.3 75.5P sychopomporus felipi 0.2 Nebrioporus rotundatus Nebrioporus clarkii Laccornis difformis Laccornellus lugubris Methles cribratellus Celina hubbelli Celina imitatrix Microdytes svensoni Hydrovatus pustulatus Queda youngi Desmopachria convexa Vatellus bifenestratus Amarodytes sp. Peschetius quadricostatus Liodessus affinis Uvarus baoulicus Hygrotus acaroides Coelambus semivittatus Herophydrus inquinatus Hyphydrus elegans Hyphydrus excoffieri Sanfilippodytes sp. Heterosternuta pulchra Haideoporus texanus Neoporus mellitus Hydrocolus paugus Hydroporus dorsalis Hydroporus angustatus Hydroporus palustris Stictonectes optatus Stictonectes rufulus Rhithrodytes sexguttatus Graptodytes ignotus Ereboporus naturaconservatus Stygoporus oregonensis Stygoporus oregonensis Canthyporus parvu s Oreodytes congruus Oreodytes scitulus Oreodytes quadrimaculatus Stictotarsus roffii Necterosoma susanna Necterosoma undecimlineatum Sternopriscus tasmanicus Chostonectes gigas Chostonectes nebulosus Megaporus howittii Megaporus hamatus Antiporus femoralis Antiporus blakei Siettitiina Figure 3. Maximum likelihood tree from concatenated dataset. Scale bar = 0.2 expected substitutions per position as estimated by RAxML. Stygoporus oregonensis in orange; other stygobitic dytiscids in blue; the epigean genus Sanlippodytes , hypothesized by Larson and LaBonte (1994) to be the closest relative to S. oregonensis , in green. Bootstrap support given at nodes for Siettitiina and S. oregonensis + Ereboporus naturaconservatus . Ereboporus naturaconservatus and S. oregonensis are recovered as sister species in all single gene ML analyses (Suppl. material 1). is relationship is moderately to highly supported across single gene bootstrap analyses except in 16S (Table 2, Suppl. material 2). Although Siettitiina ( Graptodytes group) is not equally well sampled for all genes, S. oregonensis and E. naturaconservatus are recovered within a monophyletic Siettitiina in ML analyses of 16S, COI, and wg . Support for Siettitiina (including Stygoporus ) is high in bootstrap analyses of 16S but low to non-existent in other genes. Table 2. Bootstrap support for placement of Stygoporus oregonensis . Taxonomic hypotheses are in the rst column. Bootstrap support given as a percentage for each hypothesis for all analyzed matrices. â€œConâ€ refers to the analysis of the concatenated matrix. Taxonomic hypotheses Con 12S 16S COI COII H3 wg Stygoporus oregonensis + Ereboporus naturaconservatus 99.3 79.2 41.1 60.1 70.7 86.0 90.0 S. oregonensis + Sanlippodytes 0 0 0 0 0.2 3.0 0.5 Siettitiina including S. oregonensis 75.5 4.6 87.7 45.7 0 0 31.0 Siettitiina excluding S. oregonensis 0 0 1.3 1.2 0 0 0 S. oregonensis in Hydroporina 0 0.1 0 0 0 0 0
84 Figure 4. Majority rule consensus of 1,000 bootstrap replicates performed on concatenated dataset. Bootstrap percentages given for clades recovered with more than 50% support. Branches and taxa colored as in Figure 3. 100 100 51 99.7 99.9 88.1 99.5 84.2 100 95.6 100 56.8 66.3 62.4 86.2 54.8 100 100 99.3 75.5 53.7 91 93.6 98.3 96.2 100 100 100 59 71.4 100Megaporus hamatus Laccornis difformis Canthyporus parvus Laccornellus lugubris Methles cribratellus Microdytes svensoni Celina hubbelli Celina imitatrix Desmopachria convexa Vatellus bifenestratus Hyphydrus elegans Hyphydrus excoffieri Hydrovatus pustulatus Queda youngi Hygrotus acaroides Coelambus semivittatus Herophydrus inquinatus Amarodytes sp. Peschetius quadricostatus Liodessus affinis Uvarus baoulicus Nebrioporus rotundatus Nebrioporus clarkii Oreodytes congruus Oreodytes scitulus Oreodytes quadrimaculatus Stictotarsus roffii Graptodytes ignotus Psychopomporus felipi Rhithrodytes sexguttatus Stictonectes optatus Stictonectes rufulus Ereboporus naturaconservatus Stygoporus oregonensis Stygoporus oregonensis Sanfilippodytes sp. Neoporus mellitus Heterosternuta pulchra Haideoporus texanus Hydrocolus paugus Hydroporus dorsalis Hydroporus angustatus Hydroporus palustri s Necterosoma susanna Necterosoma undecimlineatum Chostonectes gigas Chostonectes nebulosus Sternopriscus tasmanicus Antiporus blakei Antiporus femoralis Megaporus howittii Siettitiina Stygoporusoregonensis is never placed with Sanlippodytes nor in Hydroporina in ML analyses of either the concatenated or single gene ML trees, and this hypothesis has no bootstrap support across analyses.
85 Discussion In their original description of Stygoporus , Larson and LaBonte (1994) placed it in the Hydroporini based on (1) posterior margin of metacoxal lobes continuous and sinuate, (2) posterior margin of metacoxal lobes unfused to abdominal ventrites II and III, (3) metafemur broadly separated from metacoxal lobe by large metatrochanter, (4) base of metafemur hidden ventrally by metacoxal lobe, and (5) male lateral lobes with a single segment. None of these morphological characters are synapomorphic for a tribal-level clade of Hydroporinae. Historically, Hydroporini included those Hydroporinae with out a distinctive set of apomorphies, and clarifying relationships within Hydroporini has been a prominent goal of modern Dytiscidae systematics (Miller and Bergsten 2014). Recently, Miller and Bergsten (2014) reclassied the Hydroporini, giving genus group clades that were well supported with molecular and morphological data available higher-level names: Deronectina ( Deronectes group), Hydroporina ( Hydroporus group), Sternopriscina ( Necterosoma group), and Siettitiina ( Graptodytes group). While they did not have molecular sequence data for Stygoporus , they tentatively classied it within the Hydroporina (Miller and Bergsten 2014). Larson and LaBonte (1994) hypothesized that Stygoporus is sister to Sanlippodytes based on similar anteriorly produced metaventral processes, large metatrochanters, and habitat data. Contrary to this hypothesis, our molecular data places S. oregonensis with in Siettitiina and not near Hydroporina and Sanlippodytes . ough the phylogenetic analyses of Miller et al. (2013) and Miller and Bergsten (2014) strongly support the monophyly of Siettitiina, this clade is morphologically poorly dened. One potential synapomorphy is a ring-sclerite on the bursa copulatrix adjacent to the attachment of the spermathecal duct (Miller and Bergsten 2014). is structure is known to occur in Ereboporus and other siettitiines but is notably missing from Graptodytes Seidlitz (Miller et al. 2013), which is also the most diverse genus within the subtribe (Nilsson 2001). As in Graptodytes , the bursa copulatrix of S. oregonensis lacks a ring-like sclerite. We note that although the female genitalia in our specimen was damaged, it is clear that there is not a region along the bursa that looks more sclerotized or distinct from the remaining structure. ere are additional morphological characters in support of inclusion of S. oregon ensis within Siettitiina, though it remains unclear whether these characters are strong synapomorphies for Siettitiina as a whole. In particular, the pronotum of S. oregonensis has prominent paralateral longitudinal creases or striae similar to many members of the larger group (e.g. Graptodytes Seidlitz, Siettitia Abeille de Perrin and Etruscodytes Mazza, Cianferoni, and Rocchi). e prosternal process of S. oregonensis contacts the anteriorly projecting and narrowly rounded metaventral process, resting dorsad to it and altogether looks remarkably similar to the Italian stygobite Etruscodytes . is re gion of the body has received much attention from biologists interested in stygobitic beetles (Miller et al. 2013; Spangler 1986), and these sclerites are intricately involved in locomotion, particularly wedging (Evans 1977). e similarity in form of these sclerites may be evidence of recent common ancestry, but this may also be the result of
86 convergence as modications to the ventral thoracic sclerites and the loss of a stream lined body are commonly observed patterns in distantly related subterranean diving beetles (Miller et al. 2009; Spangler 1986). Other morphological features in S. oregonensis relevant to grouping within Hy droporini are known plesiomorphies. ese are, for example, the simple transverse tooth of the proventriculus, the unfused, simple gonocoxae, the basally broad and api cally narrowed elytral epipleuron, the male proand mesotarsomeres I-III with ventral adhesive setae, and the mesoventral fork separated from the anteromedial metaventral process. Most of these characters are unlike those observed in Deronectina and Ster nopriscina, and the morphological evidence separating Hydroporina from Siettitiina is limited. Based on our observations, it appears that Stygoporus retains many plesiomor phies and placement based on morphological characters alone is dicult. However, the sequence data support the inclusion of Stygoporus within Siettitiina, and they deci sively indicate that Stygoporus is closely related to Ereboporus among sampled species. e Siettitiina has a predominantly Mediterranean and European distribution and includes many epigean species as well as other subterranean species (e.g. Ribera and Faille 2010). Intriguingly, the only presently known European stygobitic dytiscids are members of Siettitiina, including some species known only from wells and aquifers (Castro and Delgado 2001; Mazza et al. 2013; Ribera and Faille 2010). Aside from S. oregonensis and two described Texas subterranean aquifer endemics, Siettitiina are not represented in the New World, which suggests an ancient origin for these species (Miller et al. 2013). e mechanism and process behind this biogeographic pattern is not known. Conclusions invoking vicariance, dispersal, and extinction can certainly be applied to this pattern, but we prefer the practical hypothesis that at least part of this result is attributable to our ignorance. Instead of being dismayed, however, we are excited by the possibility that there are many unknown stygobitic beetles in aquifers between Oregon and Texas as well as other parts of the world for which little sampling of this habitat has been done. Acknowledgements We sincerely thank the private landowners who allowed us access to their land. Por tions of this project were supported by the Harold E. and Leona M. Rice Endowment Fund at Oregon State University (to D.R. Maddison) and NSF grants #DEB-0845984 and #DEB (to K.B. Miller). References Barr C, Spangler P (1992) A new genus and species of stygobiontic dryopid beetle, Stygoparnus comalensis (Coleoptera: Dryopidae), from Comal Springs, Texas. Proceedings of the Bio logical Society of Washington 105: 40.
87 Barr CB, Gibson JR, Diaz PH (2015) Typhloelmis Barr (Coleoptera: Elmidae: Elminae), a New Stygobiontic Rie Beetle Genus with ree New Species from Texas, USA. e Coleop terists Bulletin 69: 531. doi: 10.1649/0010-065X-69.4.531 Castro A, Delgado JA (2001) Iberoporus cermenius , a new genus and species of subterranean wa ter beetle (Coleoptera: Dytiscidae) from Spain. Aquatic Insects 23: 33. doi: 10.1076/ aqin.184.108.40.20631 Darriba D, Taboada GL, Doallo R, Posada D (2012) jModelTest 2: more models, new heu ristics and parallel computing. Nature Methods 9: 772. doi: 10.1038/nmeth.2109 Evans M (1977) Locomotion in the Coleoptera Adephaga, especially Carabidae. Journal of Zoology 181: 189. doi: 10.1111/j.1469-7998.1977.tb03237.x Faille A, Bourdeau C, Fresneda J (2010) A new species of blind Trechinae from the Pyrenees of Huesca, and its position within Aphaenops ( sensu stricto )(Coleoptera: Carabidae: Trechini). Zootaxa 2566: 49. Gmez RA, Reddell J, Will K, Moore W (2016) Up high and down low: Molecular systematics and insight into the diversication of the ground beetle genus Rhadine LeConte. Molecular Phylogenetics and Evolution 98: 161. doi: 10.1016/j.ympev.2016.01.018 Green P (1999) Phrap. Version 0.990329. http://phrap.org Green P, Ewing B (2002) Phred. Version 0.020425 c. Computer program and documentation available at www.phrap.org Jean A, Telles ND, Gibson JR, Foley D, Miller KB (2012) Description of a new genus and species of stygobiontic diving beetle, Psychopomporus felipi Jean, Telles, and Miller (Co leoptera: Dytiscidae: Hydroporinae), from the Edwards-Trinity aquifer system of Texas, USA. e Coleopteristsâ€™ Bulletin 66: 105. doi: 10.1649/072.066.0202 Juan C, Guzik MT, Jaume D, Cooper SJ (2010) Evolution in caves: Darwinâ€™s â€˜wrecks of ancient lifeâ€™ in the molecular era. Molecular Ecology 19: 3865. doi: 10.1111/j.1365-294X.2010.04759.x Kanda K, Pug JM, Sproul JS, Dasenko MA, Maddison DR (2015) Successful recovery of nuclear protein-coding genes from small insects in museums using illumina sequencing. PLoS ONE 10: e0143929. doi: 10.1371/journal.pone.0143929 Katoh K, Standley DM (2013) MAFFT multiple sequence alignment software version 7: im provements in performance and usability. Molecular Biology and Evolution 30: 772. doi: 10.1093/molbev/mst010 Lanfear R, Calcott B, Ho SY, Guindon S (2012) PartitionFinder: combined selection of parti tioning schemes and substitution models for phylogenetic analyses. Molecular Biology and Evolution 29: 1695. doi: 10.1093/molbev/mss020 Larson D, LaBonte J (1994) Stygoporus oregonensis , a new genus and species of subterranean water beetle (Coleoptera: Dytiscidae: Hydroporini) from the United States. e Coleop teristsâ€™ Bulletin, 371. Larson DJ, Alarie Y, Roughley RE (2000) Predaceous diving beetles (Coleoptera: Dytiscidae) of the Nearctic Region, with emphasis on the fauna of Canada and Alaska. NRC Research Press, Canda, 982 pp. Leys R, Watts CH (2008) Systematics and evolution of the Australian subterranean hydropo rine diving beetles (Dytiscidae), with notes on Carabhydrus. Invertebrate Systematics 22: 217. doi: 10.1071/IS07034
88 Leys R, Watts CH, Cooper SJ, Humphreys WF (2003) Evolution of subterranean diving beetles (Coleoptera: Dytiscidae Hydroporini, Bidessini) in the arid zone of Australia. Evolution 57: 2819. doi: 10.1111/j.0014-3820.2003.tb01523.x Maddison D, Maddison W (2011) Chromaseq: a Mesquite package for analyzing sequence chromatograms. Version 1.0. http://mesquiteproject org/packages/chromaseq Maddison D, Maddison W (2015) Zephyr: a Mesquite package for interacting with external phylogeny inference program. Version 1.1. https://mesquitezephyr.wikispaces.com Maddison W, Maddison D (2015) Mesquite: a modular system for evolutionary analysis. v3.02. http://mesquiteproject .org Mazza G, Cianferoni F, Rocchi S (2013) Etruscodytes nethuns n. gen., n. sp.: the rst phreatic water beetle from Italy (Coleoptera: Dytiscidae: Hydroporinae). Italian Journal of Zoology 80: 233. doi: 10.1080/11250003.2013.783633 Miller KB, Jean A, Alarie Y, Hardy N, Gibson R (2013) Phylogenetic placement of North American subterranean diving beetles (Insecta: Coleoptera: Dytiscidae). Arthropod Sys tematics & Phylogeny 71: 75. Miller KB (2001) On the phylogeny of the Dytiscidae (Insecta: Coleoptera) with emphasis on the morphology of the female reproductive system. Insect Systematics & Evolution 32: 45. doi: 10.1163/187631201X00029 Miller KB (2005) Revision of the New World and southeast Asian Vatellini (Coleoptera: Dytiscidae: Hydroporinae) and phylogenetic analysis of the tribe. Zoological Journal of the Linnean Society 144: 415. doi: 10.1111/j.1096-3642.2005.00180.x Miller KB (2016) Revision of the Neotropical diving beetle genus Hydrodessus J. BalfourBrowne, 1953 (Coleoptera, Dytiscidae, Hydroporinae, Bidessini). ZooKeys 580: 45. doi: 10.3897/zookeys.580.8153 Miller KB, Bergsten J (2014) e phylogeny and classication of predaceous diving beetles (Coleoptera: Dytiscidae). In: Yee DA (Ed.) Ecology, systematics, and the natural history of predaceous diving beetles (Coleoptera: Dytiscidae). Springer, New York, USA, 49. Miller KB, Gibson JR, Alarie Y (2009) North American stygobiontic diving beetles (Coleop tera: Dytiscidae: Hydroporinae) with description of Ereboporus naturaconservatus Miller, Gibson and Alarie, new genus and species, from Texas, USA. e Coleopterists Bulletin 63: 191. doi: 10.1649/1124.1 Miller KB, Wolfe WG, Bistrm O (2006) e phylogeny of the Hydroporinae and classica tion of the genus Peschetius Guignot, 1942 (Coleoptera: Dytiscidae). Insect Systematics & Evolution 37: 257. doi: 10.1163/187631206788838617 Nilsson AN (2001) World catalogue of insects. Volume 3: Dytiscidae (Coleoptera). Apollo Books, Stenstrup, 395 pp. Post DL (2010) Habitat Identication for ree California Species of Sanlippodytes Francis colo (Coleoptera: Dytiscidae). Coleopterists Bulletin 64: 258. doi: 10.1649/0010065X-64.3.258.13 Ribera I, Castro A, Hernando C (2010) Ochthebius ( Enicocerus ) aguilerai sp. n. from central Spain, with a molecular phylogeny of the Western Palaearctic species of Enicocerus (Co leoptera, Hydraenidae). Zootaxa 2351: 1.
89 Ribera I, Faille A (2010) A new microphthalmic stygobitic Graptodytes Seidlitz from Morocco, with a molecular phylogeny of the genus (Coleoptera, Dytiscidae). Zootaxa 2641: 1. Ribera I, Hogan JE, Vogler AP (2002) Phylogeny of hydradephagan water beetles inferred from 18S rRNA sequences. Molecular Phylogenetics and Evolution 23: 43. doi: 10.1006/ mpev.2001.1080 Ribera I, Vogler AP, Balke M (2008) Phylogeny and diversication of diving beetles (Coleop tera: Dytiscidae). Cladistics 24: 563. doi: 10.1111/j.1096-0031.2007.00192.x Spangler P, Botosaneanu L (1986) Insecta: Coleoptera. In: Botosaneanu L, Stock JH (EDS) Sty gofauna Mundi A faunistic distributional and ecological synthesis of the world fauna inhab iting subterranean waters (including the marine interstitial). E. J. Brill, Leiden, 622. Spangler P, Decu V (1998) Coleoptera aquatica. In: Juberthie C, Decu V (Eds) Encyclopaedia Biospeologica II. Socit de Biospologie, Moulis, 1031. Spangler PJ, Barr CB (1995) A new genus and species of stygobiontic dytiscid beetle, Co maldessus stygius (Coleoptera: Dytiscidae: Bidessini) from Comal Springs, Texas. Insecta Mundi 9(3): 301. Stamatakis A (2014) RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics: btu033. doi: 10.1093/bioinformatics/btu033 Toussaint EF, Condamine FL, Hawlitschek O, Watts CH, Porch N, Hendrich L, Balke M (2015) Unveiling the diversication dynamics of Australasian predaceous diving beetles in the Cenozoic. Systematic Biology 64(1): 3. doi: 10.1093/sysbio/syu067 Watts C, Humphreys W (2009) Fourteen new Dytiscidae (Coleoptera) of the genera Limbodessus Guignot, Paroster Sharp, and Exocelina Broun from underground waters in Australia. Trans actions of the Royal Society of South Australia 133: 62. Wiens JJ, Chippindale PT, Hillis DM (2003) When are phylogenetic analyses misled by con vergence? A case study in Texas cave salamanders. Systematic Biology 52: 501. doi: 10.1080/10635150390218222 Young FN, Longley G (1976) A new subterranean aquatic beetle from Texas (Coleoptera: Dytiscidae-Hydroporinae). Annals of the Entomological Society of America 69: 787. doi: 10.1093/aesa/69.5.787
90 Supplementary material 1 Figure 1 Authors: Kojun Kanda, R. Antonio Gomez, Richard Van Driesche, Kelly B. Miller, David R. Maddison Data type: Adobe PDF le Explanation note: Maximum likelihood trees for single gene datasets. Scale bar indi cates the expected substitutions per site as estimated by RAxML. Copyright notice: is dataset is made available under the Open Database License ( http://opendatacommons.org/licenses/odbl/1.0/ ). e Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Supplementary material 2 Figure 2 Authors: Kojun Kanda, R. Antonio Gomez, Richard Van Driesche, Kelly B. Miller, David R. Maddison Data type: Adobe PDF le Explanation note: Majority rule consensus of 1,000 bootstrap replicates performed on single gene datasets. Bootstrap percentage given for clades recovered with more than 50% support. Copyright notice: is dataset is made available under the Open Database License ( http://opendatacommons.org/licenses/odbl/1.0/ ). e Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Supplementary material 3 Table 1 Authors: Kojun Kanda, R. Antonio Gomez, Richard Van Driesche, Kelly B. Miller, David R. Maddison Data type: MS Word le Explanation note: Collection and specimen data for material examined in this study. Copyright notice: is dataset is made available under the Open Database License ( http://opendatacommons.org/licenses/odbl/1.0/ ). e Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.
91 Supplementary material 4 Table 2 Authors: Kojun Kanda, R. Antonio Gomez, Richard Van Driesche, Kelly B. Miller, David R. Maddison Data type: MS Word le Explanation note: PCR primers and amplication conditions for sampled gene fragments. Copyright notice: is dataset is made available under the Open Database License ( http://opendatacommons.org/licenses/odbl/1.0/ ). e Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Supplementary material 5 Table 3 Authors: Kojun Kanda, R. Antonio Gomez, Richard Van Driesche, Kelly B. Miller, David R. Maddison Data type: MS Word le Explanation note: Taxa from Miller et al. (2013) sampled in this study with updated tribal and subtribal classication of Miller and Bergsten (2014). Copyright notice: is dataset is made available under the Open Database License ( http://opendatacommons.org/licenses/odbl/1.0/ ). e Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Supplementary material 6 Data availability Authors: Kojun Kanda, R. Antonio Gomez, Richard Van Driesche, Kelly B. Miller, David R. Maddison Data type: NEXUS le Explanation note: NEXUS formatted single-gene and concatenated nucleotide sequence alignments. Copyright notice: is dataset is made available under the Open Database License ( http://opendatacommons.org/licenses/odbl/1.0/ ). e Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.