Redescription of Conostigmus albovarius Dodd , 1915 ( Hymenoptera , Megaspilidae ) , a metallic ceraphronoid , with the first description of males

Conostigmus albovarius Dodd, 1915 (Hymenoptera: Megaspilidae) is a species previously known by a single female holotype. Here, we provide a redescription of this peculiar ceraphronoid based on several female specimens and describe the male of the species for the first time. Intraspecifically-variable morphological traits such as female antenna color pattern are documented and discussed. A phenotype bank of morphological characters is provided for use in future megaspilid taxonomic treatments. We also provide phenotypic data in a semantic form to allow for ease of data integration and accessibility, making taxonomic data more accessible to future systematic efforts.


Redescription of Conostigmus albovarius
(Hymenoptera, Megaspilidae), a metallic ceraphronoid, with the first description of males

Introduction
Conostigmus Dahlbom, 1858 (Hymenoptera: Ceraphronoidea; Megaspilidae) is a relatively small genus of parasitoid wasps that has been largely neglected by modern taxonomic efforts, though they are commonly collected and are worldwide in distribution (Johnson and Musetti 2004).Relatively little is known about the biology of Conostigmus

Materials and methods
Air-dried point-mounted specimens were obtained from the Canadian National Collection of Insects (CNC).To prepare specimens for dissection, air-dried point-mounted specimens were removed from their points, cleared with BioQuip Specimen Clearing Fluid #6373, and subsequently kept in glycerol for dissection and storage.Dissections were performed with #2 insect pins and #5 forceps.Air-dried point-mounted specimens and glycerol-stored specimens were examined using an Olympus SZX16 stereomicroscope with an Olympus SDF PL APO 1X PF objective (115×) and an Olympus SDF PL APO 2X PFC objective (230× magnification).Blue-Tack (Bostik, Inc., Wauwatosa, Wisconsin, USA) was used to stabilize specimens while making observations and images.Measurements were taken using a KR 851 stage micrometer (1mm in 100 divisions) attached to the same microscope.Bright field images were taken with an Olympus DP71 digital camera attached to an Olympus ZX41 compound microscope.Images were subsequently aligned and stacked using Zerene Stacker Version 1.04 Build T201404082055.
For the male and female species descriptions, 87 morphological characters were scored based on observations of air-dried point mounted specimens and glycerol stored specimens.Following the Phenotypic Quality Ontology (PATO; available at http://obofoundry.org/), the preferred label "count" was used instead of its synonyms, including "presence".It logically follows that the character "presence" should not include "absent" as a state because the correct use of this syntax would be "presence/ absence: absent", which would over-complicate descriptions.Specimen data, OTU concepts, natural language phenotypes and images of specimens were compiled in the MX database (http://mx.speciesfile.org).This software was used to render the Diagnosis, Description, and Material Examined sections.Definitions and descriptions of the morphological terms and structures were mapped to classes in the following phenotype-focused ontologies: Hymenoptera Anatomy Ontology (HAO), Phenotypic Quality Ontology (PATO), Biospatial Ontology (BSPO), OBO Relation Ontology (RO), Ontology for Biomedical Investigations (OBI), and Information Artifact Ontology (IAO), all available at http://www.ontobee.org/(Xiang et al. 2011).
Semantic statements were generated in Protégé Version 5.0.0 (Build beta-17) to build an ontology where phenotypes are represented semantically.Standardizing taxonomic data through ontology-based semantic representation could benefit future systematic work by allowing taxonomic data sets from different sources to be easily integrated, expediting computerized searches across these data sets (Deans et al. 2012;Balhoff et al. 2013;Mikó et al. 2014).Statements were written in OWL Manchester syntax, modeled after examples set by Balhoff et al. (2013), Mikó et al. (2014), and Mikó et al. (2015).A list of the semantic statements generated is presented in Supplementary file 1: Appendix 1.The data matrix file, the NEXUS file corresponding to the data matrix and the semantic annotations file are available at http://dx.doi.org/10.6084/m9.figshare.1539620,http://dx.doi.org/10.6084/m9.figshare.1544551,and http://dx.doi.org/10.6084/m9.figshare.1539621.
Abdomen and male genitalia.S1 length vs. shortest width: S1 wider than long.Distal margin of male abdominal sternum 9 shape: straight.Proximolateral corner of abdominal sternum 9 shape: blunt.Cupula length vs. gonostyle-volsella complex length: cupula less than 1/2 the length of gonostyle-volsella complex in lateral view.Proximodorsal notch of cupula count: present.Proximodorsal notch of cupula shape: arched.Distodorsal margin of cupula shape: straight.Proximodorsal notch of cupula width vs length: wider than long.Proximolateral projection of the cupula shape: blunt.Distoventral submedian corner of the cupula count: absent.Dorsomedian conjunctiva of the gonostyle-volsella complex count: present.Dorsomedian conjunctiva of the gonostyle-volsella complex length relative to length of gonostyle-volsella complex: dorsomedian conjunctiva extending 2/3 of length of gonostyle-volsella complex in dorsal view.Distal end of dorsomedian conjunctiva of the gonostyle-volsella complex shape: acute.Parossiculus count (parossiculus and gonostipes fusion): present (not fused with the gonostipes).Apical parossiculal seta number: one.Distal projection of the parossiculus count: absent.Distal projection of the penisvalva count: absent.Dorsal apodeme of penisvalva count: absent.Harpe length: harpe shorter than gonostipes in lateral view.Distodorsal setae of sensillar ring of harpe length vs. harpe width in lateral view: setae as long or shorter than harpe width.Distodorsal setae of sensillar ring of harpe orientation: medially.Sensillar ring area of harpe orientation: medially.Lateral setae of harpe count: present.Lateral setae of harpe orientation: oriented distally.
Specimens will be deposited at the Canadian National Collection of Insects (CNC), Ottawa, ON, Canada and at the Frost Entomological Museum (FEM), University Park, PA, USA.
Comments.Conostigmus albovarius stands out from other species of Conostigmus due to its unique, white color pattern, for which the species was named (albus as in "white" and varius as in "variegated") (Dessart 1997).Dessart (1997) was intrigued by the stark white color present on the back of the mesosoma and on portions of the legs and antennae.He attributed the unique color of the antenna and legs to a lack of pig- ment in these areas, but voiced concerns whether the color of the mesosoma could be an artifact due to damage of the sole holotype specimen (Dessart 1997).With the discovery of ten new specimens, it is now clear that this coloring of the mesosoma is not an artifact, and that it is a phenotype shared by both females and males of the species (Figs 1, 2, 3).
bers of Conostigmus exhibit reticulate sculpturing (Yoder et al. 2010), which refers to the polygonal microreticulation of the cuticular surface that is most likely based on the pattern of epithelial cells (Krell 1994).C. albovarius, in contrast, exhibits foveolate sculpturing.Our definition of foveolate sculpturing is based on Harris (1979), where the cuticle is divided into irregular pits with raised edges and a single seta is present at the center of each pit.In females, foveolate sculpturing is present on the head and mesosoma, while in males the foveolate sculpturing is only present on the frons.In females, the areas with foveolate sculpturing are also present with a metallic coloration ranging from a bronze sheen to a deep iridescence (Fig. 4).This metallic coloration is absent from males.
In comparing the antennae of different female specimens, it was observed that there was variation in the coloration of the apical flagellomeres (Fig. 1).Whereas F9 is always melanized and F1 through F4 always have transparent cuticle (it appears white because of the soft tissue, e.g.fat bodies and muscles, underneath), F5 through F8 vary in whether melanization is present or not.When melanization is present, it is always present in the apical flagellomeres after the melanized flagellomere, such that if F5 is melanized, then F6-9 will also be melanized.It is unclear whether this intraspecific phenotypic variation in color is influenced by genetic or environmental factors, such as temperature (Quicke 1997).Females from different sampling events in different areas sometimes shared the same pattern of melanization on the antennae, though females collected from the same sampling event sometimes had different patterns.
It is known that the antennae play important roles in the courtship of parasitic wasps in general (Ayasse et al. 2001;Romani 2008).In Cotesia rubecula (Hymenoptera: Braconidae), females use their antennae to signal their receptivity to the males (Field and Keller 1993).It is possible that the melanization seen in female C. albovarius antennae could be used for visual signaling to males during courtship.

Table 1 .
A list of the locality data for the specimens examined.