Morphology and function of the ovipositor mechanism in Ceraphronoidea (Hymenoptera, Apocrita)

The ovipositor of apocritan Hymenoptera is an invaluable source of phylogenetically relevant characters, and our understanding of its functional morphology stands to enlighten us about parasitoid life history strategies. Although Ceraphronoidea is one of the most commonly collected Hymenoptera taxa with considerable economic importance, our knowledge about their natural history and phylogenetic relationships, both to other apocritan lineages and within the superfamily itself, is limited. As a first step towards revealing ceraphronoid natural diversity we describe the skeletomuscular system of the ceraphronoid ovipositor for the first time. Dissections and Confocal Laser Scanning Microscopy 3D media files were used to visualize the ovipositor complex and to develop character concepts. Morphological structures were described in natural language and then translated into a character-character state format, whose terminology was linked to phenotype-relevant ontologies. Four unique anatomical phenotypes were revealed: 1. The first valvifer (gonangulum) of the genus Trassedia is composed of two articulating sclerites, a condition present only in a few basal insect taxa. The bipartition of the first valvifer in Trassedia is most likely secondary and might allow more rapid oviposition. 2. Ceraphronoids, unlike other Hymenoptera, lack the retractor muscle of the terebra; instead the egg laying device is retracted by the seventh sternite. 3. Also unlike other Hymenoptera, the cordate apodeme and the anterior flange of the second valvifer are fused and compose one ridge that serves as the site of attachment for the dorsal and ventral T9-second valvifer muscles. Overall, the ceraphronoid ovipositor system is highly variable and can be described by discrete, distinguishable character states. However, these differences, despite their discrete nature, do not reflect the present classification of the superfamily and might represent parallelisms driven by host biology. JHR 33: 25–61 (2013) doi: 10.3897/JHR.33.5204 www.pensoft.net/journals/jhr Copyright Andrew F. Ernst et al. This is an open access article distributed under the terms of the Creative Commons Attribution License 3.0 (CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ReseARCH ARtiCle Andrew F. Ernst et al. / Journal of Hymenoptera Research 33: 25–61 (2013) 26

The phylogenetic placement of Ceraphronoidea remains uncertain despite the recent efforts to resolve the Hymenoptera tree of life (Heraty et al. 2011, Sharkey et al. 2011).Using expressed sequence tag (EST) data and limited exemplars, Sharanowski et al. (2010) placed Ceraphronoidea as sister to Evanioidea, but with low support.Heraty et al. (2011), using data from four genetic loci, placed Ceraphronoidea as sister to Stephanidae or Stephanidae + Orussidae, depending on the analysis.Vilhelmsen et al. (2010) found Ceraphronoidea to be sister to Megalyroidea within Evaniomorpha, based on morphological data.Sharkey et al. (2011) using molecular data and morphological data -based largely on Vilhelmsen et al. (2010) -also suggested a sister relationship between Ceraphronoidea and Megalyroidea.
The most comprehensive review on the classification of the superfamily has proposed by Dessart and Cancemi (1987).Although this classification served as a template for further studies (Masner 1993, Dessart 1995a, b, Mikó and Deans 2009) the phylogenetic base has never been challenged.The utility of traditionally used morphological characters in Ceraphronoidea classification seems to be limited (Dessart 1995b(Dessart , c, 2001) ) and with the discovery of the first ceraphronid wasp with well-developed pterostigma even the family level classification of the superfamily has been challenged recently (Mikó and Deans 2009).Our recent observations on the genus Trassedia provide additional evidence that the current classification of the superfamily needs to be revised (Mikó et al. in press).The genus was erected by Cancemi (1996), who classified the type species within Megaspilinae and considered it to be closely related to Conostigmus.The original classification is supported by two distinct character states: the presence of a fore wing pterostigma (which has subsequently been found in Ceraphronidae; Mikó and Deans 2009) and the presence of 11 antennomeres (Ceraphronidae have 10).Our observations of multiple specimens of Trassedia luapi reveal numerous morphological character states that are specific to Ceraphronidae (Mikó and Deans 2009, Mikó et al. in press): absence of mesotibial apical spur, absence of narrow sclerite anterior to synsternum, presence of Waterston's evaporatorium, and numerous male genitalia characters.Based on these differences the genus has recently been transferred to Ceraphronidae (Mikó et al. in press), a classification we follow in this study.
It is evident that a comprehensive study using both morphological and molecular characters is necessary for the reevaluation of Ceraphronoidea systematics.Since traditionally used morphological characters have been phylogenetically inconsistent the utilization of unexplored character systems, such as the male and female terminalia might offer additional data relevant to more robustly estimate the phylogenetic history of this group.
Despite the extensive descriptive work done on comparative morphology of the Hymenoptera female terminalia (Oeser 1961, Smith 1970, Le Ralec et al. 1996, Quicke et al. 1992, Quicke et al. 1994, Quicke et al. 1999, Vilhelmsen 2000, Vilhelmsen et al. 2001), the available morphological data on the ceraphronoid female terminalia are restricted to the distal region of the terebra (Quicke et al. 1994, Le Ralec et al. 1996) and the accessory glands (Höller et al. 1993).The skeletomuscular system of ceraphronoid ovipositor remained, until now, relatively unexplored.The main aim of the present study is to describe morphological diversity of the female terminalia in Ceraphronoidea and compare its anatomical structures with that of other Hymenoptera taxa giving special emphasize on the skeletomuscular system.With this study, we establish a baseline for further phylogenetic analyses of the superfamily using ovipositor characters.

Taxon
Specimen identifier

Imaging techniques
Locality DOI of CLSM media files stored at http://figshare.com Specimens used in the present study were stored in 95% ethanol.Some specimens were critical point dried and dissected on Blue-Tack (Blue Tack, Bostik inc.) medium.This method is mostly used to reveal the spatial relationships between muscles.Other specimens were dissected in glycerin on a concave microscope slide or were macerated in KOH to visualize the skeletal structures.Dissections and observations were made using an Olympus SZX16 stereomicroscope and an Olympus CX41 compound microscope.
Bright field images were taken using an Olympus CX41 compound microscope, equipped with an Olympus DP71 digital camera.Image stacks were combined using CombineZP (Hadley 2010) "do stack" command.SEM micrographs were made using a Hitachi S-3200 Scanning Electron Microscope (wd=23.5, av=5kV).Specimens were critical point dried and coated with palladium prior to examination.CLSM images were made on glycerin-stored specimens between 1.5 mm thick, 24×50 mm cover glasses with a Leica LSM 710 and Olympus Fluoview 1000 confocal laser scanning microscopes (CLSM) using the 488 nm laser to excite the sample.We collected the autofluorescence of insect anatomical structures between 500 and 700 nm with two channels (500-580; pseudocolor green and 580-700 pseudocolor red) using 106 and 206 Plan Achromat objectives.Volume rendered images and media files were generated by Imaris Bitplane (Bitplane, Zürich, Switzerland) and ImageJ (Schneider et al. 2012) software.Bright field and CLSM micrographs were edited with Adobe Photoshop CS4 (Adobe) changing "Gamma Correction" value, resize images to 7.3 cm width at 400 dpi resolution, standardize style and width of scale bars and create image annotations.Media files, SEM micrographs and bright field images are available from figshare.com(Table 1).
To verify the relationships between anatomical structures, serial transverse sectioning was carried out on a Lagynodes specimen (Table 1).The specimen was embedded in Araldit®, cut at 1 µm with a Microm microtome (HM 360), and stained with toluidine blue.

Towards semantic statements
Verbatim descriptions composed in natural language serve as the traditional way for communicating observations in insect morphology.However, these descriptions can be decoded only by morphology experts, are hardly accessible to non-expert researchers and cannot be reasoned over efficiently by text-mining applications (Vogt et al. 2010, Deans et al. 2012a, b, Vogt et al. 2013).Morphological descriptions, albeit they are arguably more complex, shares numerous similarities with taxonomic descriptions which were the objects of recent efforts for altering the way of biodiversity descriptions into a more accessible format.Deans et al. (2012a) proposed a new description model for taxonomists applying semantic statements.These statements are written in a logic and queryable format and are linked to the concepts of biomedical ontologies.Semantic descriptions are therefore not only transparent for researchers unfamiliar with specific morphological jargon, but can be executed via an automated reasoning mechanism (Balhoff et al. in press, Mullins et al. 2012).
To meet the grand challenge of describing phenotypes in a semantic way, using Web Ontology Language (OWL; http://www.w3.org/TR/owl-features/) for example, one must be familiar with tools of the Semantic Web (e.g., Protégé, http://protege.stanford.edu/and Manchester Syntax, http://www.w3.org/TR/owl2-manchester-syntax/).Perhaps more importantly, one also has to provide a character/character state description with terms explicitly linked to ontologies.We provide here an example of the transformation of our natural language descriptions to character/character state format and to link the terminology to relevant phenotype ontologies.Our goal is to make this product more accessible to future reasoning applications.During the "ontologization" procedure the describer is forced to provide strict, structure-based definitions for each anatomical concept, which itself enhances the readability, objectivity, consistency and comparability of the research product.
Results ii: semantically enhanced characters and character states for ceraphronoid female terminalia.

Discussion
The enormous diversity of ovipositor phenotype in Hymenoptera reflects the manner in which the female wasp finds feasible environments for her developing larvae.Host structure and location, as well as the different ways of storing the ovipositor, are arguably the principal factors driving the structural adaptation of these structures (Quicke et al. 1999, Vilhelmsen 2000, Vilhelmsen and Turrisi 2011).Major morphological characteristics of ceraphronoid ovipositors are most likely related to these influences.

Storage of terebra
The ceraphronoid ovipositor is stored in a horizontal position inside the metasoma, with its ventral part concealed by S7 (Fig. 6E).As the first oviposition movement, the contracting muscles between the apical metasomal tergites and sternites expose the ventral part of the ovipositor by rotating it and the ninth abdominal tergite posteriorly, from the resting, horizontal to the active, vertical position.This movement is common within Apocrita (Fig. 6F; Alam 1953, Copland 1976, Fergusson 1988).
The mechanism of the extension of the terebra from the second valvifer-third valvula complex is shared between Ceraphronoidea and most other Hymenoptera.The movement is facilitated by the posterior second valvifer-second valvula muscles (M9: Figs 1A-F, 2A, B, D, 3A, B, 4B, D-F, 5A, C, D; Vilhelmsen 2000).When the muscles contract they cause the bulb to pivot anteriorly at the basal articulation that is composed of the processus articularis and pars articularis (pra, paa: Figs 2A, 2F).As the bulb pivots, the distal end of the terebra move into an extended active position -compare Fig. 1B (partially extended) with Fig. 2B (retracted).The terebra is held in the longitudinal axis during oviposition: https://scholarsphere.psu.edu/files/r494vk42v At the end of the oviposition the terebra is retracted prior to the anterior rotation of the ovipositor/T9 complex into the resting, horizontal position.The retraction of the terebra, unlike its extension, seems to be unique for Ceraphronoidea.In other Hymenoptera the movement is accomplished by the contraction of the vertically oriented anterior second valvifer-second valvula muscle that arises from the anterodorsal margin of the second valvifer and inserts on the distal region of the bulb (King and Copland 1969, 1976, Vilhelmsen 2000, Fergusson 1988, Alam 1953).However, this muscle is absent from Ceraphronoidea, and thus we hypothesize a different mechanism for retracting the terebra.We observed a relatively large muscle arising from S7 and inserting dorsally on the first valvula in Ceraphronoidea (M1: Figs 1B, D, 3A).The position of the site of attachments of the muscle suggest that the S7-first valvula muscle (Figs 1B, 3A) aids in the retraction of the terebra.The presence of muscles arising from S7 and inserting on the first valvula has been reported only in some basal Hymenoptera (Dhillon 1966, Vilhelmsen 2000) and in one braconid species, Stenobracon deesae (Alam 1953).The S7-first valvulae muscle has an entirely different configuration in basal Hymenoptera.It arises from along the anterior margin of S7 and inserts on the proximal end of the ventral ramus of the first valvula and presumably contributes to the movement of the first valvula aiding the ventral T9-second valvifer muscle (Vilhelmsen 2000).In Stenobracon, however, the S7-first valvulae muscle has a very similar structure to that we reported in Ceraphronoidea (Alam 1953) suggesting the possible involvement of this muscle into the retraction of the terebra.Alam (1953) hypothesized that the muscle is involved in the movement of the median conjunctivae of the first valvulae (dorsal wall of the egg canal) influencing the egg movement along the egg canal.Although the anterior second valvifer-second valvulae muscle is present in all non-ceraphronoid Hymenoptera, it is possible that the S7-first valvulae muscle is involved in the retraction of the terebra in other Apocrita, and that this function is not an evolutionary novelty for Ceraphronoidea.

Egg laying mechanism
The paired first and second valvifers and T9, operated together by the dorsal and ventral T9-second valvifer muscles, form the ovipositor machinery that is responsible for "drilling" the terebra into a substrate and moving the egg along the egg canal (Vilhelmsen 2000).
As the dorsal T9-second valvifer muscle (M5: Figs 1A, B, D, E, 2A, B, D, 3B, 4A, B, D, E, F, 5A-C, 6A) contracts the first valvifer pivots posteriorly (in anterior to the left position) at the intervalvifer articulation while contraction of the antagonistic ventral T9-second valvifer muscle (M6: Figs 1A, B, D, E, 2A, B, D, 3B, 4A, B, D, F, 5A, B, 6A) pulls the first valvifer in the opposite direction (compare the position of the tergo-valvifer (tva) and intervalvifer articulations (iva) on Figures 3A and 3E).As this movement is what slides the first valvula along the second valvulae, the distance the first valvifer moves determines the distance the first valvula moves.The left and right first valvulae slide back and forth alternately during the alternate contraction of the left and right T9-second valvifer muscle pairs (1vf left, 1vv right: Fig. 3B; Vilhelmsen 2000).This alternate movement is what is underlying the advance of the egg inside the egg canal and the drilling of the terebra into a substrate.The former function is facilitated by the presence of internal, posteriorly oriented cuticular modifications (Austin and Browning 1981) while the latter could be aided by anchoring structures on the first valvulae (Vilhelmsen 2000).
Adaptations effecting the alternate movements and configuration of the first valvulae might be mostly affected by the hardness of the substrate and constraints for fast oviposition.Oviposition into a concealed, and therefore relatively immobile host requires a robust system that has to be strong enough to drill or break the barrier.On the other hand, parasitization of an exposed, mobile but relatively soft host requires fast and perhaps less robust mechanism.Two major egg laying habits have been recorded within Ceraphronoidea: oviposition inside a mobile host and oviposition trough a hard but relatively thin barrier enclosing the host, which has restricted movement (Dessart 1995b, c).Ceraphron and numerous Aphanogmus species were reported to parasitize free living Diptera larvae (Laborius 1972) whereas most Dendrocerus species and some Aphanogmus parasitize hosts hidden by the hardened integument of the primary host (Fergusson 1980), the wall of the cocoon (Peter andDavid 1990, Alam 1985) or galls (Bakke 1955) developed around the host.
The relative distance between the anterior angle of the first valvifer and the intervalvifer articulation (ang, iva: Figs 1B, 2C, E, 3C, E, 4A, E, F, 5A, 6A, C, D) is most probably positively correlated with the degree of the sliding motion of the first valvula (Prentice 1998).The posterior margin of the first valvifer angled at the tergo-valvifer articulation (tva: Fig. ) in most Hymenoptera (Oeser 1961, Vilhelmsen 2000).It is easy to see that the distance between the anterior angle of the first valvifer and the intervalvifer articulation and thus the degree of motion of the first valvula is larger with less acute angle at the tergo-valvifer articulation.A straight posterior margin of the first valvifer has been reported in a few Chalcidoidea (Epidinocarsis, Le Ralec et al. 1996;Spalangia, fig. 6 in Copland and King 1972) and in Apoidea (Pryonx, Prentice 1989).The only ceraphronoid taxon with a angled margin is Dendrocerus, while the rest have straight or concave (Trassedia) posterior margins.The location of the tergo-valvifer articulation on the posterior margin of the first valvifer seems to be also influenced by the way the first valvifer is moved.The closer the tergo-valvifer articulation is to the intervalvifer articulation and the further from the anterior angle, the further the first valvula will slide on the second valvula.The tergo-valvifer articulation is adjacent to the anterior angle of the first valvifer in Aphanogmus sp1.This indicates a first valvula sliding motion of very short distance but, considering the extended site of origin of the T9-second valvifer muscles, with relatively great power.The presence of an anterior angle corresponding to the tergo-valviferal articulation is unique in Hymenoptera.
Two ridges/apodemes are present on T9 in most Hymenoptera, the anterior flange of T9 and the cordate apodeme.The anterior flange extends along the anterior margin of the tergite and might be homologous with the antecosta of the ninth abdominal tergum of other insects because it receives the site of attachment of the dorsal T8-T9 muscle in Macroxyela (Vilhelmsen 2000).The dorsal T9-second valvifer muscle arises at least partly from the flange in the rest of Hymenoptera.The cordate apodeme is close to the tergo-valvifer articulation and receive the site of attachment of the ventral T9-second valvifer muscle.It is apophysis-like and extends internally in most basal hymenopterans but ridge-like and extended posteriorly in Siricoidea, Orussidae, and numerous Apocrita ("diagonal ridge" sensu Fergusson 1988).The apodeme is usually well separated from the anterior flange of T9 in Apocrita, except in Bruchophagus, where they are seemingly fused anteriorly (Copland and King 1971).Only one ridge, the anterior ridge of T9 extends along the anterior margin of T9 and receives the site of attachment of both the ventral and the dorsal T9-second valvifer muscles in Ceraphronoidea.This condition is unique in Hymenoptera.As described above, the dorsal and ventral T9-second valvifer muscles move the first valvifer indirectly.Host relationships of Megaspilus remain unknown, but it is possible that some bands of the ventral muscle insert on the interarticular ridge of the first valvula and thus the contraction of it might cause the direct movement of the sclerite.
The first valvifer is composed of two articulating sclerites in Trassedia.Although this condition is possibly plesiomorphic for Insecta (Klass et al. 2013) it is present only in a few taxa i.e.Archaeognatha, most ovipositor bearing Odonata and some Dermaptera (Klass personal communication).When the ventral T9-second valvifer muscle contracts the two sclerites of the first valvifer pivot anteriorly together as one unit.The two sclerites articulate with one another anteriorly, however, allowing the dorsal sclerite to pivot posteriorly on the ventral sclerite at the intravalvifer articulation when the dorsal T9-second valvifer muscle is contracted (Figs 3A-E).The first valvulae are thus enabled to slide a very long distance along the second valvulae in this unique system, probably allowing the egg to move quickly down the length of the ovipositor.
The presence or absence of annuli at the tip of the ovipositor may depend on the hardness of the substrate into which the wasp is ovipositing, as well as the circumstances under which oviposition is taking place (Quicke et al. 1999, Le Ralec et al. 1996, Gerling et al. 1998).Megaspilidae, similar to numerous other non-ichneumonoid apocritans do have annuli apically only on the second valvulae whereas Ceraphronidae lack them from both valvulae.In general it seems that the harder the substrate is, the more developed the ovipositor sculpture is (Le Ralec et al. 1996).
A minute gland (dgl?: Figs 3B, 5A) and a relatively larger gland reservoir (res: Figs 3A-E, 5A, B, E, F, 4C), enclosed by the second valvifers, have been detected in Ceraphronidae including Trassedia and Lagynodes.The Dufour's gland and the venom gland reservoir has a similar location in some Chalcidoidea (Copland and King 1971) where it was hypothesized that the ventral second T9-second valvifer muscles might aid to discharge the reservoir (res: Fig. 5B).The gland extract of this possible venom reservoir is resin-like (hard, transparent and amber colored) in critical point dried Cer-aphron specimens (res: Fig. 5F) implying its possible cement nature that might be used for coating the eggs or fastening them on a surface.Ceraphron and some Aphanogmus species are reported to be endoparasitoids (Cordero and Cave 1992) in which lifestyle the egg coating can be crucial for avoiding the host immune response.We did not observe any glands or resin containing reservoirs in Megaspilinae.Höller et al. (1993) identified the Dufour's gland in Dendrocerus carpenteri outside of the second valvifers and reported the absence of the venom gland in this taxon.Nevertheless, more accurate, TEM based examination of the accessory gland system of Ceraphronoidea is needed for clarifying the function of the gland and gland reservoir located inside the second valvifer.
In general, the ovipositors of Ceraphron, Trassedia and Aphanogmus sp. 2 are less robust and capable of a very large degree of motion, corresponding to the available data about ovipositing in exposed and active hosts.Trassedia represents, perhaps, a more extreme version of the "quickly into soft substrate" oviposition type.Megaspilidae, on the other hand, have a stronger, more robust ovipositor systems, which afford the smaller degree of motion required for handling a harder substrate concealing a static host.Dendrocerus, for example, exhibits extended sites of origins for muscles and a very small degree of motion for the first valvulae.Aphanogmus sp. 1, although it belongs to Ceraphronidae, shares numerous characteristics with Dendrocerus and therefore may represent the Aphanogmus-group that parasitizes hosts obscured by harder barrier, e.g., the wall of a plant gall, and thus is a case of parallelism driven by the same environmental constraints.
So far it is widely accepted that Ceraphronoidea is composed of two extant families, Ceraphronidae and Megaspilidae, plus two fossil families not treated here.The limits between the two families, however, have been challenged recently (Mikó and Deans 2009, Mikó et al. in press).Although the presence of the annuli in Megaspilidae and absence from Ceraphronidae supports the traditional classification, the location of a resin producing gland inside the second valvifers is shared by the megaspilid subfamily Lagynodinae and Ceraphronidae.
for his assistance with SEM and Rolf Beutel.This research was funded in part by the U. S. National Science Foundation (grants DBI-0850223, DEB-0842289) and benefited from discussions initiated through the Phenotype Research Coordination Network (NSF DEB-0956049).
Anatomical terms used, cross-referenced to an ontological (formal) definition (Hymenoptera Anatomy Ontology; URI = Uniform Resource Identifier).

First
elongated in lateral view Anterior margin of first valvifer: shape (0) convex in lateral view Posterior margin of first valvifer: shape (0) straight in lateral view (1) concave in lateral view (2) bent at tergo-valvifer articulation Transvalvifer conjunctiva: count (0) present (1) absent Dorsal sclerite of the first valvifer: shape (0) elongated in lateral view Ventral sclerite of the first valvifer: shape (anterior region of first valvifer Ventral margin of dorsal sclerite of the first valvifer: thickness (0) thickened Anterodorsal margin of ventral sclerite of the first valvifer: thickness (0) thickened Anterior flange of the first valvifer: count (0) present (1) absent Second valviferal condyle of the first valvifer: position (0) at the posteroventral corner of the first valvifer Ninth tergal condyle of the first valvifer: position (0) posterior margin of the first valvifer Distance between tergo-valvifer articulation and intravalvifer articulation (D1): proportion to the distance between tergo-valvifer articulation and anterior angle of first valvifer Ninth tergal condyle of the first valvifer: position (0) on dorsal sclerite of the first valvifer Anterior area of the second valvifer: height vs. height of posterior area of second valvifer Length of dorsal projection of second valvifer in lateral view (L1) vs. length of anterior area of second valvifer in lateral view (L2) (0) L1<L2 (1) L1>L2 Attachment site of S7-first valvula muscle on first valvula: position (0) anterior to attachment site of S7-first valvula muscle on S7 Attachment site of dorsal T8-T9 muscle on T8: position (0) adjacent to the anterior margin of T8 Attachment site of dorsal T8-T9 muscle on T9: position (0) adjacent to anterior margin of the dorsomedial region of T9 Attachment site of lateral T8-T9 muscle on T8: position (0) posterodorsal to the attachment site of dorsal T8-T9 muscle on T8 Attachment site of lateral T8-T9 muscle on T9: position (0) adjacent to anterior ridge of T9 Attachment site of T8-first valvifer muscle on T8: position (0) dorsal to the attachment site of lateral T8-T9 muscle on T8 Attachment site of T8-first valvifer muscle on first valvifer: position (0) adjacent to the ninth tergal condyle of the first valvifer Attachment site of T8-first valvifer muscle on first valvifer: position (0) on ventral sclerite of the first valvifer adjacent to the intravalvifer articulation Attachment site of dorsal T9-second valvifer muscle on T9: position (0) adjacent to the anterior ridge of T9 and the region of T9 dorsal to and ventral to the anterior ridge of T9 Attachment site of dorsal T9-second valvifer muscle on second valvifer: position (0) adjacent to the anterior section of the dorsal flange of the second valvifer Ventral T9-second valvifer muscle: count value (0) 1 Attachment site of ventral T9-second valvifer muscle on second valvifer: position (0) posterior area of the second valvifer Attachment site of ventral T9-second valvifer muscle on T9: position (0) adjacent to anterior region of anterior ridge of T9 First valvifer-second valvifer muscle: count (0) present (1) absent Attachment site of posterior T9-second valvifer muscle on T9: position (0) on the dorsal region of T9 Attachment site of posterior T9-second valvifer muscle on second valvifer: position (0) on the median bridge First valvifer-genital membrane muscle: shape (0) fan shaped in dorsal view Attachment site of first valvifer-genital membrane muscle on first valvifer: position (0) on the medial side of the first valvifer adjacent to the intervalvifer articulation Attachment site of first valvifer-genital membrane muscle on genital membrane: position (0) along the median line of the genital membrane Attachment site of first valvifer-genital membrane muscle on first valvifer: position (0) on the medial side of the dorsal sclerite of the first valvifer Attachment site of posterior second valvifer-second valvula muscle on second valvifer: position (0) on the medial side of the posterior area of the second valvifer (1) on the medial side of the posterior area of the second valvifer and the anterior area of the second valvifer Attachment site of posterior second valvifer-second valvula muscle on second valvula: position (0) on the processus musculares T9-genital membrane muscle: count (0) absent Lateral T9-second valvifer muscle: count (0) absent second valvifer-genital membrane muscle: count (0) absent anterior second valvifer-second valvula muscles: count (0) absent The area of the second valvifer which is anterior to the anatomical line that is the shortest distance from the first valviferal fossa of the second valvifer and the ventral margin of the second valviferThe flange that extends anteriorly on the first valvifer and overlaps with the posterior margin of the anterior area of the second valvifer. http://purl.obolibrary.org/obo/PATO_0001591