See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/297518833

†Alienoptera - A new order in the roach- mantodean twilight zone

Article in Gondwana Research · March 2016 DOI: 10.1016/j.gr.2016.02.002

CITATIONS READS 12 1,036

6 authors, including:

Ming Bai Rolf Beutel Chinese Academy of Sciences Friedrich Schiller University Jena

127 PUBLICATIONS 374 CITATIONS 287 PUBLICATIONS 6,272 CITATIONS

SEE PROFILE SEE PROFILE

Klaus-Dieter Klass Benjamin Wipfler Senckenberg Research Institute Friedrich Schiller University Jena

76 PUBLICATIONS 1,451 CITATIONS 60 PUBLICATIONS 931 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Insect Systematics and Evolutionary Biology Forum View project

Genitalic region in Archaeognatha, Zygentoma and Odonata: comparative morphology, functional reconstructions, phylogenetic value, and evolution View project

All content following this page was uploaded by Rolf Beutel on 04 April 2016.

The user has requested enhancement of the downloaded file. GR-01582; No of Pages 10 Gondwana Research xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Gondwana Research

journal homepage: www.elsevier.com/locate/gr

†Alienoptera — A new insect order in the roach–mantodean twilight zone

Ming Bai a, Rolf Georg Beutel b,⁎, Klaus-Dieter Klass c,WeiweiZhanga,d,XingkeYanga,BenjaminWipfler a,b a Key Laboratory of Zoological Systematics and Evolution, Institute of Zoology, Chinese Academy of Sciences, Box 92, Beichen West Road, Chaoyang District, Beijing 100101, China b Entomology Group, Institut für Spezielle Zoologie und Evolutionsbiologie mit Phyletischem Museum, Friedrich-Schiller-Universität Jena, Erbertstrasse 1, D-07743 Jena, Germany c Senckenberg Natural History Collections Dresden, Königsbrücker Landstrasse 159, D-01109 Dresden, Germany d P.O. Box 4680, Chongqing 400015, China article info abstract

Article history: A new insect species (†Alienopterus brachyelytrus Bai, Beutel, Klass, Wipfler et Zhang gen. et sp. nov.) of a new Received 20 November 2015 order and family is described, based on a single male embedded in (ca. 99 Ma). Un- Received in revised form 29 January 2016 usual characters are shortened forewings combined with fully developed, operational hindwings, similar as in Accepted 2 February 2016 Dermaptera, and specialized attachment pads otherwise only found in mantophasmatodeans (heelwalkers). A Available online xxxx cladistic analysis suggests a placement as sister to Mantodea, supported by a profemoral brush and other charac- fl Handling Editor: I.D. Somerville ters. The male genitalia show the same pattern in both groups. Specialized features are the unusual ight appa- ratus, attachment structures adapted for locomotion on leaves, and a dense profemoral setation suitable for Keywords: catching small prey. †Alienopterus was apparently able to fly and likely a predator of small in bushes Alienoptera or trees. An impressive radiation of Mantodea started in similar habitats at least 35 Ma later in the early Cenozoic. New order In contrast, †Alienopterus was an evolutionary dead end in the roach– transition zone. Insecta © 2016 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved. Mantodea Amber Cretaceous

1. Introduction distinctly shortened ovipositor distinguishes “modern” dictyopterans from the “roachoids”, which persisted up to the end of the Mesozoic In this study we describe a new extinct insect species in a new (Grimaldi and Engel, 2005; Legendre et al., 2015). monotypic order. Its potential role as a link between roach-like In the typical case roaches are saprophagous and solitary with a and the praying is evaluated. The single specimen with a high- ground-oriented lifestyle (Bell et al., 2007), and the same is likely true ly unusual character combination is embedded in Cretaceous Burmese for most of the “roachoids”. However, during their history of more amber, which has turned out as a very rich source of insect fossils than 300 Ma, more or less highly specialized groups have evolved sever- (Grimaldi et al., 2002). al times, for example the Mesozoic †Umenocoleidae, a group strongly The earliest roach-like insects known from the Upper resembling beetles, with heavily sclerotized tegmina and a small were already remarkably diverse (Labandeira, 1994; Grimaldi and pronotum (Vršanský, 2003; Grimaldi and Engel, 2005; Nel et al., Engel, 2005; Legendre et al., 2015). Similar to extant , 2014), or the and Cretaceous †Raphidiomimidae (Liang et al., Palaeozoic forms with a preserved body had a discoid pronotum partly 2009), predacious forms with raptorial forelegs (Grimaldi and Ross, covering the head, flattened posteriorly slanting coxae, and leathery 2004). forewings (tegmina). In Carboniferous coal swamps they were abun- In contrast to these lineages, which vanished from the scene in the dant and arguably the dominant group of insects (Grimaldi and Engel, late Mesozoic, “modern” Dictyoptera are a quite successful group with 2005). The phylogenetic relationships among these Palaeozoic almost 10,000 described species. Aside from the more or less general- “roachoids” are unresolved as yet (Legendre et al., 2015). However, it ized roaches (“Blattaria”), the group contains two highly specialized lin- is widely accepted that one of these lineages gave rise to Dictyoptera, eages: the wood- or fungus-feeding and eusocial (Isoptera), most likely in the Jurassic (Grimaldi and Engel, 2005: fig. 7.60; Misof today recognized as a subordinate group of (Klass and et al., 2014; Legendre et al., 2015; Tong et al., 2015). This clade com- Meier, 2006; Inward et al., 2007), and the predacious praying mantises prises the “modern” roaches (Blattodea) including termites (Isoptera) (Mantodea), which are characterized by specifically modified raptorial and the praying mantises (Mantodea) (Grimaldi, 2003; Grimaldi and forelegs and an elaborate prey catching mechanism. With about 3000 Engel, 2005; Legendre et al., 2015). Despite a general similarity, the and 2300 extant species (Ehrmann, 2002; Krishna et al., 2013), respec- tively, termites and praying mantises are slightly less species-rich than “ ” ⁎ Corresponding author. their generalized relatives, the paraphyletic roaches ( Blattaria ) E-mail address: [email protected] (R.G. Beutel). (Beutel et al., 2014).

http://dx.doi.org/10.1016/j.gr.2016.02.002 1342-937X/© 2016 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002 2 M. Bai et al. / Gondwana Research xxx (2016) xxx–xxx

Looking at extant species, Mantodea can be easily recognized by and Odonata were added as outgroup terminals(Kristensen, 1991; conspicuous modifications linked with the specialized predacious life Beutel and Gorb, 2006; Misof et al., 2014), without enforcing the mono- style. Even the few known early-diverging extinct forms (Grimaldi, phyly of the ingroup. Additionally, we included three extinct roaches 2003) differ distinctly from non-mantodean dictyopterans, including which have been considered as raptorial (†Ponopterix, †Jantaropterix those considered as predators (Grimaldi and Ross, 2004; Hörnig et al., and †Manipulator). As mantodean terminals we included four extant 2013; Lee, 2014; Vršanský and Bechly, 2015). In the present study we genera (Chaeteessa, Metallyticus, Hymenopus and Stagmomantis)and describe and document a Cretaceous fossil in detail using modern tech- all described fossils with at least some accessible characters niques. The morphological features are evaluated with respect to the (†Burmantis, †Jersimantis, †Santanmantis and †Ambermantis). Supple- phylogenetic position. Implications for the life style and habitats of the mentary Method S1 provides detailed information about the studied extinct species are discussed. An evolutionary scenario is developed taxa including the source of information. The 58 characters (from based on the phylogenetic conclusions, with a special focus on character head, mouthparts, pronotum, pterothorax, wings, legs, and transformations in the “roach–mantis transition zone”. postabdomen) are listed in Supplementary Method S2. To address the relationships of major lineages a few selected characters were included 2. Material and methods which cannot be scored for fossil terminals (e.g., internal parts of the exoskeleton, muscles, mode of insemination). We analyzed one matrix 2.1. Material and photography (Supplementary Method S3) with all taxa included and a second one (Supplementary Method S4) with fossil taxa with many inaccessible The single specimen, a male, was obtained from an amber deposit in characters excluded (fossil roaches and mantodean nymphs). The data the Hukawng Valley of Myanmar. The age has been estimated as were gathered in a matrix with WinClada and parsimony analyses ca. 99 Ma (98.8 ± 0.6; earliest ) based on U–Pb dating of were conducted with NONA (ratchet, 1000 replicates) and TNT (tradi- zircons from the volcaniclastic matrix of the amber (Shi et al., 2012). tional search; 99,999 random seeds, 1000 replicates) (Goloboff et al., The mining locality is at Noije Bum, near Tanai Village (26°21′33.41″N, 2003, 2008). Bremer support was calculated with TNT. Supplementary 96°43′11.88″E) (Grimaldi et al., 2002; Cruickshank and Ko, 2003). The Table S1 provides a list with all retrieved apomorphies for both analyses. deposit has been investigated and dated in detail by Cruickshank and An extensive discussion on the various relevant character systems and Ko (2003) and Shi et al. (2012). their phylogenetic evidence is provided in Supplementary Note S1. The piece of amber containing the specimen was ground and polished to a 23.5 × 14.5 × 12.5 mm cube (1.58 g) and examined with 3. Results a LEICA MZ 12.5 dissecting microscope with a drawing tube attachment. Photographs were taken using a Nikon D610 digital camera fitted to a 3.1. Systematic palaeontology Zeiss Stemi 2000-C stereomicroscope and processed in Helicon Focus 5.1 software and lastly Adobe Photoshop CS5 to deal with the images. Order †Alienoptera Bai, Beutel, Klass, Wipfler et Zhang ord. nov. By merging several photographs of one series, at different focal planes, http://zoobank.org/A4D93942-79D7-4B45-8E24-1E6CFFB4FB1A a single final photograph was created in which the entire sample was Family †Alienopteridae Bai, Beutel, Klass, Wipfler et Zhang fam. in focus. nov. http://zoobank.org/11ACE691-DCB4-4FC5-AB5B-EFEE5C3EB4BA 2.2. Micro-CT scanning and 3D reconstruction Type genus: †Alienopterus Bai, Beutel, Klass, Wipfler et Zhang gen. nov. The amber specimen was scanned with an X-radia 400 (Carl Zeiss †Alienopterus Bai, Beutel, Klass, Wipfler et Zhang gen. nov. X-ray Microscopy, Inc., Pleasanton, USA) at the Institute of Zoology, http://zoobank.org/04CAB2C7-3E25-4FD4-BC21-562330C7AEB1 Chinese Academy of Sciences. The scan of the entire (Fig. 2) Type species: †Alienopterus brachyelytrus Bai, Beutel, Klass, Wipfler was done with a beam strength of 40 kV, absorption contrast and a spa- et Zhang sp. nov. tial resolution of 15.4732 μm. The detail scans of the head (Fig. 4) and †Alienopterus brachyelytrus Bai, Beutel, Klass, Wipfler et Zhang the postabdomen (Fig. 5) were done with 60 kV, absorption contrast sp. nov. and a spatial resolution 4.1866 μm. Based on the obtained image stacks, http://zoobank.org/8D5503C4-2368-49D2-A950-926AA174641B structures of the specimen were reconstructed and separated with Amira 5.4 (Visage Imaging, San Diego, USA). The subsequent volume 3.1.1. Material rendering and animation were performed with VG Studiomax 2.1 (Vol- Currently only one nearly complete specimen is known (No. BU- ume Graphics, Heidelberg, Germany). For selected illustrations parts of 001057). This adult male is designated as the holotype of the new spe- the animal (e.g., the wings) were virtually removed. The same was done cies. The (type) specimen is currently on long-term loan in the Institute with air bubbles in the amber. For the reconstruction of the hindwing of Zoology, Chinese Academy of Sciences (IZAS) (specimen available for (Fig. 4E) the folded part of the wing was reconstructed and rendered study by contacting MB or WWZ). From 2026 it will be deposited in the separately. Both parts of the wing were combined using Photoshop currently established Three Gorges Entomological Museum, Chongqing, CS5 (Adobe, San Jose, USA). Final figures were prepared with Photoshop China. CS5 (Adobe, San Jose, USA) and Illustrator CS5 (Adobe, San Jose, USA). Resolution was not sufficient for showing setae or spines. 3.1.2. Etymology The genus name “Alienopterus” refers to the unusual (Latin 2.3. Taxon sampling, characters and phylogenetic analysis “alienus”) combination of characters including the wings (New Latin “-pterus” = -winged, from Greek “pterón” =wing)ofthenewspecies. The main aim of the cladistic analysis was to clarify the phylogenetic The species epithet refers to the short (Greek “brachys”)andapparently position of †Alienopterus. Based on the character combination wings hardened forewings (“elytron”, as generally used for the hardened folded back over the abdomen, well-developed cerci, and pad-like forewings of beetles). euplantulae, †Alienopterus brachyelytrus Bai, Beutel, Klass, Wipfler et Zhang gen. et sp. nov. can be assigned to Polyneoptera with reasonable 3.1.3. Diagnosis certainty. We thus selected a taxon sample containing extant represen- The description of a new species placed in a new taxon of higher tatives of all ten polyneopteran orders (ingroup) plus one member of rank requires the erection of a series of new taxa, in this case a new Psocodea and Hymenoptera; genera of Zygentoma, Ephemeroptera, genus, family, and order. In the ideal case each taxonomic level should

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002 M. Bai et al. / Gondwana Research xxx (2016) xxx–xxx 3 be associated with a diagnosis. However, in such a case it is impossible prementum and very large postmentum (about twice as wide as to determine which character is diagnostic at which rank. This depends prementum); palps three-segmented; endite lobes prominent (glossa on the character combination in the closest relatives of the new species, and paraglossa not clearly discernible as individual structures); which are unknown or may not exist. We avoid this procedural conflict postmentum with distinctly visible anterolateral expansions; elongate by providing a diagnosis currently valid for the entire series of newly posterior part comparatively narrow, tapering posteriorly, posterior erected taxa. Elements of this diagnosis can be appropriately associated edge reaching foramen occipitale. Posterior tentorial grooves not recog- with a particular taxon as soon as further species of the new order will nizable. Gula not developed, at least not recognizable as defined unit. be discovered. Thorax (Figs. 1–4): Paired lateral cervical sclerites and intercervical The new species (and the newly defined taxa of higher rank) can be sclerites (Wieland, 2006) present; median unpaired ventral cervical recognized by the following combination of characters: orthognathous sclerites not distinctly recognizable. Pronotum slightly longer than head with biting mouthparts, long filiform antennae, three ocelli, large wide; very deep transverse furrow (supracoxal furrow, Fig. 2A) sepa- and strongly convex compound eyes; pronotum with very deep trans- rates posterior quarter of pronotum from longer anterior part, corre- verse groove in posterior third, lateral parts bent downward to cover sponds with internal supracoxal ridge; anterior pronotal part strongly pleural elements; forewings short and apparently hardened, covering convex, constricted by U-shaped horizontal groove; posterior ends of anterior part of hind wings; hind wings well-developed, surpassing tip horizontal groove connected with supracoxal groove; central anterior of abdomen, folded longitudinally but not transversely, resting on ab- part of pronotum hemispherical in lateral view, round in dorsal dominal dorsum in repose; profemora with antero- and posteroventral view; lateral pronotal region strongly bent downward, almost vertical, row of densely placed setae; patch of setae inserted on distal anterior reaching beyond lateral base of coxa, largely covering pleural surface (profemoral brush); meso- and metafemora lacking comparable region; distinct lateral lamellar pronotal extensions absent. Meso- setal rows and patches; all tibiae with only two ventral terminal spines, thorax distinctly shortened; dorsally with triangular scutellar of similar size and shape on all legs; tarsi five-segmented, four proximal shield. Mesosternum well-developed, flat and plate-like; paired trian- tarsomeres each with paired pad-like euplantulae; large pretarsal gular fold-like structures present anterior to mesocoxae (possibly the arolium present between claws; cerci well-developed, multisegmented; trochantin), mesofurcal origin probably indicated by median invagina- male subgenital plate undivided, bearing small styli; male genitalia tion between them. Only small part of mesopleura exposed, partly cov- asymmetric (mantodean type, see below). Differing from “roachoids” ered by pronotum and partly by forewing; mesopleural ridge visible. (stemgroup Dictyoptera) and Blattodea (including termites [Isoptera]) Metathorax much larger than pro- and mesothorax; large triangular by the presence of a profemoral brush and interantennal and parietal metascutellum recognizable dorsally. Metapleuron largely exposed; ridges. Differing from Blattodea by the presence of three ocelli metapleural ridge very distinct, originates from posterolateral (plesiomorphy) instead of two or zero, and from Mantodea by the sim- metacoxal margin. Individual metasternal elements not recognizable. ilar terminal spines on all tibiae (plesiomorphy) instead of a distinctly Legs moderately long, length slightly increasing from prothorax to enlarged anterior terminal protibial spine, and by protibia lacking a dor- metathorax. Pro- and mesocoxae quite elongate, conical, slightly sal curvature (see character 22 in Supplementary Method S2). The new narrowing from base to apex; distinct concavity present on lateral side taxon differs from both dictyopteran orders by the combination of short of distal part, most distinctly on procoxae; metacoxae enlarged, triangu- sclerotized forewings and long hind wings, and by the deep posteriorly lar, with basal part extending far laterad; elevated mesal part forms located transverse pronotal groove. even platform; straight mesal metacoxal edges adjacent; distinct later- Description (adult male) (Figs. 1–5). Slender, body length 14.5 mm ally directed process present at laterodistal end of metacoxal platform. (anterior face of head to tip of tergite X); maximum width 3.2 mm (at Pro-, meso- and metatrochanters small, roughly triangular. Femora larg- forewing bases, the “shoulder” region). Entire cuticular surface glossy, er than other leg elements, about twice as long as coxae, very slightly ta- with well-developed vestiture of long, erect and semi-erect, yellowish pering distally; diameter of profemur not enlarged compared to meso- setae. and metafemora; each profemur with distinct anterior and posterior Head (Fig. 4): Orthognathous head wider than pronotum, nearly as longitudinal rows of setae on ventral surface; each row formed by nu- wide as mesothoracic shoulder region; distinctly triangular, widest at merous (more than 45) densely spaced setae; proximal setae longer dorsolateral ocular area; posterior cephalic region completely exposed, than distal ones; a dense group of setae (profemoral brush) located on strongly constricted in postoccipital area; narrow neck suggests high distal half of anterior surface; meso- and metafemora (Supplementary movability of head; posterior side of head capsule concave; distinct Fig. S1) lack ventral rows of setae (or setae minute) and femoral ridge originates near ventral margin, ascends dorsad (occipital ridge; brush; genicular spine on anterodorsal femoral apex lacking on all part of parietal ridge sensu Wipfler et al., 2012), then bends anterad, ex- legs. All tibiae very similar, more slender than femora, about equally tends further longitudinally (parietal ridge s.str.), and obliterates be- long, slightly widening distally; protibia not curved, i.e. dorsal edge tween compound eyes; indistinct epistomal ridge recognizable on straight, not convex; tibiae with numerous setae on all sides, appar- frontal side; antennal sockets mesally connected by interantennal ently longer on mesal surface; larger setae or spines arranged in two ridge; epicranial sutures not visible. Compound eyes strongly convex ventral rows absent; spines on dorsal side of tibiae also missing; two (hemispherical), with large near-circular base and small ommatidia; ventral terminal spines (Tt1, Tt5; Supplementary Fig. S1) present three large ocelli present anterodorsally, arranged in an almost straight distally on all tibiae, all of similar size and shape, not placed on tibial transverse line between compound eyes, with median ocellus placed projection, i.e. protibia lacking apical spur. Tarsus five-segmented; slightly more anterad. Antennae inserted below lateral ocelli, very tarsomeres one to four of all legs with two pad-like euplantulae; long and slender, filiform, multi-segmented; flagellum without region- size decreasing from one to four; the small tarsomere four distinctly ally differentiated flagellomeres; scapus distinctly longer and slightly protruding; elongate and slender apical tarsomere five strongly bent wider than other antennomeres; pedicellus slightly longer than upward. Pretarsus with paired claws equal, flanking a large pan- flagellomeres, narrowest near mid-length; transverse subdivision of shaped arolium. flagellomeres (Hockman et al., 2009) not recognizable. Labrum well- Two pairs of wings present, folded back on dorsum of pterothorax developed, free; clypeolabral suture recognizable, quite indistinct. and abdomen in repose. Forewings covering anterior parts of Mouthparts orthopteroid. Mandibles apparently unmodified but mesal hindwings; highly modified, forming scale-like shortened pseudoelytra parts not visible (enclosed between labrum and maxillae). Maxillae ar- with leathery cuticular structure; shoulder region strongly pronounced, ticulate with cardines laterad of anterior postmental region; stipes elon- part of fore wing mesad of shoulder horizontal, lateral part nearly verti- gate, bearing fairly long five-segmented palps; distal palpomere slightly cal; veins not recognizable in forewing; callous stigma possibly present widened distally; endite lobes not recognizable. Labium composed of but very indistinct; epipleura not present as delimited region; strongly

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002 4 M. Bai et al. / Gondwana Research xxx (2016) xxx–xxx

Fig. 1. Photographs of †Alienopterus brachyelytrus Bai, Beutel, Klass, Wipfler et Zhang gen. et sp. nov., holotype specimen (No. BU-001,057). (A) Lateral view of specimen. (B) Dorsal view of specimen. (C) Complete view of the amber cube with scale. Scale bars: 1 mm (pink). diverging mesal edges of the pseudoelytra fit with scutellar shield; dis- broadly rounded apex; dorsal process (paa) slender, curved dorsad; tinct mesal interlocking device lacking; articulation of forewings not bases of processes pda and paa separated by transverse cleft (en- recognizable. Hindwings (Fig. 4E) fully developed, wrapped around trance of a pouch, lve); interpretation of other elements of left side upper part of abdomen, folded longitudinally but not transversely; (L1 in Fig. 5A) ambiguous, posterior part apparently wrapped in posterior edge slightly surpasses abdominal apex; wing articulation membranous foldings, anterior part concealed by tergite X; L1 (as not clearly recognizable; main longitudinal veins partly visible, espe- suggested by visible part) probably developed as narrow sclerite, cially in proximal region (visible parts conforming with the typical possibly flexed ventrad along edge-like right extremity. Right part dictyopteran pattern (Smart, 1951, 1956)). of genitalia mainly represented by a large compact lobe (fda in Abdomen (Fig. 5): Abdomen composed of ten distinctly developed Fig. 5B–D); mesal part of fda (labeled fdm) projects slightly beyond segments and reduced segment XI with cerci. Sternite II fully developed remainder of lobe fda, vertically orientated, slightly bent towards right as anteriormost visible ventrite, sternite I at least strongly reduced side; lobe fda probably giving rise to a ventral lobe or tooth (pia) (as (or absent). Cerci distinctly developed, conical, about as long as last vis- suggested by ventral view) projecting to left side (i.e. lobes fda and ible sternite (IX); composed of large basal cercomere, three distinctly pia diverge to left side like two halves of a slightly opened book; shorter intermediate cercomeres, and elongate and slender distal white arrows in Fig. 5B indicate free posterior edge of lobe pia); dorsal cercomere. Subanal lobes bearing paraproct sclerites (PP-l, PP-r in and lateral walls of lobe fda probably continuously sclerotized (sclerite Fig. 5C) large, slightly asymmetric (or distorted); supraanal lobe not vis- R1); sclerite R1 apparently forming a mesally directed arm dorsally ible. Sternite IX forms posteriorly expanded subgenital plate, apically (white arrow in Fig. 5D); lateroventrally R1 apparently broadly bearing paired, short styli positioned close to midline; apparently continuous with sclerotization of lobe pia. See character discussion in uniformly sclerotized (no subdivision recognizable). Supplementary Note S1 for more details on phallic organs. Male genitalia (phallic organs) originate from large cavity bor- dered by sternite IX ventrally and subanal lobes and cercal bases dor- 3.2. Phylogenetic analyses sally; only apical projecting parts of genitalia visible; basal parts enclosed in genital cavity or embedded in debris in interspaces of In the first analysis with 58 morphological characters and 31 cavity; internalized parts (e.g., apodemes) not accessible; phallic terminals (Supplementary Methods S1 and S2; matrix provided in organs strongly asymmetric. Left part with two apical processes: Supplementary Method S3) the strict consensus tree of 25 equally par- ventral one (pda) short, wide, directed towards right side, with simonious trees (98 steps) (Supplementary Fig. S2) was poorly

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002 M. Bai et al. / Gondwana Research xxx (2016) xxx–xxx 5

Fig. 2. 3D reconstruction based on a μ-CT scan of †Alienopterus brachyelytrus Bai, Beutel, Klass, Wipfler et Zhang gen. et sp. nov. (A) Lateral view; left foreleg partly missing due to scan limitations. (B) Dorsal view. (C) Ventral view, with larger parts of legs removed. Abbreviations: an = antenna, ar = arolium, ce = compound eye, cer = cercus, CS9 = abdominal coxosternum IX, cx1 = procoxa, cx2 = mesocoxa, cx3 = metacoxa, fe = femur, fw = forewing, hw = hind wing, lp = labial palpus, mp = maxillary palpus, mss = mesoscutellum, mts = metascutellum, pn = pronotum, sf = supracoxal furrow, ti = tibia. Scale bars: 1 mm (red).

resolved, especially within Polyneoptera. This was obviously due to the The combination of short sclerotized forewings and fully developed, op- included fossil roach taxa (†Manipulator, †Ponopterix, †Jantaropterix). As erational hindwings resembles the condition found for instance in ear- many characters could not be scored for them, they switched between wigs (Dermaptera; Haas, 2003), staphylinid beetles (Thayer, 2005)and different positions in the tree, thereby blurring the phylogenetic results. some tettigoniids (Orthoptera; Bailey, 1979). The large, pan-shaped In the second analysis we excluded the three fossil roaches and two fos- arolium (an attachment device) and the upward lift of the ultimate sil mantodean nymphs (†Jersimantis, †Burmantis) with important inac- tarsomeres are typical for heelwalkers (Mantophasmatodea; Beutel cessible characters (matrix provided in Supplementary Method S4). and Gorb, 2008). The saddle-shaped pronotum is similar to conditions This analysis including 26 terminals yielded only three most parsimoni- found in grasshoppers and relatives (Orthoptera; Snodgrass, 1908). ous trees (96 steps) (Fig. 6) (see Supplementary Table S1 for potential Other features place †Alienopterus in a lineage comprising roaches, autapomorphies for all nodes in both analyses). termites and praying mantises (Dictyoptera), as the sister group of the The second analysis supported Metapterygota (=Odonata + order Mantodea (Fig. 6; see Supplementary Note S1 for detailed Neoptera), Neoptera, and tentatively Polyneoptera. All insect orders discussions). with more than one representative included were obtained as mono- Unfortunately, most derived (apomorphic) features characterizing phyletic. †Alienopterus was unambiguously placed as sister group of Dictyoptera are not visible in the single fossil specimen of †Alienopterus. Mantodea including extinct and extant terminals. Basal relationships This includes internal structures like the inner skeleton of the head in Mantodea could not be resolved. Only the “higher” mantodean taxa (tentorium, character 15 in Supplementary Method S2), soft tissue Stagmomantis and Hymenopus are combined in a clade. It should be (muscles; antennal vessels, character 56), female postabdomen noted that our dataset was neither composed to address deep insect re- (subgenital plate formed by coxosternite VII, character 48), characters lationships (Misof et al., 2014; Wipfler et al., 2015), nor the intraordinal related to reproduction (ootheca, character 45), and details of the phylogeny of Mantodea (Svenson and Whiting, 2004, 2009; Wieland, mouthparts (postmola on mandible, character 50; lacinia in cavity of 2013). The focus was on the placement of †Alienopterus. galea, character 55). However, a specific detail of the leg supporting an assignment to Dictyoptera is visible in †Alienopterus:aventralscler- 4. Discussion ite located between the margins of the tibia and the basal tarsomere (intertibiotarsal sclerite, character 52). As far as currently known, this †Alienopterus exhibits a seemingly weird combination of characters of sclerite occurs only in the dictyopteran orders. different groups of insects. This mosaic of structural features results in a Despite the incomplete accessibility of characters, a placement of very unusual appearance, which is reflected by the name “Alienoptera”. †Alienopterus in Dictyoptera is also clearly implied by derived features

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002 6 M. Bai et al. / Gondwana Research xxx (2016) xxx–xxx

Fig. 3. Morphological details of †Alienopterus brachyelytrus Bai, Beutel, Klass, Wipfler et Zhang gen. et sp. nov. (A) Forelegs. (B) Head. (C) Arolium. (D) Forewing. (E) Mesofemur and mesotibia. (F) Metafemur and metatibia. Abbreviations: ar = arolium, ce = compound eye, cla = pretarsal claw, dft = dorsal fold on distal margin of tibia, fe = femur, feb = femoral brush, Fva = anterior row of setae on ventral surface of profemur, Fvp = posterior row of setae on ventral surface of profemur, lp = labial palpus, mp = maxillary palpus, ta = tarsus, ti = tibia, Tt1 = tibial terminal spine 1, Tt5 = tibial terminal spine 5. Scale bars: 1 mm (pink); 0.5 mm (blue).

shared with Mantodea. The profemoral brush (character 21), a cleaning potential synapomorphy of Mantodea and †Alienopterus is a feature of device, was hitherto only known from mantises (including all fossils the legs: the ventral femoral rows of setae (or spines) are more strongly where this body part is visible). Several other derived features are developed on the forelegs (Fig. 3A) than on the middle and hind legs shared by both taxa, but are not unique to them: a notch and lobe at (Fig. 3E–F). Another character set shared by both taxa is likely linked the tibial apex (character 51), and interantennal (character 5) and pari- with raptorial habits, the triangular, very movable head with large, etal (character 2) head ridges on the head. The presence of a callous strongly protruding compound eyes. Similar conditions occur in extinct stigma (a thickened area of the forewing; character 34) would be an ad- predacious roaches (Vršanský, 2003; Hörnig et al., 2013; Lee, 2014; ditional argument, but the condition in †Alienopterus is ambiguous. The Vršanský and Bechly, 2015), possibly as a result of parallel evolution. strongest argument for †Alienopterus + Mantodea comes from the sim- Compared to all known extant and extinct mantodean species, ilar structure of the male genitalia. Male genital structures are highly di- †Alienopterus has retained a considerable number of plesiomorphies. vergent between insect orders and also often on lower taxonomic levels One of them is the presence of an arolium, an attachment structure oc- (e.g., Zoraptera (Dallai et al., 2014), Blattodea (Klass, 1997)). The visible curring in many groups of Neoptera (winged insects able to fold back projecting parts of †Alienopterus (ca. 30% of the genital apparatus) fully the wings) and arguably a groundplan apomorphy of this lineage conform with conditions found in Mantodea. This is a rare case where (Beutel and Gorb, 2001, 2006; character 27). The arolia are exceptional- the genital morphology of an extinct taxon can be reconstructed to an ly large in †Alienopterus, similar to the condition in Mantophasmatodea extent allowing a reliable phylogenetic interpretation. Another (character 28), but absent in all extant mantodeans, and also missing

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002 M. Bai et al. / Gondwana Research xxx (2016) xxx–xxx 7

Fig. 4. Details of †Alienopterus brachyelytrus Bai, Beutel, Klass, Wipfler et Zhang gen. et sp. nov. A–C + E: 3D reconstructions based on μ-CT data; D: line drawing. (A) Head in lateral view. (B) Head in posteroventral view. (C) and (D) Head in frontal view. (E) Unfolded hindwing. Abbreviations: A1 = anal vein, an = antenna, as = antennal socket, ce = compound eye, cl = clypeus, cs = cervical sclerite, CuA = anterior cubital vein, CuP = posterior cubital vein, er = epistomal ridge, fo = frontal ocellus, ias = interantennal ridge, lb. = labrum, lo = lateral ocellus, lp = labial palpus, M = medial vein, md = mandible, mp = maxillary palpus, pm = prementum, pn = pronotum, po = postmentum, pr = parietal ridge, RA = anterior radial vein, RP = posterior radial vein, ScP = posterior subcostal vein. Scale bars: 1 mm (pink).

(or perhaps vestigial) in Santanmantis (Grimaldi and Engel, 2005)and defining apomorphy of Mantodea. †Alienopterus does not have this spe- other extinct species of this order. The most conspicuous differences cialization (Fig. 3A). Thus we place it in a new order, †Alienoptera. This concern the forelegs, which are distinctly modified in Mantodea as is also in agreement with our cladistic analysis including the fossil specialized organs for catching prey: the thickness of the spines in the roaches (Supplementary Fig. S2), where †Alienopterus is not unambigu- ventral rows on the femur and tibia is increased (e.g., compared to ously placed with Mantodea. Blattodea), the anterior terminal tibial spine (Tt1) is distinctly enlarged Aside from preserved plesiomorphic features, †Alienopterus displays and placed on a projection, and the entire tibia is curved (at least in the an entire series of apomorphies unknown in Mantodea, including mod- ground plan of the group; characters 22, 23, 47). Due to these modifica- ifications of the forelegs: the setae of the ventral profemoral rows are tions, which are apparently completely missing in †Alienopterus,the delicate compared to the corresponding spines of Mantodea and most mantodean forelegs form an efficient grasping and holding device. We blattodean groups (thin setae also occur in corydiid cockroaches). Addi- follow Klass and Ehrmann (2003) and others (e.g., Wieland, 2013)by tionally, the ventral rows of protibial spines are missing in †Alienopterus, considering this specialized raptorial foreleg as a key innovation and in clear contrast to nearly all groups of Mantodea and Blattodea.

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002 8 M. Bai et al. / Gondwana Research xxx (2016) xxx–xxx

Fig. 5. 3D reconstruction of postabdomen of †Alienopterus brachyelytrus Bai, Beutel, Klass, Wipfler et Zhang gen. et sp. nov. (A) Dorsal view (arrows: margins of sclerite L4). (B) Ventral view (arrows: posterior edge of process pia). (C) Posterior view. (D) Lateral view (arrow: anteromesal extension of sclerite R1). Abbreviations: cer = cercus, CS9 = coxosternite of abdominal segment IX, PP-l, PP-r = left and right paraproctal sclerites (on subanal lobes), sl9 = stylus of abdominal segment IX, TG9, TG10 = tergites of abdominal segments IX and X. Genitalic sclerites (L1, L4, R1), projections (afa, fda, fdm, loa, paa, pda, pia, and vla), and pouches (lve) are labeled according to the terminology of Klass (1997). Scale bars: 1 mm (red).

Conspicuous apomorphies of †Alienopterus are the strongly shortened A flight apparatus with large hindwings and shortened forewings is forewings (in combination with fully developed hindwings), the unknown in extant Dictyoptera, where both fore- and hindwings are ei- distinctly enlarged arolium, elevated terminal tarsomeres, and the ther well developed or reduced to varying degrees. The large hindwings saddle-shaped pronotum. Even though these features occur in other in- suggest that †Alienopterus could fly well, while the forewings played a sect groups, suggesting possible phylogenetic affinities, the analyses minor role in this functional context if at all. †Alienopterus is a case of presented here show that they evolved independently in †Alienopterus functional dipterism, with flight mechanics likely differing considerably as autapomorphic conditions, contributing to a weird appearance of from other dictyopterans, but similar to conditions occurring in earwigs this extinct insect. †Alienopterus is embedded in 99 Ma old amber. The (Dermaptera) and beetles (Coleoptera). These groups fold their mem- age of the clade Mantodea is discussed controversially (Grimaldi, branous flight organs under the forewings for mechanical protection, 2003; Béthoux and Wieland, 2009; Svenson and Whiting, 2009; Misof with a derived transverse folding mechanism (Haas, 2006). This is obvi- et al., 2014; Legendre et al., 2015), but based on the fossil evidence, ously not the case in †Alienopterus, where the hindwings are largely ex- the minimum age is 130 million years. Consequently, the last common posed, and only folded longitudinally (as in most dictyopterans). ancestor of Mantodea and †Alienoptera must have lived at least However, the sclerotized forewings form a cavity containing the proxi- 30 million years earlier than †A. brachyelytrus. In this (minimum) time mal part of the hindwings. It is conceivable that the resulting protection this unique combination of characters and adaptations evolved in the of the wing joint is an advantage, as it is also the case in earwigs, beetles, lineage including the species described here. and some stick insects (Phasmatodea) with shortened leathery fore- All in all, the characters evaluated here and the results of our cla- wings (Bradler, 2003). Apparently this is a trade-off between easy and distic analysis place †Alienopterus as the sistergroup to Mantodea, rapid unfolding (without transverse folding mechanism) and full pro- with retained ancestral conditions, but also an array of specialized tection of the membranous flight organs in repose. Beetles and earwigs features. Both phylogenetically and in its structural features, are adapted to penetrate narrow crevices and usually dorsoventrally †Alienopterus is a connecting link in the narrow space between flattened (Beutel, 1997). This is apparently not the case in †Alienopterus “roach-like” ancestors and the successful praying mantises as we and Phasmatodea, which are characterized by a more or less cylindrical know them today. body, suggesting a more exposed lifestyle. The reconstruction of the lifestyle of extinct species is greatly The three pairs of legs of †Alienopterus are structurally very similar impeded by the complete lack of direct observations. However, the and were probably used for normal hexapod walking. However, a rather accessible morphological features of †Alienopterus allow several unspecialized raptorial function of the forelegs is conceivable, as long conclusions. and densely arranged ventral setae are present on the profemora but

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002 M. Bai et al. / Gondwana Research xxx (2016) xxx–xxx 9

The complex genitalia of Dictyoptera are bilaterally symmetric in fe- males (Klass, 1998) but strongly asymmetric in males (Klass, 1997). A particular architecture of complex male genitalia usually requires a de- fined way of how to connect them with the female organs. This usually correlates with a particular posture (or series of postures) for initiating copulation. In Mantodea the male climbs or jumps on the back of the fe- male and clings to the thorax with his forelegs. Then the male curves the postabdomen to the right (or to the left in the rare cases of mirror- inverted genitalia; see Supplementary Note S1) and pushes his phallic organs into the genital pouch of the female (Kumar, 1973). The great similarity of the male apparatus of †Alienopterus strongly suggests a similar pattern in this extinct taxon. Based on our structural observations, we conclude that †A. brachyelytrus was a predator with a preference for trees or bushes, with suitable foliage and open spaces. Similar environments are preferred by many recent species of Mantodea. However, early mantises probably hunted on bark, a lifestyle still observed in various Recent branches of the order (Wieland, 2010). The transition to an open three-dimensional environment such as trees, bushes or larger herbaceous plants occurred in the Tertiary, thus at least 35 million years after †Alienopterus lived (Grimaldi and Engel, 2005). †Alienoptera is thus an unsuccessful lineage of predators in the roach–mantis transi- tion zone.

Fig. 6. Result of the cladistic analysis. Strict consensus tree of three most parsimonious 5. Conclusions trees with 96 steps. Cladistic analysis based on 58 characters (SI Appendix, Method S2) for 26 taxa (SI Appendix, Method S1). Polyneoptera (excl. following groups) in dark † blue, Dictyoptera stem in light blue, Blattodea in green, †Alienoptera in orange, The detailed morphological investigation of the new species A. Mantodea in brown. Numbers on branches indicate Bremer support above 1. Potential brachyelytrus from Burmese amber and the phylogenetic analyses autapomorphies for all clades are provided in SI Appendix, Table S1. Insect (sub-)orders (including extant and extinct taxa) revealed that it belongs to are represented by icons that do not necessarily show the sampled species. Dictyoptera. The profemoral brush, male genitalia, and other features suggest that it is the sistergroup of the order Mantodea, despite the seemingly unorthodox combination of features (e.g., shortened tegmi- na, enlarged arolium). A placement within mantodeans (incl. the stemgroup) can be excluded. †Alienopterus was characterized by a spe- lacking on the middle and hindlegs. The delicate condition of the cialized efficient flight apparatus, attachment structures adapted for lo- profemoral vestiture and the absence of counteracting tibial spines comotion on leaves, and a dense profemoral setation suitable for (both arguably apomorphic conditions) suggest that prey-grasping catching small prey. These features also suggest that it lived in bushes was distinctly less powerful than in Mantodea. The forelegs were likely or trees. However, it failed to succeed in this environment, which was suitable for grasping small items from the substrate in front of the in- later successfully populated by the praying mantises with their distinct- sect, perhaps with movements similar to those of mantodean forelegs. ly modified raptorial forelegs. †Alienopterus was thus an evolutionary This may have included small and delicate arthropods such as for in- dead end in the roach–mantis-transition zone. stance midges or aphids. A raptorial lifestyle is also suggested by the †Alienoptera is the fourth newly discovered insect taxon of recent high movability of the head. Beyond this, the large, widely separated years assigned ordinal rank, following the Mantophasmatodea (with compound eyes and the presence of a profemoral brush, which like in extant and extinct taxa; Klass et al., 2002), †Nakridletia (extinct; extant Mantodea may have served as a cleaning device for the eyes, in- Vršanský et al., 2010)and†Coxoplecoptera (extinct; Staniczek et al., dicate that as in Mantodea the raptorial habits of †Alienopterus were 2011). However, an improved resolution of phylogenetic relationships mainly supported by vision. Regardless of the precise target and tech- among early (Palaeozoic) pterygote insects may require the erection nique of food acquisition, it is likely that an early stage of raptorial of further “new insect orders”. This will likely concern the multitude prey-catching had evolved in the common ancestor of †Alienopterus of groups assembled in “Grylloblattida” (see Cui et al., 2015), but also and Mantodea. the paraphyletic stem-Dictyoptera, the roachoids (see Grimaldi and The enlarged arolium of †Alienopterus is a unique feature in Engel, 2005: fig. 7.60), and possibly also other groups. Dictyoptera, with parallel evolution in heelwalkers (Mantophasmatodea) Supplementary data to this article can be found online at http://dx. and “basal” stick insects (Timematodea) (Beutel and Gorb, 2006, 2008). doi.org/10.1016/j.gr.2016.02.002. Timematodeans spend their lives on leaf surfaces and show cryptic be- haviour to a degree that certain species or morphs are adapted to leaves Acknowledgments of specificplantspecies(Nosil and Crespi, 2004). Additionally they rely on defensive glands. As an almost unique feature, heelwalkers can lift or We thank Sven Bradler (Georg-August-Universität Göttingen) for lower the terminal tarsomere, thus either keeping the arolium off the sub- his helpful comments concerning stick insects and all curators who strate (walking on the “heels”) or in touch with it, depending on the spe- hosted the first author or loaned or supplied material for this study. cific situation (Beutel and Gorb, 2006; Roth et al., 2014). The arolia are This research was supported by the National Basic Research Program only used for walking on smooth surfaces such as leaves, for striking of China (973 Program) (No. 2011CB302102), the National Natural prey, when additional grip is required (e.g., transportation of larger Science Foundation of China (Nos. 31172143, 31350110218 prey), or for walking in an upside down position (Roth et al., 2014). Like (B.W.)), the National Science Fund for Fostering Talents in Basic in Mantophasmatodea, the large arolium of †Alienopterus could be lifted Research (Special Subjects in Animal Taxonomy, NSFC-J1210002), with the last tarsomere. This suggests that it was used in a similar way Research Equipment Development Project of Chinese Academy of and probably also in similar habitats. Sciences (YZ201509), the Special Fiscal Funds of Shaanxi Province

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002 10 M. Bai et al. / Gondwana Research xxx (2016) xxx–xxx

(No. 2013-19), by a Humboldt Fellowship (M.B.) from Alexander von Klass, K.-D., Meier, R., 2006. A phylogenetic analysis of Dictyoptera (Insecta) based on morphological characters. Entomologische Abhandlungen 63, 3–50. Humboldt Foundation, by the BIG4 project (MSCA-ITN-2014-ETN: Klass, K.-D., Zompro, O., Kristensen, N.P., Adis, J., 2002. Mantophasmatodea: a new insect Marie Skłodowska-Curie Innovative Training Networks [ITN-ETN], order with extant members in the Afrotropics. Science 296, 1456–1459. 642241—BIG4—H2020-MSCA-ITN-2014), and by the Deutsche Krishna, K., Grimaldi, D.A., Krishna, V., Engel, M.S., 2013. Treatise on the Isoptera of the world. Bulletin of the American Museum of Natural History 377, 1–2704. Forschungsgemeinschaft DFG (KL1162/7-1; WI 4324/1-1). Kristensen, N.P., 1991. Phylogeny of extant hexapods. In: Naumann, J.D., Carne, P.B., Lawrence, J.F. (Eds.), Insects of Australia. Melbourne University Press, Melbourne, References pp. 125–140. Kumar, R., 1973. The biology of some Ghanaian mantids (Dictyoptera: Mantodea). Bulle- Bailey, W.J., 1979. A review of Australian Copiphorini (Orthoptera: Tettigoniidae: tin de Institut Fondamental d'Afrique Noire 35, 551–578. Conocephalinae). Australian Journal of Zoology 27, 1015–1049. Labandeira, C.C., 1994. A compendium of fossil insect families. Milwaukee Public Bell, W.J., Roth, L.M., Nalepa, C.A., 2007. Cockroaches: Ecology, Behavior, and Natural Museum. Contributions in Biology and Geology 88, 1–71. History. JHU Press. Lee, S.W., 2014. New lower cretaceous basal mantodean (Insecta) from the Crato Forma- Béthoux, O., Wieland, F., 2009. Evidence for carboniferous origin of the order Mantodea tion (NE Brazil). Geologica Carpathica 65, 285–292. (Insecta: Dictyoptera) gained from forewing morphology. Zoological Journal of the Legendre, F., Nel, A., Svenson, G.J., Robillard, T., Pellens, R., Grandcolas, P., 2015. Phylogeny Linnean Society 156, 79–113. of Dictyoptera: dating the origin of cockroaches, praying mantises and termites with Beutel, R.G., 1997. Über Phylogenese und Evolution der Coleoptera (Insecta), molecular data and controlled fossil evidence. PloS One 10, e0130127. insbesondere der Adephaga. Verhandlungen des Naturwissenschaftlichen Liang, J.H., Vršanský, P., Ren, D., Shih, C.K., 2009. A new Jurassic carnivorous Vereins in Hamburg 31, 1–64. (Insecta, Blattaria, Raphidiomimidae) from the Inner Mongolia in China. Zootaxa Beutel, R.G., Gorb, S., 2001. Ultrastructure of attachment specializations of hexapods 1974, 17–30. (Arthropoda): evolutionary patterns inferred from a revised ordinal phylogeny. Misof, B., Liu, S., Meusemann, K., Peters, R.S., Donath, A., Mayer, C., Frandsen, P.B., Ware, J., Journal of Zoological Systematics and Evolutionary Research 39, 177–207. Flouri, T., Beutel, R.G., et al., 2014. Phylogenomics resolves the timing and pattern of Beutel, R.G., Gorb, S., 2006. A revised interpretation of the evolution of attachment insect evolution. Science 346, 763–767. structures in Hexapoda with special emphasis on Mantophasmatodea. Nel, A., Prokop, J., Grandcolas, P., Garrouste, R., Lapeyrie, J., Legendre, F., Anisyutkin, L.N., Systematics & Phylogeny 61, 3–35. Kirejtshuk, A.G., 2014. The beetle-like Palaeozoic and Mesozoic roachoids of the so- Beutel, R.G., Gorb, S., 2008. Evolutionary scenarios for unusual attachment devices called “umenocoleoid” lineage (Dictyoptera: Ponopterixidae fam. nov.). Comptes of Phasmatodea and Mantophasmatodea (Insecta). Systematic Entomology 33, Rendus Palevol 13, 545–554. 501–510. Nosil, P., Crespi, B.J., 2004. Does gene flow constrain adaptive divergence or vice versa? A Beutel, R.G., Friedrich, F., Ge, S.Q., Yang, X.K., 2014. Insect Morphology and Phylogeny. De test using ecomorphology and sexual isolation in Timema cristinae walking-sticks. Gruyter, Berlin, New York. Evolution 58, 102–112. Bradler, S., 2003. 16. Phasmatodea, Gespenstschrecken. In: Dathe, H. (Ed.), Lehrbuch der Roth, S., Molina, J., Predel, R., 2014. Biodiversity, ecology, and behavior of the recently dis- Speziellen Zoologie, Band 1: Wirbellose Tiere, 5. Teil: Insecta, second ed. Spektrum covered insect order Mantophasmatodea. Frontiers in Zoology 11, 70. Akademischer Verlag, Berlin, Heidelberg, pp. 251–261. Shi, G.H., Grimaldi, D.A., Harlow, G.E., Wang, J., Wang, J., Yang, M.C., Lei, W.Y., Li, Q.L., Li, Cruickshank, R.D., Ko, K., 2003. Geology of an amber locality in the Hukawng Valley, X.H., 2012. Age constraint on Burmese amber based on U–Pb dating of zircons. Creta- northern Myanmar. Journal of Asian Earth Sciences 21, 441–455. ceous Research 37, 155–163. Cui, Y.Y., Béthoux, O., Klass, K.-D., Ren, D., 2015. The Jurassic Bajanzhargalanidae (Insecta: Smart, J., 1951. The wing-venation of the American cockroach Periplaneta americana Linn Grylloblattida?): new genera and species, and data on postabdominal morphology. (Insecta — blattidae). Proceedings of the Zoological Society of London 121, 501–509. Arthropod Structure & Development http://dx.doi.org/10.1016/j.asd.2015.04.008. Smart, J., 1956. On the wing-venation of Chaeteessa and other mantids (Insecta: Dallai, R., Gottardo, M., Mercati, D., Machida, R., Mashimo, Y., Matsumura, Y., Beutel, R.G., Mantodea). Proceedings of the Zoological Society of London 127, 545–553. 2014. Giant spermatozoa and a huge spermatheca: a case of coevolution of male and Snodgrass, R.E., 1908. A comparative study of the thorax in Orthoptera, Euplexoptera, and female reproductive organs in the ground louse Zorotypus impolitus (Insecta, Coleoptera. Proceedings of the Entomological Society of Washington 9, 95–108. Zoraptera). Arthropod Structure & Development 43, 135–151. Staniczek, A.H., Bechly, G., Godunko, R.J., 2011. Coxoplectoptera, a new fossil order of Ehrmann, R., 2002. Mantodea. Gottesanbeterinnen der Welt. Natur und Tier - Verlag, Palaeoptera (Arthropoda: Insecta), with comments on the phylogeny of the stem Münster. group of mayflies (Ephemeroptera). Insect Systematics & Evolution 42, 101–138. Goloboff, P.A., Farris, J., Nixon, K., 2003. Tree Analysis Using New Technology [Internet]. Svenson, G.J., Whiting, M.F., 2004. Phylogeny of Mantodea based on molecular data: evo- Program and Documentation. Distributed by the Authors. Available from: http:// lution of a charismatic predator. Systematic Entomology 29, 359–370. www.zmuc.dk/public/phylogeny. Svenson, G.J., Whiting, M.F., 2009. Reconstructing the origins of praying mantises Goloboff, P.A., Farris, J., Nixon, K., 2008. TNT, a free program for phylogenetic analysis. (Dictyoptera, Mantodea): the roles of Gondwanan vicariance and morphological con- Cladistics 24, 774–786. vergence. Cladistics 25, 468–514. Grimaldi, D.A., 2003. A revision of Cretaceous mantises and their relationships, including Thayer, M.K., 2005. 11. Staphylinoidea. 11.7. Staphylinidae Latreille, 1802. In: Beutel, R.G., new taxa (Insecta: Dictyoptera: Mantodea). American Museum Novitates 3412, 1–47. Leschen, R.A.B. (Eds.), Coleoptera. Morphology and Systematics (Archostemata, Grimaldi, D.A., Engel, M.S., 2005. Evolution of the Insects. Cambridge University Press, Adephaga, Myxophaga, Polyphaga partim). De Gruyter, Berlin, New York, New York. pp. 296–344. Grimaldi, D.A., Ross, A.J., 2004. Raphidiomimula, an enigmatic new cockroach in Creta- Tong, K.J., Duchêne, S., Ho, S.Y., Lo, N., 2015. Comment on “Phylogenomics resolves the ceous amber from Myanmar (Burma) (Insecta: Blattodea: Raphidiomimidae). Journal timing and pattern of insect evolution”. Science 349, 487. of Systematic Palaeontology 2, 101–104. Vršanský, P., 2003. Umenocoleoidea — an amazing lineage of aberrant insects (Insecta, Grimaldi, D.A., Engel, M.S., Nascimbene, P.C., 2002. Fossiliferous Cretaceous amber from Blattaria). AMBA Projekty 7, 1–32. Myanmar (Burma): its rediscovery, biotic diversity, and paleontological significance. Vršanský, P., Bechly, G., 2015. New predatory cockroaches (Insecta: Blattaria: American Museum Novitates 3361, 1–71. Manipulatoridae fam.n.) from the Upper Cretaceous Myanmar amber. Geologica Haas, F., 2003. 12. Ordnung Dermaptera. In: Dathe, H. (Ed.), Lehrbuch der Speziellen Carpathica 66, 133–138. Zoologie, Band 1: Wirbellose Tiere, 5. Teil: Insecta, second ed. Spektrum Vršanský, P., Ren, D., Shih, C.K., 2010. Nakridletia ord. n. — enigmatic insect parasites sup- Akademischer Verlag, Berlin, Heidelberg, pp. 173–180. port sociality and endothermy of pterosaurs. AMBA Projekty 8, 1–16. Haas, F., 2006. Evidence from folding and functional lines of wings on inter-ordinal Wieland, F., 2006. The cervical sclerites of Mantodea discussed in the context of relationships in Pterygota. Arthropod Systematics & Phylogeny 64, 149–158. dictyopteran phylogeny (Insecta: Dictyoptera). Entomologische Abhandlungen 63, Hockman, D., Picker, M.D., Klass, K.-D., Pretorius, L., 2009. Postembryonic development of 51–76. the unique antenna of Mantophasmatodea (Insecta). Arthropod Structure & Develop- Wieland, F., 2010. The Phylogenetic System of Mantodea (Insecta: Dictyoptera). Unpub- ment 38, 125–133. lished thesis, Georg-August-Universität Göttingen, Germany Hörnig, M.K., Haug, J.T., Haug, C., 2013. New details of Santanmantis axelrodi and the Wieland, F., 2013. The phylogenetic system of Mantodea (Insecta: Dictyoptera). Species, evolution of the mantodean morphotype. Palaeodiversity 6, 157–168. Phylogeny and Evolution 3, 3–222. Inward, D., Beccaloni, G., Eggleton, P., 2007. Death of an order: a comprehensive molecu- Wipfler, B., Wieland, F., DeCarlo, F., Hoernschemeyer, T., 2012. Cephalic morphology of lar phylogenetic study confirms that termites are eusocial cockroaches. Biology Hymenopus coronatus (Insecta: Mantodea) and its phylogenetic implications. Arthro- Letters 3, 331–335. pod Structure & Development 41, 87–100. Klass, K.-D., 1997. The external male genitalia and the phylogeny of Blattaria and Wipfler, B., Klug, R., Ge, S.Q., Bai, M., Göbbels, J., Yang, X.K., Hörnschemeyer, T., 2015. The Mantodea. Bonner Zoologische Monographien 42, 1–341. thorax of Mantophasmatodea, the morphology of flightlessness, and the evolution of Klass, K.-D., 1998. The ovipositor of Dictyoptera (Insecta): homology and ground-plan of the neopteran insects. Cladistics 31, 50–70. the main elements. Zoologischer Anzeiger 236, 69–101. Klass, K.-D., Ehrmann, R., 2003. 13. Ordnung Mantodea. In: Dathe, H. (Ed.), Lehrbuch der Speziellen Zoologie, Band 1: Wirbellose Tiere, 5. Teil: Insecta, second ed. Spektrum Akademischer Verlag, Berlin, Heidelberg, pp. 182–197.

Please cite this article as: Bai, M., et al., †Alienoptera — A new insect order in the roach–mantodean twilight zone, Gondwana Research (2016), http://dx.doi.org/10.1016/j.gr.2016.02.002

View publication stats