The male genital and pregenital skeleton and musculature were studied in males of the following species of the Muscidae subfamily Azeliinae: Drymeia firthiana (Huckett, 1965), Drymeia longiseta Sorokina & Pont, 2015, Drymeia segnis (Holmgren, 1883), Thricops nigritellus (Zetterstedt, 1838), Thricops hirtulus (Zetterstedt, 1838), Hydrotaea dentipes (Fabricius, 1805), Muscina stabulans (Fallén, 1817), and Muscina levida (Harris, 1780). Descriptions and figures of the genital sclerites and muscles of D. firthiana and M. stabulans are given. A comparison was made between the genital segments and muscles of previously studied species of Mydaeinae and Muscinae and those of the Azeliinae. Based on the structure of the skeleton and muscles of syntergosternite VII + VIII and the phallapodeme muscles, significant differences were found between the subfamily Azeliinae and the subfamilies Mydaeinae and Muscinae. The basal position of the Azeliinae within the family Muscidae was confirmed. A comparison of the genital segments and muscles of the Muscidae with those of the Scathophagidae (Scathophaga stercoraria (Linnaeus, 1758)) and Anthomyiidae (Delia platura (Meigen, 1826)) was made. Tendencies in reduction of the pregenital segments and musculature, as well as of the phallapodeme muscles in the evolution of the Muscoidea have been revealed. The complete set of phallapodeme muscles in the Scathophagidae and Anthomyiidae corresponds to the basal state, and therefore the structure of the genital sclerites and muscles in the Muscidae shows a certain degree of reduction. The progressive changes in the Muscidae from the Azeliinae through the Mydaeinae to the Muscinae were traced.

abdominal segments, Calyptratae, flies, male genitalia, muscles, Muscoidea, pregenital segments, sclerites

The Muscidae is one of the largest family of the Calyptratae (Diptera) and the largest of the Muscoidea. The world fauna includes approximately 5000 species in 180 genera (Pape et al. 2011). Despite the use of various modern methods of phylogenetic analysis, the classification of the family is still unstable and changeable, and sometimes controversial (Carvalho 1989; Carvalho et al. 2005; Schuehli et al. 2007; Fan 2008; Kutty et al. 2010, 2014, 2019; Haseyama et al. 2015; Grzywacz et al. 2017). The monophyly of the family and of some of its subfamilies (i.e., Azeliinae, Muscinae, and Coenosiinae) has been established beyond doubt, based on morphological characters and molecular data. However, the monophyly of some subfamilies (i.e., Mydaeinae and Phaoniinae) has not been supported and the traditional tribal classification has been rejected (Kutty et al. 2014, 2019; Haseyama et al. 2015).

Among the morphological characters used in phylogenetic reconstructions and classification systems, the characters describing the morphology of the muscles of the genital and pregenital structures are usually more stable than those of the sclerites (Matsuda 1976; Ovtshinnikova 1989; Friedrich and Beutel 2008). Moreover, study of the muscles helps to clarify function and homology and reveals parallelisms in the pregenital and genital sclerites (Ovtshinnikova 1989, 1994; Ovtshinnikova and Yeates 1998; Galinskaya and Ovtshinnikova 2015; Galinskaya et al. 2018; Ovtshinnikova et al. 2019).

Very few papers have dealt with the study of the male genital muscles of the Muscoidea. Hennig (1976) produced the first work on the Anthomyiidae, describing in detail the male genital muscles of Delia platura (Meigen, 1826) and Fucellia tergina (Zetterstedt, 1845), while the musculature of the pregenital sclerites was not studied. Later Ovtshinnikova (1989, 1994) studied the muscles in Musca domestica Linnaeus, 1758 and Scathophaga stercoraria (Linnaeus, 1758). Most recently, studies of the musculature of the male terminalia of the Muscoidea, and in particular that of the Muscidae, have continued with Ovtshinnikova et al. (2018, 2019) detailing the muscles of the genital and pregenital structures in some members of the Muscinae (Musca autumnalis De Geer, 1763, Pyrellia rapax (Harris, 1780)) and the Mydaeinae (Mydaea urbana (Meigen, 1826), Graphomya maculata (Scopoli, 1763)).

This paper continues the series of publications devoted to the structure of the sclerites and muscles of the abdominal segments and male terminalia in Muscidae and presents the results of our study of another subfamily, the Azeliinae. The Azeliinae was recognized as a subfamily following cladistic analyses by Carvalho (1989, 2002) and Carvalho et al. (2005). Previously, according the Hennig’s classification, the Azeliini was treated as a tribe within the subfamily Muscinae and this was followed by many authors (e.g., Hennig 1965; Pont 1986; Gregor at al. 2002). However, Skidmore (1985) maintained the Azeliinae and Reinwardtiinae as separate subfamilies mainly because the larvae of Azeliinae are trimorphic or dimorphic, facultative or obligate carnivores, or parasites, unlike the larvae of other Muscinae (Skidmore 1985). Before this study, Lobanov (1979, 1984) had written about a branch of the Azeliinae (“Hydrotaeinae” in Lobanov) on the basis of his studies of the female ovipositor. Savage and Wheeler (2004) conducted a genus-level phylogenetic analysis of the tribe Azeliini within the composition of the subfamily Azeliinae.

The subfamily Azeliinae is currently divided into two tribes: Azeliini and Reinwardtiini. Most of the members of the subfamily are known as anthophilous insects, but others are known as sweat flies or synanthropic flies. The larvae are mainly carnivores and can develop in humus soil, or are saprophages developing in various decomposing substrates including human and animal feces, or are even parasites of birds. The subfamily is cosmopolitan, but with a much higher diversity in the northern hemisphere for the Azeliini (389 species in 12 genera) and in the southern hemisphere for the Reinwardtiini (128 species in 17 genera) (A. C. Pont, pers. comm.).

The monophyly of the subfamily Azeliinae is still a matter for discussion. Only the monophyly of the tribe Azeliini has been established, based on morphological characters, molecular data and also the structure and lifestyle of the larval stage (Savage and Wheeler 2004; Schuehli et al. 2007; Kutty et al. 2014; Haseyama et al. 2015). The Azeliinae are paraphyletic in all molecular analyses because Muscinae is sister-group to the Azeliini while the Reinwardtiini are polyphyletic. There are no morphological synapomorphies to support monophyly of the current Azeliinae (Azeliini + Reinwardtiini), nor do the Reinwardtiini emerge as sister-group of the Azeliini, though according to other classifications the tribe Reinwardtiini has also been shown to be the sister-group of the Azeliini (Carvalho 2002; Savage and Wheeler 2004; Carvalho et al. 2005; Savage 2009). It has even been suggested that the Reinwardtiini should be treated as a separate subfamily (Skidmore 1985; Couri and Carvalho 2003). The monophyly of the Azeliinae has been confirmed in the recent paper by Kutty et al. (2019). However, only one species (Muscina stabulans Fallén, 1817) was used in this analysis.

This paper presents the results of our study of the sclerites and muscles of the male abdominal segments and terminalia in members of the subfamily Azeliinae belonging to the genera Drymeia Meigen, 1826, Thricops Rondani, 1856, Hydrotaea Robineau-Desvoidy, 1830 (Azeliini), and Muscina Robineau-Desvoidy, 1830 (Reinwardtiini).

The muscid material used in this paper is deposited in the collection of the Institute of Systematics and Ecology of Animals, Russian Academy of Sciences, Siberian Branch, Novosibirsk, Russia (SZMN).

To study the genital sclerites, dry specimens were softened in a hydration chamber; the abdomen was then detached, treated with 10% KOH solution, and dissected. The sclerites are designated here following the terminology of Sinclair (2000). The male abdomen consists of five segments; the pregenital segments VI–VIII are strongly modified as a result of the clockwise rotation of the male genitalia by 360°; the genital segments IX–XI are strongly modified.

The muscles of the male genitalia were studied by manual dissection of specimens preserved in 70% ethanol, using microknives, under a Leica MZ9⁵ stereomicroscope. The illustrations were made in Photoshop CS6 and CorelDRAW X6, based on digital images of muscles and sclerites captured with a Canon EOS 77D camera mounted on the Leica MZ9⁵ trinocular head. The genital muscles are classified into the following groups: abdominal, pregenital and genital muscles (tergosternal muscles, muscles of the hypandrial complex, and muscles of the epandrial complex). The muscles are numbered according to the classification of Ovtshinnikova (1989, 2000) and grouped by the sites of their origin.

The following abbreviations are used in the text: c – cercus; dph – distiphallus; ej – ejaculatory apodeme; ep – epandrium; eph – epiphallus; hyp – hypandrium; l – left muscle; r – right muscle; pgt – postgonite; phap – phallapodeme; prgt – pregonite; sbeps – subepandrial sclerite; sp – spiracle; st – sternite; stgst – syntergosternite; sur – surstylus; tg – tergite; ISM – abdominal and pregenital intersegmental sternal muscles; ITM – abdominal and pregenital intersegmental tergal muscles; M1–M26 – pregenital and genital muscles; TSM – abdominal and pregenital tergosternal muscles.

The muscle M18 includes asymmetric muscles which are designated in this paper as M18 r and M18 l. This corresponds to the previously accepted designations M18¹ and M18² in Scathophaga (Ovtshinnikova 1994).

Because of genital rotation, sclerites of the pregenital segments do not always lie in the usual position. For this reason, characteristics such as “wide” or “narrow” in the descriptions describe only the geometric shape of the sclerites, regardless of their orientation relative to the body axis.

In the Muscidae as well as in other Cyclorrhapha, both the sclerites and the muscles of abdominal segments VI–VIII and partly IX are asymmetrical as a result of the clockwise rotation of the male genitalia by 360°. The pregenital sclerites of segments VI–VIII are partly reduced, modified, and fused. In our previous studies we used the characteristic features of the musculature to clarify the homologies of some male pregenital sclerites in the Muscidae. The homologies of the pregenital sclerites in members of different subfamilies of the Muscidae, in particular the nature of tergite VI, sternites VI and VII, syntergosternite VII + VIII, and of the hypandrial appendages, was confirmed by analysis of the muscle connections (Ovtshinnikova et al. 2018, 2019). It was confirmed that syntergosternite VII + VIII in the Muscidae consists of tergites VII and VIII, and at least part of sternite VIII.

The results on the muscles of the Azeliinae and their connections with the sclerites support our previous conclusions about the presence and the order of certain genital sclerites: tergite VI – syntergosternite VII + VIII – epandrium (tergite IX); sternite VI – sternite VII – syntergosternite VII + VIII – hypandrium (sternite IX).

The structure of the sclerites of the terminal segments in the examined species of Azeliini (Drymeia, Thricops, and Hydrotaea) is very similar. The main differences are the shape and the degree of development of some sclerites, for example the extent of sclerotization of the connection point of the muscle TSM 7 r. Unlike other examined species of Azeliini, both Hydrotaea ignava (Harris, 1780) and H. aenescens (Wiedemann, 1830) have a very strongly sclerotized distiphallus, very small ejaculatory apodeme, and tergite VI divided into two sclerites.

The pregenital and genital musculature in all the examined Azeliini is also very similar. The main differences are in some features of the connection points of the muscles and in their development. In members of different genera, the muscle ISM 6 extending to sternite VII can connect with the lateral margin of sternite VI or the adjacent membrane with sternite VI (desclerotized part of sternite VI). In H. dentipes, the muscles of phallapodeme M 2¹ extending from the hypandrial arms are much more developed than in Drymeia and Thricops.

The study of the sclerites and muscles of the terminal segments of the Reinwardtiini (Muscina stabulans, M. levida) has shown some differences from the Azeliini. Sternite VI is completely membranous. Nevertheless, as in the Azeliini, muscles ISM 5 in Muscina extend in two layers, one above the other, from sternite V to the membrane between sternite V and sternite VII (membranous sternite VI) and to sternite VII at its connection with the membrane of sternite VI. However, muscles of the lower layer of the Reinwardtiini are less powerful and spread along the membrane. The joints of the phallapodeme muscles M 2¹ are different between the Azeliini and the Reinwardtiini. In particular, in the Azeliini these muscles extend from the hypandrial arms close to the base of the epiphallus and they are opposite the epandrium, whereas in the Reinwardtiini M 2¹ extend from the hypandrium, i.e. are on the other side of the phallapodeme. In addition, unlike the large pregonites and postgonites of the Azeliini these structures are smaller in the Reinwardtiini. Compared with other Muscidae previously examined by us, the ejaculatory apodeme in M. stabulans and M. levida is a very large sclerotized plate located inside syntergosternite VII + VIII and the wide and powerful constrictors of ejaculatory apodeme M 23 extend from one margin of this plate to the other. The same genitalic structures of the Reinwardtiini, in particular the small pregonites and postgonites, and the very large ejaculatory apodeme were described in the Neotropical genus Callainireinwardtia by Savage (2009). However, in the genus Passeromyia of the Reinwardtiini the ejaculatory apodeme is small, not enlarged, but pregonites and postgonites are also small (Pont 1974). Since the structures of the male genitalia in this tribe were studied in fragments and not for all genera, it is currently difficult to say how much the ejaculatory apodeme size is an important feature of the tribe. Some molecular data demonstrated the differences between the Azeliini and the Reinwardtiini where the Reinwardtiini is not sister-group of the Azeliini (Schuehli et al. 2007; Kutty et al. 2014; Haseyama et al. 2015). In these analyses the affinity between most genera of the Reinwardtiini and Cyrtoneurininae has been shown and only Reinwardtia has been related with the Azeliini. The authors of these works suggested that the Reinwardtiini are polyphyletic, but we cannot confirm or refute it in present work because only one genus was studied by us. Thus, further study of these structures as well as the genital and pregenital muscles of different genera of the current Reinwardtiini can confirm polyphyly of the Reinwardtiini and change the position of some genera in the Muscidae and probably this tribe itself.

In our earlier papers on the study of muscles in the subfamilies Muscinae and Mydaeinae, we suggested that the features of the genital musculature in Scathophagidae were basal (plesiomorphic) (Ovtshinnikova et al. 2018, 2019). In these papers, the reduction tendencies in the structure of the sclerites and the genital and pregenital musculature in the Muscinae as compared with the Mydaeinae, basically the pregenital sclerites and phallapodeme, were also found. In the Muscinae syntergosternite VII + VIII is narrow, while it is wider (less strongly reduced and membranous) in the Mydaeinae. Correspondingly, the pregenital muscles are paired and better developed in the Mydaeinae. In the Azeliinae as compared with the Muscinae and Mydaeinae, the pregenital sclerites as well as the pregenital muscles extending from syntergosternite VII + VIII to the epandrium (M 19) and to the hypandrium (M 18) are very well developed. In particular, members of the Azeliinae have three well-developed muscles M 19 and a long M 18, whereas in members of the other subfamilies only one pair of M 19 and a small M 18 were found. Since the set of the pregenital muscles is a stable feature compared with the sclerotization of sternite VI, the complete set of pregenital muscles, in addition to the presence of non-membranous sclerites (joint of the muscle TSM 7 r) close to syntergosternite VII + VIII (a demonstration of a lesser reduction of the pregenital sclerites), indicates the basal position of the Azeliinae in the Muscidae.

In the Muscidae, differences in the development of hypandrial appendages, parts of the phallus (epiphallus, distiphallus, basiphallus) and their musculature as a result of the reduction processes, as well as their homologies, were noted earlier (Ovtshinnikova et al. 2018, 2019). These appendages are of great importance in copulation. In the studied genera of the Azeliinae, the pregonites, postgonites, and epiphallus are well developed.

The main difference in the genital musculature of the Azeliinae (Reinwardtiini and Azeliini) compared with other subfamilies of the Muscidae previously examined by us is the presence of the same large set of the phallapodeme muscles, specifically four pairs of phallapodeme muscles: M 1, M 2¹, M 2², M 2³ (from the hypandrium, pregonites and the epiphallus). The set of the phallapodeme muscles in different subfamilies of the Muscidae is different, but in the Mydaeinae and Muscinae there are only two pairs of the muscle M 2: in the Muscinae M 2¹, M 2² (from the hypandrium and pregonites) and in the Mydaeinae M 2², M 2³ (from pregonites and epiphallus). Since the set of phallapodeme muscles is a stable feature, the complete set of phallapodeme muscles (M 1, M 2¹, M 2², M 2³) in the Azeliinae (as in Scathophaga) is considered to be the basal state and confirms the basal position of the subfamily Azeliinae within the entire family Muscidae. Moreover, our results confirms well separated subfamily Azeliinae (Carvalho 1989; Carvalho et al. 2005; Savage and Wheeler 2004; Kutty et al. 2014) but refuted newly proposed classification with only three subfamilies (Haseyama et al. 2015) because the Azeliinae and the Muscinae have different set of genital and pregenital muscles and the structure of genital and pregenital segments and they cannot be in one subfamilies as authors suggested.

Comparison of the genital skeleton and muscles in the studied species of Muscidae with those of Scathophaga stercoraria has shown that males of most Muscidae as well as S. stercoraria possess well-developed pregonites, postgonites and epiphallus. However Scathophaga has a larger set of muscles of the hypandrial complex: 4 pairs of phallapodeme muscles M 1, M 2¹, M 2², M 2³ and the ejaculatory apodeme muscle M 23, muscles M 41 extending from the hypandrium to the basiphallus processes, muscles M 42 extending from the pregonites to the hypandrium and 2 pairs of tergosternal muscles M 5. Within the Muscidae, the set of phallapodeme muscles varies among the subfamilies, but members of all the genera we have studied possess only 1 pair of tergosternal muscles M 5, and lack muscles M 41 and M 42.

Comparison of the phallapodeme muscles of Scathophaga with those of Delia platura by Hennig (1976) has shown that D. platura has the same phallapodeme muscles as in Scathophaga (in Hennig: M 35 – 37, M 38, M 40, M 41) and one extra pair of muscles extending from the postgonites to the phallapodeme (in Hennig: M 39). In our opinion, the complete set of phallapodeme muscles in Scathophagidae and Anthomyiidae corresponds to the basal state, and the structure of the genital sclerites and muscles in Muscidae therefore reveals a certain degree of reduction. Our results about the relationships among studied families of the Calyptratae are congruent with the previous molecular hypotheses by Kutty et al. (2010, 2019).

The genital and pregenital modifications that we have detected in the Muscidae, in particular the reduction of pregenital sclerites and musculature, as well as the phallapodeme muscles, have thus allowed us to trace the progressive changes from the Azeliinae through the Mydaeinae to the Muscinae.

We would especially like to thank Adrian Pont (Oxford, UK) for discussion of this manuscript and for checking the English. We thank our reviewers Dr Bradley Sinclair (Ottawa, Canada) and Prof. Claudio Carvalho (Curitiba, Brazil) for significant comments on our manuscript. We also are very grateful to A.V. Barkalov (Novosibirsk) and O. Makarova (Moscow) for providing alcohol material from various regions of Russia.

The work of OGO was carried out at the Zoological Institute of Russian Academy of Sciences (State Research Program AAAA-A19-119020690082-8) and financially supported by the Russian Foundation for Basic Research (project 18-04-00354-a). The work of VSS was carried out at the Institute of Systematics and Ecology of Animals, Siberian Branch of the Russian Academy of Sciences, and financially supported by the Russian Foundation for Basic Research (project 18-04-00354-a) within the framework of the State Research Program VI.51.1.5 for 2013–2020 (AAAA-A16-116121410121-7).