Anatomic glossary of mesopleural structures in Bethylidae (Hymenoptera: Chrysidoidea)

The mesopleuron of Bethylidae has many structures that are used in taxonomic and phylogenetic studies. The lack of understanding of these structures has generated independent terminologies and a series of confusing terms, hampering effective scientific communication. A morphological study and literature review were made in order to solve these problems. Our study resulted in an anatomic glossary with 49 terms that presented a large number of synonyms and polysemies. The glossary standardizes the terms used in the Bethylidae mesopleuron and in other Hymenoptera groups, which will facilitate hypotheses of primary homology in comparative biology.

Although these studies provided great contributions to the understanding of Hymenoptera and insect anatomy, anatomical studies restricted to more limited anatomical regions allow their characters to be discussed in more detail, and facilitates a more comprehensive comparative analysis. Examples are Shcherbakov (1980Shcherbakov ( , 1981 and Gibson (1993), who contributed considerably to the understanding of the thoracic pleura of Hymenoptera, accomplishing a detailed study of the external and internal morphology, comparing structures, standardizing terms and discussing the variations in different taxa.
Despite the various morphological studies, many cladistic and taxonomic studies explore the anatomy of taxa in a restricted and particular way, creating new terms for the analyzed char-acters or using previous terms without further exploring the homology between the characters. Such factors contribute to a progressive increase in the number of polysemy and synonyms in the terminology. This is evident in groups within Hymenoptera, in which evolutionary events have caused the loss or fusion of various anatomical structures (e.g., mesopleural and mesopleurosternal sutures). To try to minimize these problems in Bethylidae, Kawada et al. (2015) carried out a study of the skeletomusculature of the metapostnotum to solve problems of delimiting the boundary between the metanotum and the propodeum. However, there are areas that still need a skeletomuscular study to understand their structures and propose homologies, such as the mesopleuron.
This study aims to propose a detailed glossary of the terms and concepts of mesopleural structures, as well as a list of synonyms, which will assist in future investigations of mesopleuron in Bethylidae and other groups of Hymenoptera, serving as a basis for comparison, understanding and character recognition as well as their use in phylogenies.

Sample
Specimens of 33 genera of Bethylidae belonging to all five extant subfamilies were examined covering most of the diversity of the mesopleuron (Table 1)

Dissection techniques and images
Critical-point-dried and air-dried specimens (Gibson, 1985;Vilhelmsen et al., 2010;Kawada et al., 2015) were used for morphological analysis, dissected with insect pins and razor blades. The specimen's parts were glued on a card with information on the sample's provenance and stored as vouchers.
The external and internal part of the mesopleuron was photographed with a Leica Z16 APO stereomicroscope coupled to a Leica DFC 2 video camera by Leica Microsystems (Switzerland) following the lighting procedures of Kawada & Buffington (2016). The image stacks were combined with Leica Application Suite V3.6.0 by Leica Microsystems (Switzerland) or Helicon Focus (HeliconSoft).

Terminology
The terms and descriptions of the morphological characters and muscles follow The Hymenoptera Anatomy Ontology project (Seltmann et al., 2012) available through The Hymenoptera Glossary (HAO http:// glossary.hymao.org) and Lanes et al. (2020). The homologisation of structures (primary homology) between Bethylidae and other Hymenoptera was made by comparing the shape, location and muscles associated with the structures. When it was not possible to establish homology through association with the musculature, we applied the criterion of the topographical sameness of structures (see Seltmann et al., 2012). Lanes et al. (2020) and Seltmann et al. (2012) use the term homonymy when the same name is used for more than one structure. However, according to Martins (2015), Table 1. Genera examined.

Subfamily
Genera Bethylinae the homonymy is related to cases in which two or more meanings are presented by words with the same spelling and/or sound, but have different semantic roots, while polysemy is characterized when a word with the same semantic root has two or more meanings. Furthermore, Alves (2000) explains that polysemy and homonymy may represent cases in which the same term is used to name different concepts, but polysemy deals with concepts from the same area or conceptual network, while homonymy from different conceptual areas or networks.
From that, we will adopt polysemy rather than homonymy when one name is applied to two or more structures and synonymy when the same structure has more than one different name applied to itself. We establish three criteria according to the order of the importance to define the preferential term: (1) was preferred the term that best represents the skeletal structure (shape and location in the body) according to Lanes et al. (2020), Ronquist & Nordlander (1989) and HAO. (2) more widely used term in the scientific community; (3) oldest proposition of the term. The criteria adopted are informed in the description of the structure.
In polysemy cases or if the adopted terminology was not adequate (second criterion described in the previous paragraph), a new name was proposed. Some descriptions have been expanded because the previous definitions were insufficient to describe the characters. These definitions were structured following the genus differentia format, using the existing definition and from it we add the new part to cover the structural concept.
The positioning of characters in relation to the body follows the Biological Spatial Ontology -BSPO (Dahdul et al., 2014;BSPO) and the interpretation follows The Hymenoptera Anatomy Ontology (Yoder et al., 2010;Seltmann et al., 2012;HAO).
From the terminological glossary by Lanes et al. (2020) and the bibliography consulted, which presents information about structures of the mesopleuron in Hymenoptera (mainly morphological terms and concepts) and Bethylidae specifically (including taxonomic, morphological and cladistic terms applied to structures in mesopleuron), a new glossary for the mesopleural structures was made, in which the detailed description of the morphological characters and the commonly used synonyms were included. Valid terms and abbreviations are in bold and new terms are marked with one asterisk (*). In case of articles written in a non-English language, we translated the terms and kept the term in the original language in parentheses and marked with a double asterisk (**). When the term represents only part of a structure, we include the term "in part".

General description of the mesopleuron in Bethylidae
Mesopleuron occupies a large part of the mesopectus (mesopleuron + mesosternum), being divided into lateral and ventral surfaces. Located dorsally in the lateral surface, the mesepimeral sulcus (epm2s) separates the posterior mesepimeral area (epm2pa) from the rest of the mesopleuron, which is composed of the mesepisternum and part of the mesepimeron. The posterior mesepimeral area is projected anteriorly, forming the mesepimeral lobe (mlo). The subalar area (saa) is located anterodorsally in the mesopleuron and is ventrally limited by the subalar impression (sai), within which are the anterior subalar pit (asup) and posterior subalar pit (psup). Ventral to the subalar impression may have the anterior mesopleural fovea (pl2af). Posteriorly to this fovea there are a mesopleural callus (pl2c) or an upper mesopleural fovea (pl2uf) or a posterior oblique sulcus of the mesopleuron (pl2pos). Centrally in the mesopleuron, the mesopleural pit (pl2p) is present, which can be internal to the upper mesopleural fovea or to the posterior oblique sulcus of the mesopleuron. Ventrally in the lateral surface, the lower mesopleural fovea (pl2lf) is usually present and at the posterior extremity there is the mesopleural epicoxal lobe (pl2el), which may be limited by the mesopleural epicoxal sulcus (pl2es). In the ventral surface, subpleural signum (sps) and sternaulus (st) are usually present.

Integumentary structures
Acropleural apodeme (acra: Figs. 1A, 2D) = The apodeme located anterodorsally on the mesopleuron and correspondent to the site of origin of the third mesopleuro-mesonotal muscle (HAO_0000083). Although the acropleural apodeme is present in Bethylidae, we were not able to see the third mesopleuro-mesonotal muscle. Anterior subalar pit (asup: Fig. 1C) = The pit located anterodorsally on the mesopleuron that corresponds internally to the acropleural apodeme*. Criterion 1.

Axillary sclerites
First axillary sclerite of the forewing (1ax2: fig. 44 in Gibson, 1986) = The axillary sclerite that articulates anteriorly and medially with the anterior notal wing process laterally with the second axillary sclerite, and posteriorly with the posterior notal wing process (HAO_0001772). Criterion 1.

DISCUSSION
As proposed by Freixa (2006), denominative variation in terminology responds to linguistic needs of different natures, being that they occur in a natural way and, sometimes, they can result from the enhancement about the interpretation of the studied objects (evolution of concepts). However, contrary to what was defended by Alves (2000) and Martins (2015), we believe that, in some cases, the presence of synonyms, homonyms and polysemies can be considered problematic in specialized communication, especially when different denominations for the same concept are a product of imprecise and/or ambiguous definitions of conceptual boundaries (i.e., denominative variation caused by conceptual imprecision; see cognitive causes in Freixa, 2006).
Based on the morphological analysis of several Bethylidae taxa and their respective descriptions and terminologies adopted, we verify that over the years, a large number of imprecise terms have been proposed to name the same mesopleural characters applied to Bethylidae or even to Hymenoptera. We believe that the huge number of terminological synonyms presented in the anatomical glossary can be evidence of that.
Many terms are not identified due to the conceptual imprecision regarding the structure to which it was referring, usually caused by incomplete descriptions and lack of illustrations. Among the characters identified, many are difficult to interpret and the factors that contribute to this are: the structural equivalence between the structure is difficult to establish, the lack of knowledge about what it represents morphologically and the lack of a clear delimitation of the structures in some cases (e.g., subalar impression, mesepimeral sulcus, sternaulus). Moreover, many areas and characters are little explored in taxonomic and cladistic contexts (e.g., anterior and posterior subalar pit, sternaulus, mesopleural epicoxal lobe and subalar area).

Synonymous and polysemous terms
In this topic we will discuss the main cases of the synonymous and polysemous terms found in Bethylidae literature (mainly) and, in some cases, groups of Hymenoptera.
In Bethylidae, the posterior oblique sulcus of the mesopleuron is present in Pristocerinae and Mesitiinae, and despite showing little variation in size, position and format, it has been labelled differently in both subfamilies (see Argaman, 2003;Barbosa & Azevedo, 2011a, b;Gobbi & Azevedo, 2016).
The use of names such as mesopleural sulcus (see Barbosa & Azevedo, 2011c;, for example, can lead to misidentification with another sulcus present in the mesopleuron e.g., sometimes, the lower margin of the mesopleural fovea form a structure like a sulcus externally. Names such as transversal furrow (see Gobbi & Azevedo, 2016) and other terms present in Bethylidae literature can make identification by systematists difficult, because in the name there is no inference about the location of the character, making it difficult to correlate the name and the structure it represents through literature, especially when interpreted by non-specialists in Bethylidae. Furthermore, this sulcus is also found in other Hymenoptera groups, e.g., Chrysididae and Pompilidae, and was termed posterior oblique sulcus of the mesopleuron by Richards (1956), the same term established in Bethylidae as described by Lanes et al. (2020).
Establishing the correspondence of the structure can be difficult, even when there is little structural variation, as the mesopleural callus. This area was named by Evans (1961) as "a more or less well defined swelling just below the base of the hind wings" to the genus Pseudisobrachium that belongs to Pristocerinae. Because of this, the term was widely used in workers of this subfamily. However, in some genera of the Epyrinae, this area is more accentuated, which took Alencar & Azevedo (2013) to name it as upper mesopleural elevation. Lanes et al. (2020) standardized the term acropleural area to this region, but our muscular analysis is in disagreement with the definition presented by them (see definition proposed by Lanes et al. (2020) and our definition in the glossary section). Because of this, we will keep the term mesopleural callus (as proposed by Evans (1961), but with a new conception. This terminological unification based on an anatomic study will make the works in Bethylidae safer and more understandable and it can serve as an area to be explored and studied in other groups of Hymenoptera.
The subalar impression has recently been proposed for Bethylidae and it has received many terms. Externally, this structure is fused posteriorly with mesepimeral sulcus, being considered and named as a single structure in the publications of the Bethylidae. Therefore, the terms found in the literature review refers to the subalar impression and part of the mesepimeral sulcus. Of all citations of this term, only the publications by Barbosa & Azevedo (2014a, b;2015;, and Barbosa et al. (2017) seem to have followed the original proposition of Gibson (1986), which used the term acropleural sulcus to this area.
The problems related to the terminology of the mesepimeral sulcus arises mainly from the lack of the knowledge and the misinterpretation of the type of structure it represents, whether a groove (e.g., episternal groove and mesopleural groove), suture (e.g., mesepimeral suture) or carina (e.g., postpectal carina) that do not represent the structure, more complete definitions and names about the structures can be found in Lanes et al. (2020), Ronquist & Nordlander (1989) and HAO. Despite these terms, the names mostly used in Bethylidae are episternal furrow and episternal groove, which was cited for the first time by Azevedo (2008a) and . However, they did not define the term.
The mesopleural pit is an example of the lack of uniformity in terminology not just in Bethylidae, but also in other families of Hymenoptera (e.g., episternal scrobe to bees - Michener (1944); pleural apophysis to bees - Snodgrass (1942); mesopleural scrobe to Braconidae - ; pleural pit Platygastroidea - Mikó et al. (2007)). Although there are no doubts about the identity of the mesopleural pit, several terms have been applied to this structure, which shows the importance in using the uniform terminology in taxonomy, cladistic and morphologic works. To adopt this practice can make the use of data from the literature more reliable and less laborious.
The lack of consistent criteria for proposing primary homologies (sensu De Pinna, 1991) may be one of the factors that promotes proliferation of terminological polysemy or synonyms. Even when there is concern with the proposal of homologies through comparative anatomy, the superficial analysis of the characters is not always sufficient, since some of them may be externally similar, but internally they represent totally different structures. An example of the lack of the consistent criteria is the case of the "transepisternal line". Lanes et al. (2020) used the term "transepisternal line" to name a ventral sulcus of the lateral surface of the mesopleuron, but for Gibson (1985) the "transepisternal line" marks the origin of muscles that does not originate from this sulcus, which, in fact, appears to be only the inferior margin of the lower mesopleural fovea.
Many works, mainly taxonomic, prioritize the description of the specimens, with no concern for recognizing whether the analyzed characters have already been described in other taxa that do not belong to the group of the study, which leads to the creation of new terms for such characters. In addition, terminological polysemy or synonyms may be the result of a lack of extensive research on the terminology (e.g., databases such as HAO and Antkey) and of the different motivations by systematics and morphologists. Thus, the stability in terms requires agreement as to which terms should be used to name the structures without affecting the scientific communication (Gibson, 1993).
The "subalar pit" is an example of the importance in agreement between systematics and morphologists. The mesopleuron has two small pits in the subalar region, anterior and posterior, however, in some taxa only one of the pits is present, so that Richards (1956) used the term "subalar pit" to name the anterior subalar pit, whereas Gibson (1985) used the same term to name the posterior subalar pit.

Poorly explored characters
The study of the mesopleuron showed not only the large number of synonyms existing for its structures, but also revealed many little explored characters that can be useful in systematics.
Some of them are certainly under-explored because of the location and size of the structure, e.g., the anterior subalar pit, posterior subalar pit, subalar area and mesepimeral lobe. All of them are located at the anterior extremity of the mesopleuron, and are often difficult to observe because the wing generally covers them. They are small compared to other structures. In these cases, the state of preparation of the specimen can greatly influence observability, because the position of the wing can not only make it difficult to observe, but to correctly identify and interpret the structures. Thus, this area of mesopleuron probably is neglected due to the uncertainty of their homology, or because of the risk of damage to the specimen while studying them.
However, sternaulus and epicoxal mesopleural lobe are easy to observe, but little investigated in Bethylidae. The sternaulus, for example, was mentioned only three times for Bethylidae, as sub-anterior fovea by Alencar & Azevedo (2011a) and Stein et al. (2011) and sternaulus by Lanes et al. (2020). The mesopleural epicoxal lobe was mentioned only once by Ramos & Azevedo (2012). In our study we observed that these structures vary considerably and can be useful for the taxonomy of the group, as well as Hymenoptera.
Although we did not used any specific methodology to explicitly evaluate character usage throughout the literature, we believe that the little use of them may be related to some factors: (i) researchers generally follow standardized descriptions previously adopted, without considering their inherited descriptive and/or explanatory limitations for particular groups; (ii) researchers explore only a fixed set of characters determined by their historical usage in a particular group; and (iii) specialized literature that provides explicit information and explanation of characters is virtually absent for a particular group.

Negative effects of the absence of uniform terminology
The lack of a specific literature and the wide range of morphological variability lead specialists of different taxonomic groups to develop their terminologies independently, without standardized vocabularies, which generates many synonyms (Gibson, 1985;Seltmann et al., 2013). This may result in the accumulation of terms that often are confusing or that do not represent the structure and uncertain homology, making the taxonomic and cladistic work mostly ineffective and constitutes a barrier to clear and accurate scientific communication.
According to Komarek & Beutel (2006) insufficient morphological knowledge of the characters can result in imprecise or inadequate observations, and important features may be overlooked or misinterpreted. This shows the importance of morphological knowledge and consequently terminological standardization in a taxonomic context, especially to new taxonomists. At the beginning of their careers, many of them are unable to have a clear understanding of some structures, may replicate any eventual error or find the work dull and difficult, especially for under-studied taxonomic groups.
For cladistic analysis, the effects on the lack of uniformity of terminology can result in a character matrix with many problems, because the incorrect identification of the structure may not correspond to the primary hypotheses of homology. Examples of this problem can be found at Waichert & Azevedo (2009) and Alencar & Azevedo (2013). In the first situation, the term subtegular groove is used in their matrix of the characters, but it corresponds to two distinct structures, the subalar impression and mesepimeral sulcus, which are not delimited externally, and, therefore, seem to form a single structure. In the second situation, the terms supra-posterior callus and posterior elevation on upper fovea in Alencar & Azevedo (2013), are used to name the same structure, present in different groups that is the mesopleural callus.
Furthermore, in groups with sexual dimorphism, some structures are difficult to correlate. Thus, the standardization of terms and determination of homologous regions among species is important to infer phylogenetic relationships (Gibson, 1986).

CONCLUSION
The study of the mesopleuron of bethylids showed how much a study of certain regions of the body is necessary and how much it can not only add new characters, but also understanding about certain structures. The standardization of the terminology of the mesopleural structures will facilitate the study of the mesopleuron of Bethylidae, generating uniformity in future studies, which can serve for comparison and understanding not only within the family, but in other groups.
Even understanding the importance of terminological variation, especially when it relates to the improvement of concepts, we believe that the standardization of terminologies should be managed with due care, respecting recommended standards and established by consensus, considering terminological variation as part of the process.
Thereby, the efforts of The Hymenoptera Anatomy Ontology (HAO) and many other morphological studies have contributed to reducing the problem related to this subject. Gibson (1993) stated that systematics and morphology are inextricably related because each is dependent on the other for its own success, which is increasingly necessary.

ACKNOWLEDGMENTS
We thank the curators of the collections for authorization to use the analyzed specimens. We thank Ricardo Kawada for loaning specimens for study. We thank the Estação Biológica de Santa Lúcia for allowing us to collect specimens used in this study. We are also very grateful to the subject editor Helena Onody and reviewers Lars Vilhelmsen and anonymous for the revision and suggestions regarding the manuscript. The first author was supported with a Ph.D. bursary of Demanda Social by CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior); third author was supported with a bursary by CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico), INCT-CNPq HymPar grant # 303748/2018-4.