A New Distinctive Darwin Wasp Represents the First Record of the Ophion minutus Species-group (Hymenoptera: Ichneumonidae: Ophioninae) from Japan and the Far East, with an Analysis of DNA Barcode-based Species Delimitation in Ophion
Abstract
Ng and Hsi-Te Shih, Peter K. L. (2023): A New Distinctive Darwin Wasp Represents the First Record of the Ophion minutus Species-group (Hymenoptera: Ichneumonidae: Ophioninae) from Japan and the Far East, with an Analysis of DNA Barcode-based Species Delimitation in Ophion. Zoological Studies 62 (27): 1-14, DOI: 10.6620/ZS.2023.62-27, URL: http://dx.doi.org/10.5281/zenodo.13917036
Full text
© 2023 Academia Sinica, Taiwan Open Access A New Distinctive Darwin Wasp Represents the First Record of the Ophion minutus Species-group (Hymenoptera: Ichneumonidae: Ophioninae) from Japan and the Far East, with an Analysis of DNA Barcode-based Species Delimitation in Ophion So Shimizu1,2,* and Kaoru Maeto2 1Institute for Agro-Environmental Sciences, NARO, Kannon-Dai, Tsukuba-Shi, Ibaraki, 305-8604, Japan. *Correspondence: E-mail: [email protected] (Shimizu) 2Laboratory of Insect Biodiversity and Ecosystem Science, Graduate School of Agricultural Science, Kobe University, Rokkodaicho 1–1, Nada, Kobe, Hyogo, 657–8501, Japan. E-mail: [email protected] (Maeto) Received 9 January 2023 / Accepted 21 February 2023 / Published 26 May 2023 Communicated by Y. Miles Zhang A new Darwin wasp species, Ophion kobensis Shimizu sp. nov. (Hymenoptera: Ichneumonidae: Ophioninae), is described using the integrated morphological and molecular species delimitation approaches. Our results indicate that the new species is closely related to European O. ventricosus Gravenhorst, 1829 of the O. minutus species-group but can be distinguished using morphological characters, such as entirely black body colour with some light-yellow marks and not inclined epicnemial carina in lateral view. This record of the new species represents the first record of O. minutus speciesgroup from Japan and the Far East. Phylogenetic analysis indicate that the O. minutus species-group is weakly recovered as monophyletic and sister to Ophion s. str. The analysis also indicated that two clades within the O. minutus species-group (O. minutus and O. ventricosus) diverged significantly. This suggests that the species-group, as well as the two included clades, could potentially be treated as separate species-groups or genera. The present study supports previous integrative taxonomic and phylogenetic studies of Ophion and represents a first fundamental step for studies focused on the challenging Japanese and Far Eastern Ophion. Key words: bPTP, Integrative taxonomy, MtCO1, Parasitoid wasps, Phylogeny. Citation: Shimizu S, Maeto K. 2023. A new distinctive Darwin wasp represents the first record of the Ophion minutus species-group (Hymenoptera: Ichneumonidae: Ophioninae) from Japan and the Far East, with an analysis of DNA barcode-based species delimitation in Ophion. Zool Stud 62:27. doi:10.6620/ZS.2023.62-27. BACKGROUND Darwin wasps of the predominantly temperate genus Ophion Fabricius, 1798 (Hymenoptera, Ichneumonidae, Ophioninae) are renowned as some of the most common midto large-sized insects that frequently visit lights at night, together with many other nocturnal insects (Townes 1971; Gauld and Mitchell 1981; Gauld 1985; Schwarzfeld et al. 2016). They are solitary koinobiont endoparasitoids of midto largesized lepidopteran larvae (Uchida 1928 1954; Townes 1971; Gauld and Mitchell 1981). Adult wasps of Ophion exhibit the convergent morphology of “ophionoid facies”, characterised by, among others, an orangeZoological Studies 62:27 (2023) doi:10.6620/ZS.2023.62-27 1
© 2023 Academia Sinica, Taiwan brown body, large ocelli, and long antennae, associated with their nocturnal behaviour (Gauld and Huddleston 1976). Although over 150 valid species of Ophion have been recognised worldwide (Yu et al. 2016; Johansson and Cederberg 2019; Johansson et al. 2021), their taxonomy does not reflect their true species richness, and most species, like other Darwin wasps (Klopfstein et al. 2019), still need to be described. Although many studies have attempted to perform morphology-based taxonomy of Ophion (Gauld and Mitchell 1981; Brock 1982), it is very difficult to delimit species of Ophion based solely on morphological characters due to a wide range of intraspecific morphological variations (Townes 1971; Brock 1982). Integrative taxonomy aims to correctly delimit and describe taxa by incorporating multiple perspectives (e.g., morphology, ecology, and genetics). It is one of the best solutions to resolve taxonomic problems in morphologically challenging and poorly known diverse groups and to understand their biodiversity (Dayrat 2005; Schlick-Steiner et al. 2010; Padial et al. 2010; Ito et al. 2015; Shimizu et al. 2019 2020). Therefore, recent studies on Ophion have employed integrative taxonomic frameworks (at least integrating morphology and DNA barcodes) (Schwarzfeld and Sperling 2014 2015; Johansson and Cederberg 2019). The Ophion minutus species-group consists of small-sized wasps with typical ophionoid facies, which parasitise geometrid moths. This species-group is a wellrecovered monophyletic basal lineage within the genus Ophion and is considered to be a potential candidate for a separate genus (Schwarzfeld et al. 2016; Johansson and Cederberg 2019). It is distinguishable from all other Ophion species by the proximally thickened and slightly angled fore wing vein 2r&RS (Gauld 1985; Schwarzfeld et al. 2016), similar to related non-Ophion genera, such as Eremotylus Förster, 1869 and Stauropoctonus Brauns, 1889. After the phylogenetic reconstruction of the species-groups of Ophion by Schwarzfeld et al. (2016), the number of species in the species-group increased by the great taxonomic studies of a Swedish Entomologist, Mr. Niklas Johansson (Johansson and Cederberg 2019; Johansson et al. 2021). Gauld (1985) suggested that O. minutus could belongs to the O. bicarinatus speciesgroup, previously consisting of Australian species, but reliable evidence has not been presented and further phylogenetic studies are required (Schwarzfeld et al. 2016). Most records of the O. minutus species-group are from Europe, with a few from the western portion of the Eastern Palaearctic region (Yu et al. 2016; Johansson and Cederberg 2019; Johansson et al. 2021), and no Darwin wasps of the species-group have been recorded from the Far East. The Japanese fauna of Ophion has been poorly studied (Smith 1874; Uchida 1928 1954), with only 10 valid species records (Yu et al. 2016), although over 50 morphospecies have been recognised by the first author (Shimizu unpublished data). Therefore, revisional studies are necessary but are difficult at present due to the enormous species richness, taxonomic difficulties, and incomplete sampling. The Ophion minutus speciesgroup is not known from Japanese fauna, but the first author of this paper recently recognised specimens of the species-group from Japan for the first time. Therefore, as a first step in the taxonomy of the Japanese Ophion, the present study aims to report the O. minutus species-group from Japan for the first time and to classify it using integrated morphology and molecular species delimitation methods, with molecular phylogenetic reconstruction for the genus Ophion to test the phylogenetic position of the Japanese species. MATERIALS AND METHODS Terms and indices Morphological terms and indices followed those of Broad et al. (2018), Johansson and Cederberg (2019), and Shimizu et al. (2019 2020). The density of punctures was indicated as follows: “densely” = punctures separated by approximately their own diameter or less, “moderately” = punctures separated by approximately twice their own diameter, “sparsely” = punctures separated by over triple their own diameter. The “mandibular gape” was used for the angle between mandibular teeth (Brock 1982; Johansson and Cederberg 2019). The following abbreviations were used: FL = flagellomere (e.g., FL1 = 1st and FL2 = 2nd flagellomeres), IOD (inter-ocellar distance) = shortest distance between inner margin of lateral ocelli, LOD (lateral-ocellar diameter) = maximum diameter of lateral ocelli, OOD (orbito-ocellar distance) = shortest distance between outer margin of lateral ocellus and orbit of eye, POD (post-ocellar distance) = shortest distance between posterior margin of lateral ocellus and anterior margin of occipital carina, S = metasomal sternite (e.g., S1 = 1st and S2 = 2nd metasomal sternites), and T = metasomal tergite (e.g., T1 = 1st and T2 = 2nd metasomal tergites). Examined samples The materials were deposited at the Institute for Agro-Environmental Sciences, NARO (= National Institute for Agro-Environmental Sciences), Tsukuba, Japan (NIAES). page 2 of 14Zoological Studies 62:27 (2023)
© 2023 Academia Sinica, Taiwan Morphological observation Morphology of the examined adult wasps was observed under a stereoscopic microscope (SMZ1500, Nikon, Tokyo, Japan). For accurate measurement, left wings were removed from the body, and the wing wrinkles were pressed out by putting them between two microscope slides in 80% ethanol. Then, wing photos were taken (cf. below for photo technique). Finally, the wing measurements were conducted using Adobe Photoshop CC v.20.0.4 (Adobe Systems Inc., San Jose, CA, USA). Body length was measured between the wasp head and the terminus of the metasoma, not including antennae and ovipositor. Multi-focus photographs were taken using a single lens reflex camera (α7II, Sony, Tokyo, Japan) with a micro-lens (LAOWA 25 mm F2.8 2.5–5× ULTRA MACRO, Anhui Changgeng Optics Technology Co., Ltd, Hefei, China and A FE 50mm F2.8 Macro SEL50M28, Sony, Tokyo, Japan), captured in RAW format, developed using Adobe Lightroom CC v.2.2.1 (Adobe Systems Inc., San Jose, CA, USA), and stacked using Zerene Stacker v.1.04 (Zerene Systems LLC., Richland, WA, USA). Molecular species delimitation Partial sequences of a mitochondrial proteinencoding gene, cytochrome c oxidase 1 (CO1), frequently referred to as the DNA barcoding gene for arthropods, were sequenced from the Japanese species of the O. minutus species-group for the delimitation of molecular species. DNA was extracted from the right fore or mid-leg, and DNA extraction and sequencing protocols followed those described by Shimizu et al. (2020). Sequences of the global Ophion species were compiled from BOLD systems (Ratnasingham and Hebert 2007) (available at: http://v3.boldsystems.org/) (accessed on 14 Dec. 2022), and together with the sequences of the Japanese species of the O. minutus species-group, a multiple sequence alignment (MSA) was conducted using MEGA v.10.0.5 (Kumar et al. 2018) based on both nucleotides and amino acids (see Fujie et al. 2021 for detailed protocols). Poorly aligned regions and identical haplotypes were removed from the MSA using trimAl v.1.2 (Capella-Gutiérrez et al. 2009) and the web server of ALTER (Glez-Peña et al. 2010) (available at: http://sing.ei.uvigo.es/ALTER/) respectively. Uncorrected pairwise nucleotide genetic distances (p-distances) were calculated using MEGA. As Schwarzfeld and Sperling (2015) concluded in the comparison of the species delimitation methods using Ophion, tree-based General Mixed Yule Coalescent (GMYC) (Pons et al. 2006; Fontaneto et al. 2007; Fujisawa and Barraclough 2013) and Poisson Tree Processes (PTP) (Zhang et al. 2013) models were less reliant on arbitrary parameters than the distancebased threshold and Automatic Barcode Gap Discovery (ABGD) (Puillandre et al. 2012) methods. They also suggested that PTP was less successful than GMYC in delimiting test species. However, PTP is fast and easy to run without ultrametricised trees. This is a significant advantage over GMYC, which requires ultrametricised trees. Therefore, bPTP (Zhang et al. 2013), an updated version of the maximum likelihood PTP with Bayesian support values for delimited species, was selected to delimit the species in the present study. The bPTP input tree was generated with IQTREE v.2.1.2 (Minh et al. 2020) under the maximum likelihood framework. The optimal substitution models and partitioning schemes were determined using PartitionFinder v.2.1.1 (Lanfear et al. 2017) with the greedy search algorithm under the corrected Akaike information criterion (AIC). A Shimodaira-Hasegawalike approximate likelihood ratio test (SH-aLRT) (Guindon et al. 2010) and ultrafast likelihood bootstrap replicates (UFBoot2) (Minh et al. 2013; Hoang et al. 2018) were performed with 10,000 replicates. To reduce the risk of overestimating nodal supports, the ”-bnni” option was also employed. A standalone version of bPTP was run with the following MCMC parameters: iterations = 5,000,000, sampling interval = 200, burn-in = 0.50, and seed = 7777. The Bayesian support values of posterior probabilities (PP) for molecular operational taxonomic units (MOTUs, considered to be approximately equivalent to species) were interpreted as follows: strongly supported: 0.95 ≤ PP ≤ 1.00, moderately supported: 0.90 ≤ PP < 0.95, weakly supported: 0.80 ≤ PP < 0.90, and not supported: PP < 0.80. Molecular phylogeny A phylogenetic analysis was conducted to determine the phylogenetic position of the Japanese species of the Ophion minutus species-group. One sequence was selected per MOTU delimited by bPTP analysis to reduce sampling bias per species. Thirteen sequences from other ophionine genera, i.e., Afrophion hynnis (Gauld & Mitchell, 1978) (ASQIC051-09), Barytatocephalus mocsaryi (Brauns, 1895) (ASQIC059-09), Dicamptus sp. (ASQAS285-11), Dictyonotus sp. (GBMIN25686-13), Enicospilus ramidulus (Linnaeus, 1758) (COLHH305-18), Eremotylus sp. (BBHYG815-10), Hellwigiella dichromoptera (Costa, 1886) (ASQIC060-09), page 3 of 14Zoological Studies 62:27 (2023)
© 2023 Academia Sinica, Taiwan Leptophion sp. (GBMND19516-21), Riekophion sp. (GBAH21297-19), Skiapus sp. (ASQIC197-09), Stauropoctonus bicarinatus (Cushman, 1947) (STAT1173-07), Thyreodon sp. (ASINB1539-12), and Xylophion sevrapek Villemant, 2012 (ICHSA002-12), were selected as outgroups. Two non-ophionines, i.e., Habronyx (Camposcopus) nigricornis (Wesmael, 1849) (Anomaloninae) (COLHH348-18) and Cremastus incompletus (Provancher, 1875) (Cremastinae) (HYMBB978-10), were also selected to define a distinct root of Ophioninae. The phylogenetic tree was reconstructed in IQ-TREE using the same process as described above for the input tree preparation for bPTP (cf. above “Molecular species delimitation”). As recommended by the manual of IQ-TREE, nodal support values were interpreted as follows: supported = SH-aLRT ≥ 0.80 and UFBoot2 ≥ 0.95, weakly supported = SH-aLRT ≥ 0.80 or UFBoot2 ≥ 0.95, and not supported = SH-aLRT < 0.80 and UFBoot2 < 0.95. Figure editing All figures were edited in Adobe Illustrator CC v.23.0.2 or Photoshop CC v.20.0.4 (Adobe Systems Inc., San Jose, CA, USA). Phylogenetic trees were edited with the interactive Tree of Life (iTOL) (available at: https://itol.embl.de) (Letunic and Bork 2021). RESULTS Morphological species delimitation Based on two female specimens from the Kinki region, the western portion of Honshu Island in the Japanese Archipelago, only a single morphospecies of the O. minutus species-group was recognised in the Japanese fauna, as described as a new species below (O. kobensis Shimizu sp. nov.). The species displayed the most distinctive morphology (i.e., entirely black body with yellowish wings, and proximally thickened and slightly angled fore wing vein 2r&RS) among the Japanese species of Ophion but resembled the European O. ventricosus Gravenhorst, 1829. However, the body colour of O. kobensis sp. nov. was entirely black with some small light-yellow marks, and that of the European O. ventricosus was entirely testaceous with some small black marks; therefore, they could easily be distinguished from each other by their colouration. Additionally, they can be distinguished from each other by the characters summarized in table 1. Molecular species delimitation Dataset: A total of 1,606 public records with CO1 sequences of Ophion, forming 148 BINs, were compiled from BOLD systems. Among them, the deposited photograph of a single record of NBINS376-15 was not ophionine Darwin wasp; seven sequences (i.e., GBMIN71038-17 = GenBank KU753332 of O. obscuratus obscuratus, GBMIN77568-17 = KU753331 of O. luteus, GBAH23440-19 = KU753333 of O. obscuratus obscuratus, GBAH23448-19 = KU753334 of O. takaozanus, SSEIB12923-13, DBFCI467-15, and ACGBA1775-12 of Ophion spp.) were enigmatic with non-triplet indels or suspicious regions. Therefore, these eight sequences were removed from the dataset (1,598 sequences, forming 140 BINs). Additionally, two sequences of O. kobensis sp. nov. were obtained (DDBJ: LC756991 and LC756992). The aligned and trimmed dataset contained 1,600 sequences of 567 bp in length. This represented 616 haplotypes. Therefore, the final dataset without identical haplotypes consisted of 616 sequences (Supplementary file S1). It included 243 conserved and 324 variable sites, of which 268 were parsimony informative. Tree and phylogeny: The TVM+I+G model was selected as the optimal substitution model for each codon position under AIC. All branches of the BOLD systems project code of “HYSAF” (South African Hymenoptera) and “HYSA” (South Australian Hymenoptera) were significantly longer than those of the other Ophion (Fig. S1). However, most of the reconstructed topologies were consistent with those reported by Schwarzfeld et al. (2016) and Johansson and Cederberg (2019). For the O. minutus species-group (Fig. 1), the monophyly of the O. ventricosus clade (O. kobensis Table 1. Diagnostic characters between O. kobensis sp. nov. and O. ventricosus Character O. kobensis sp. nov. O. ventricosus Colour Entirely black with some small light-yellow marks Entirely testaceous with some small black marks Epicnemial carina in lateral view Not inclined to anterior Strongly inclined to anterior Mesopleural furrow Moderately crenulated Strongly and roughly crenulated Posterior transverse carina of propodeum Strong laterally and vestigial centrally Entirely very strong Propodeal profile Almost evenly rounded Irregularly angulated page 4 of 14Zoological Studies 62:27 (2023)
© 2023 Academia Sinica, Taiwan Ophion minutus ASQIC110-09 Ophion minutus GBMIN77574-17 Ophion minutus GMGMJ636-14 Ophion minutus GBMIN77573-17 Ophion minutus GBMIN77570-17 Ophion minutus GBMIN77575-17 Ophion minutus GMGMI290-14 Ophion minutus GBAH10275-15 Ophion minutus GBMIN77571-17 Ophion ventricosus GBMIN78012-17 Ophion ventricosus GBMIN78011-17 Ophion ventricosus GBMIN78009-17 Ophion sp. TDAAT106-19 Ophion ventricosus GMGMI283-14 Ophion kobensis sp. nov. SHOP24 LC756991 Ophion kobensis sp. nov. SHOP190 LC756992 0.488 0.474 0.440 0.10 0.756 0.866 0.05 0.00 Nodal supports ● 0.80≤ SH-aLRT and 0.95≤ UFBoot2 ◐ 0.80≤ SH-aLRT and UFBoot2 < 0.95 Colour stripes ■ 0.95 ≤ PP ≤ 1.00 ■ 0.90 ≤ PP < 0.95 ■ 0.80 ≤ PP < 0.90 ■ 0.50 ≤ PP < 0.80 ■ 0.00 ≤ PP < 0.50 sp. nov. + O. ventricosus) was highly supported (SHaLRT = 1.00 and UFBoot2 = 0.99), and O. minutus was also strongly recovered as a monophyletic clade (the O. minutus clade; SH-aLRT = 0.98 and UFBoot2 = 1.00). However, the monophyly of the O. minutus species-group was not recovered (SH-aLRT = 0.66 and UFBoot2 = 0.69). The two clades in this species-group were more or less divergent. Divergences in p-distances were 0.2–2.6% among the O. minutus clade, 2.7–4.2% among the O. ventricosus clade (0.2–0.7% among O. ventricosus and 0.7% among O. kobensis sp. nov.), and 12.2–14.3% between the two clades (Table 2). bPTP: The bPTP analysis estimated between 168 and 265 species (mean = 213.11), and the most supported result indicated that 199 species, consisting of 80 supported (i.e., 57 strongly, 7 moderately, and 16 weakly supported species) and 119 unsupported species, were delimited (Figs. 1, S1, Supplementary file S2). Ophion kobensis sp. nov. of the O. minutus speciesgroup was recovered as a weakly supported single species (PP = 0.866). Phylogeny Dataset: The final dataset consisted of 214 sequences of the CO1 gene (199 from the bPTP MOTUs of Ophion, 13 of the other ophionine genera, and two of the non-ophionines) and 555 bp in total length (Supplementary file S3), comprising 224 conserved and 331 variable sites, with 264 parsimony informative sites. Phylogeny: The TIM+I+G model for the 1st and 2nd codon positions and the GTR+I+G for the 3rd codon were selected as the optimal substitution model under AIC. Terminals with the BOLD systems project code of “HYSAF” and “HYSA” were strongly nested into ophionine outgroups (Figs. 2, S2). Our phylogeny result was mostly consistent with that reported by Schwarzfeld et al. (2016) and Johansson and Cederberg (2019). Except for the O. minutus species-group, Ophion species were weakly supported as a monophyletic clade (referred to as Ophion s. str. here; Figs. 2, S2), with SH-aLRT = 0.93 and UFBoot2 = 0.72. However, Hellwigiella dichromaptera (ASQIC060-09) was nested in the Ophion s. str. Among the Ophion s. str., Fig. 1. Result of the bPTP molecular species delimitation test of Ophion kobensis Shimizu sp. nov. among the genus Ophion, based on the maximum likelihood CO1 gene topology reconstructed by IQ-TREE2. This figure only shows the O. minutus species-group but whole result is available at figure S1. Stripes indicate delimited molecular operational taxonomic units (MOTUs), and stripe colours and numbers on them indicate a degree of the Bayesian supports of posterior probabilities (PP) to delimited species. Pie charts on each node indicate nodal support of SH-aLRT ≥ 0.80 and/or UFBoot2 ≥ 0.95. Ophion kobensis Shimizu sp. nov. was recovered as a single MOTU (PP = 0.866). page 5 of 14Zoological Studies 62:27 (2023)
© 2023 Academia Sinica, Taiwan the monophyly of the following species-groups were strongly recovered: the O. scutellaris species-group (SH-aLRT = 0.98 and UFBoot2 = 0.99) and the O. slossonae species-group (SH-aLRT = 0.99 and UFBoot2 = 1.00). The O. minutus species-group (Figs. 2, S2) was weakly recovered as a monophyletic clade (SH-aLRT = 0.84 and UFBoot2 = 0.76) and placed as a sister to Ophion s. str., together with Xylophion sevrapek (ICHSA002-12) and Ophion sp. (GBAH2130419). However, the monophyly of Ophion s. lat. (the O. minutus species-group + Ophion s. str.) was not recovered. Ophion kobensis sp. nov. of the group was strongly recovered as a member of the O. ventricosus clade (SH-aLRT = 1.00 and UFBoot2 = 1.00). The O. minutus clade was also strongly recovered (SH-aLRT = 0.96 and UFBoot2 = 0.99). TAXONOMY As mentioned above, O. kobensis sp. nov. was recovered as a single species and was easily distinguishable from other previously described species based both on the morphological and molecular species delimitations. Therefore, it is described as a new species below. Ophion kobensis Shimizu sp. nov. [Japanese name: Kobe-amebachi] (Figs. 3–5) urn:lsid:zoobank.org:act:AD2F2A64-1324-4B7B-A3394963585070F9 Materials examined (type series): Holotype ♀ (NIAES) (SHOP24) (Figs. 3–5): 34°43'27"N 135°13'58"E (110 m alt.), Kobe Univ., Kobe City, Hyogo Pref., Japan, 29.IV.2015, Masato Ito leg. Paratype ♀ (NIAES) (SHOP190): N34.281 E135.4077, Kokawa, Kinokawa City, Wakayama Pref., Japan, 1–31. V.2016, Takuto Hirooka leg. by light trap. Etymology: Named after the type locality, Kobe, Hyogo, Japan. Description of holotype female (Figs. 3–5): Body entirely more or less shiny (Figs. 3, 4), and length ca. 13.5 mm. Head buccate behind compound eyes in dorsal view (Fig. 4c) with GOI = 1.3 (Fig. 4b). Antenna with 49 flagellomeres; FL1 3.2× longer than wide and 1.5× longer than FL2; FL20 and FL40 1.6× longer than wide. Lower face transverse and 1.2× wider than high, strongly polished, and densely punctate with setae (Fig. 4a); clypeus 2.2× wider than high, slightly convex in profile, and ventral margin evenly rounded in frontal view and weakly impressed and sub-blunt in profile (Fig. 4a, b). Malar space 0.4× longer than basal mandibular width. Mandibular gape slightly acute angle, with internal angle. Posterior ocelli not touching eye (Fig. 4c). OOD / LOD = 0.4; IOD / LOD = 0.5; POD / LOD = 1.0. Occipital carina complete, mediodorsally evenly curved; ventral end joining oral carina. Mesosoma entirely weakly shagreened to polished with sparse to dense punctures and setae (Fig. 4d–f). Mesoscutum 1.3× longer than width (Fig. 4e); polished (Figs. 4d–f); notauli absent (Fig. 4d–f); and evenly rounded in profile (Fig. 4d). Scutellum moderately convex in profile (Fig. 4d) and narrowed posteriorly in dorsal view with lateral longitudinal carinae along anterior 0.4 of scutellum (Fig. 4e, f). Epicnemial Table 2. Divergences of p-distances among the Ophion minutus species-group Species Sequence ID ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 O. minutus GBMIN77575-17 1 O. minutus GMGMJ636-14 2 0.004 O. minutus GMGMI290-14 3 0.002 0.004 O. minutus GBMIN77573-17 4 0.002 0.002 0.002 O. minutus GBMIN77570-17 5 0.004 0.004 0.004 0.002 O. minutus ASQIC110-09 6 0.006 0.002 0.004 0.004 0.006 O. minutus GBAH10275-15 7 0.024 0.025 0.021 0.023 0.019 0.026 O. minutus GBMIN77571-17 8 0.023 0.023 0.019 0.021 0.018 0.024 0.002 O. minutus GBMIN77574-17 9 0.004 0.002 0.006 0.004 0.006 0.002 0.026 0.024 O. ventricosus GBMIN78012-17 10 0.139 0.139 0.136 0.138 0.134 0.143 0.136 0.134 0.138 O. ventricosus GBMIN78011-17 11 0.137 0.138 0.134 0.136 0.133 0.141 0.134 0.133 0.137 0.000 O. ventricosus GMGMI283-14 12 0.128 0.127 0.123 0.125 0.122 0.133 0.125 0.123 0.129 0.006 0.004 O. ventricosus GBMIN78009-17 13 0.139 0.140 0.136 0.138 0.135 0.143 0.136 0.135 0.139 0.000 0.000 0.006 O. sp. TDAAT106-19 14 0.136 0.136 0.134 0.134 0.131 0.139 0.136 0.134 0.135 0.002 0.002 0.007 0.002 O. kobensis sp. nov. SHOP24 15 0.128 0.127 0.124 0.125 0.126 0.133 0.134 0.133 0.127 0.035 0.034 0.027 0.036 0.038 O. kobensis sp. nov. SHOP190 16 0.133 0.128 0.128 0.130 0.132 0.134 0.137 0.137 0.128 0.039 0.038 0.031 0.040 0.042 0.007 page 6 of 14Zoological Studies 62:27 (2023)
© 2023 Academia Sinica, Taiwan Fig. 2. Maximum likelihood CO1 gene phylogeny of the genus Ophion based on bPTP MOTUs. Gray terminals indicate the apparently non-Ophion species although they have been deposited in the database as Ophion species. Ophion s. str. was collapsed (cf. Fig. S2 for whole tree). Pie charts on each node indicate a nodal support of SH-aLRT ≥ 0.80 and/or UFBoot2 ≥ 0.95. The Ophion minutus species-group and two well-recovered clades were represented as coloured clades. Ophion kobensis Shimizu sp. nov. was recovered as a sister to O. ventricosus. (SH-aLRT = 0.99 and UFBoot2 = 1.00). ventricosus clade minutus clade minutus group outgroups Ophion minutus GMGMJ636-14 Ophion minutus GMGMI290-14 Ophion minutus GBAH10275-15 Ophion ventricosus GBMIN78012-17 Ophion kobensis sp. nov. SHOP24 LC756991 Xylophion sevrapek ICHSA002-12 Ophion sp. Oz350 GBAH21304-19 [non-Ophion] Ophion sp. HYSAF038-12 [non-Ophion] Afrophion hynnis ASQIC051-09 Dicamptus sp. ASQAS285-11 Eremotylus sp. BBHYG815-10 Stauropoctonus bicarinatus ASTAT1173-07 Ophion sp. BBHYA4098-12 [non-Ophion] Ophion sp. HYSAF041-12 [non-Ophion] Thyreodon sp. Janzen12 ASINB1539-12 Riekophion sp. Oz313 GBAH21297-19 Ophion sp. ASQIC092-09 [non-Ophion] Enicospilus ramidulus COLHH305-18 Ophion sp. HYAS1036-11 [non-Ophion] Dictyonotus sp. GBMIN25686-13 Barytatocephalus mocsaryi ASQIC059-09 Ophion sp. HYAS1301-12 [non-Ophion] Ophion sp. HYAS934-11 [non-Ophion] Ophion sp. HYAS1030-11 [non-Ophion] Leptophion sp. GBMND19516-21 Skiapus sp. ASQIC197-09 Cremastus incompletus HYMBB978-10 Habronyx nigricornis COLHH348-18 Nodal supports ● 0.80≤ SH-aLRT and 0.95≤ UFBoot2 ◐ 0.80≤ SH-aLRT and UFBoot2 < 0.95 Ophion s. str. 0.25 0.00 page 7 of 14Zoological Studies 62:27 (2023)
© 2023 Academia Sinica, Taiwan carina, in lateral view, almost straight and not inclined to anterior and upper end evenly curved to anterior (Fig. 4d); and, in antero-ventral view, virtually evenly curved, therefore pleurosternal angles strongly obtuse and sternal angles almost absent. Submetapleural carina broadened anteriorly (Fig. 4d). Propodeum rugose and almost evenly rounded in profile (Fig. 4d, f); anterior transverse carina absent (Fig. 4d, f); posterior transverse carina strongly defined laterally but vestigial and ill-defined centrally; lateromedian longitudinal carina present between about anterior transverse carina and posterior end of propodeum (Fig. 4d, f); lateral longitudinal carina complete between posterior longitudinal carina and posterior end of propodeum (Fig. 4d, f); spiracle elliptical and joining pleural carina (Fig. 4d, f). Wings (Fig. 5). Fore wing (Fig. 5a) length ca. 10.5 mm with AI = 1.4, CI = 0.5, DI = 0.6, ICI = 0.7, SDI = 1.2, SRI = 0.5; vein 2r&RS weakly thickened and angulated proximally; vein 1m-cu&M between ramellus and bulla as long as 1m-cu&M between bulla and 2m-cu; ramellus short and less longer than bulla of 1m-cu&M; angles between 1m-cu&M and M (between 2rs-m and 2m-cu) 153°; vein 1cu-a postfurcal to M&RS by 0.1× length of 1cu-a. Hind wing (Fig. 5b) with NI = 1.2; vein RA with 8 uniform hamuli. Legs. Hind leg with trochantellus shorter than wide in dorsal view; femur 5.1× longer than wide; inner spur 1.6× longer than outer one and 0.5× longer than basitarsus; basitarsus 2.5× longer than 2nd tarsomere; tarsal claw simply pectinate. Metasoma (Fig. 4g) with DMI = 0.9, PI = 1.6. T1–2 polished; T3 and after that shagreened; posterior margin of anterior sclerotised section of S1 distinctly posterior to spiracle at nearly equal distance to shortest distance between spiracle and ventral margin of T1; T2 2.2× longer than deep; T6 distinctly vertically long and 0.4× longer than deep. Ovipositor sheath not longer than posterior depth of metasoma. Colour (Figs. 3–5). Entirely blackish. Head with antennae brown; lower face black, and inner orbit and clypeus yellow; mandible yellow and apical teeth black; outer orbit yellow; frons, stemmaticum, vertex, and occiput black. Mesosoma entirely black, except for ventral end of pronotum, anterior margin of mesopleuron, and scutellum yellow. Wing membrane Fig. 3. Habitus of female holotype of Ophion kobensis Shimizu sp. nov. 2.0 mm page 8 of 14Zoological Studies 62:27 (2023)
© 2023 Academia Sinica, Taiwan strongly yellowish; veins dark brown to black. Legs with all coxae and hind femur black; fore and mid legs excluding coxae, hind trochanter and trochantellus more or less testaceous; hind legs after tibia testaceous to brownish. Metasoma entirely dark brown to black except for testaceous parts on T1–3 and S1–3 and ovipositor. Variation in females (n = 2): Body length 13.5– 15.5 mm. Head with GOI = 1.3–1.5. Antenna with 49–55 flagellomeres; FL1 3.2–3.3× longer than wide; FL20 1.3–1.6× and FL40 1.5–1.6× longer than wide. Clypeus 2.1–2.2× wider than high. Malar space 0.3–0.4× longer than basal mandibular width. OOD / LOD = 0.3–0.4. Mesoscutum 1.3–1.4× longer than width. Fore wing length ca. 10.5–12.5 mm with AI = 1.3–1.4, ICI = 0.7–0.8, SDI = 1.0–1.2; angles between 1m-cu&M and M (between 2rs-m and 2m-cu) 148–153°; vein 1cu-a postfurcal to M&RS by 0.1–0.2× length of 1cu-a. Hind Fig. 4. Detailed photographs of Ophion kobensis Shimizu sp. nov. (a) head, frontal view; (b) head, lateral view; (c) head, dorsal view; (d) mesosoma, lateral view; (e) mesoscutum and scutellum, dorsal view; (f) mesosoma, dorso-lateral view; (g) metasoma, lateral view. abc df eg page 9 of 14Zoological Studies 62:27 (2023)