A primitive sawfly, Madygella humioi sp. nov., belonging to the family Xyelidae (Hymenoptera: Symphyta), is newly described from the Upper Triassic Mine Group, Yamaguchi Prefecture, southwest Japan. The new species differs from the five previously known Madygella species in having a cell length of 1r plus 2r shorter than that of 3r + 4r and a cell height of 3r + 4r lower than 2r plus pterostigma in a forewing. To date, this is the oldest fossil record of sawflies in East Asia. Regarding genus Madygella, this is the first example found outside of the Kyrgyz Republic. This discovery provides an insight into the early evolution of the order Hymenoptera and suggests a widespread distribution of the pioneering genus Madygella during the Triassic period.
Introduction
The oldest representatives of the order Hymenoptera are xyelid sawflies, belonging to the family Xyelidae. Their fossil records include specimens from Middle to Upper Triassic deposits in the Kyrgyz Republic (Rasnitsyn, 1964, 1969; Kopylov, 2014) and from Upper Triassic deposits in Australia (Riek, 1955; Grimaldi and Engel, 2005), Argentina (Lara et al., 2014), and South Africa (Schlüter, 2000). The Xyelidae family comprises four subfamilies: Xyelinae, Macroxyelinae, Archexyelinae, and Madygellinae. The extinct subfamily Madygellinae was a pioneering group of sawflies restricted to the Triassic period, comprising eight different species belonging to three genera occurring from the Middle to Upper Triassic in the Kyrgyz Republic (Rasnitsyn, 1969; Kopylov, 2014). Their existence in other areas is not known.
The Upper Triassic Mine Group, belonging to the Inner Belt of southwest Japan, yields abundant insect fossil assemblages (Fujiyama, 1973, 1991; Ueda, 1991; Figure 1). Fujiyama (1973) firstly described some parts of the orders Blattodea, Parapalecoptera, Hemiptera, and Coleoptera from this Group. Subsequently, examples of the orders Mecoptera (Ueda, 1991), Odonata (Fujiyama, 1991), Diptera, Ephemeroptera, Hymenoptera, Neroptera, Orthoptera, Plecoptera, and Trichioptera (Takahashi et al., 1997) were reported. More than half of specimens described consists of wings of individuals of Coleoptera and Blattodea (Fujiyama, 1991), while the occurrence of Diptera is rare and important for the Triassic (Kopylov, personal comm.).
The insect fauna of the Mine Group is closely related to those of the Mount Crosby Formation in Australia, the Madygen Formation in the Kyrgyz Republic, the Djam Djun Formation in Vietnam, and the Yan-chen Formation in Inner Mongolia (Fujiyama, 1973). Particularly, the Madygen Formation yields one of the richest Triassic insect fauna comprising 20 orders (Shcherbakov, 2008), and among them, 13 orders are common to the Mine Group.
From the Mine Group, more than 6,000 specimens have been collected since the initial discovery. Most specimens are stored at the Mine City Museum of History and Folklore, Mine City, Yamaguchi Prefecture, Japan. However, these samples remain largely to be exploited in details.
This paper focuses on the first discovery of sawflies in Triassic deposits of East Asia and sheds light on the early evolution of sawflies in this region. It provides the first example of sawflies from the pioneering genus Madygella found outside of the Kyrgyz Republic.
Geological setting and fossil locality
The Triassic Mine Group, a thick clastic sequence, is lithologically divided into three units: the Hirabara, Momonoki, and Aso formations in ascending order in the Omine area, Mine City, Yamaguchi Prefecture, southwest Japan (Katayama, 1939; Kametaka, 1999; Figure 1). The lowermost Hirabara Formation is a marine deposit with a thickness of 250–1,000 m, which consists of conglomerate, sandstone, and mudstone intercalated with coal seams containing abundant marine bivalves (Hase, 1950; Tokuyama, 1960; Kametaka, 1999). The Momonoki Formation is a nonmarine deposit with a thickness of 800–1,200 m, which is composed of conglomerate, sandstone, and mudstone intercalated with coal seams (Takahashi and Mikami, 1975; Kametaka, 1999). The uppermost Aso Formation is a marine deposit with a thickness more than 1,500 m and is composed of sandstone and mudstone with marine bivalves (Hase, 1950; Tokuyama, 1958; Kametaka, 1999). Some bivalves (Tokuyama, 1960), ammonites (Ishibashi et al., 1990), plants (Katayama, 1939; Takahashi, 1973), and vertebrates (Hasegawa and Ota, 1975; Uyeno et al., 1996) have been reported from the Mine Group. The geological age of the Momonoki Formation is assigned to the Carnian stage of the Upper Triassic (Fujiyama, 1973; Kametaka, 1999).
The material reported in this paper was collected from the Momonoki Formation exposed along Route 435 at Okubata, Mine City, Yamaguchi Prefecture (Figure 1). From the same locality, Fujiyama (1991) and Ueda (1991) previously reported some fossil insects. The specimen described in the present study (MMHF3-00091) is stored at the Mine City Museum of History and Folklore.
Abbreviations of morphological terms utilized in this paper are as follows (according to Rasnitsyn, 1969; Lara et al., 2014): C, costal vein; SC, subcostal vein; Pt, pterostigma; R, radial vein; RS (1-RS, 2-RS, 3-RS, 4-RS, 5-RS), radial sector and its sections; M (1-M, 2-M, 3-M, 4-M, 5-M), medial vein and its sections; Cu (1-Cu, 2-Cu, 3-Cu, 4-Cu), cubital vein and its sections; r (1r, 2r, 3r, 4r), radial cells; rm (1rm, 2rm, 3rm), radial-medal cells; mcu (1mcu, 2mcu, 3mcu), medio-cubital cells; cua (1cua, 2cua), cubito-anal cells; r-rs (1r-rs, 2r-rs), crossveins between R and RS; r-m (1r-m, 2r-m, 3r-m), radial-medal crossveins; m-cu (1m-cu, 2m-cu), medio-cubital crossveins; cu-a (1cu-a), cubito-anal crossveins.
Systematic description
Order Hymenoptera Linnaeus, 1758
Suborder Symphyta Linnaeus, 1758
Superfamily Xyeloidea Newman, 1834
Family Xyelidae Newman, 1834
Subfamily Madygellinae Rasnitsyn, 1969 emend.
Kopylov, 2014
Type genus.—Madygella Rasnitsyn, 1969.
Diagnosis of subfamily.—Small sawfly. Forewings 2.8–5.8 mm long according to reconstructed proportions. Costal space narrow. C thickened toward apex, sometimes almost reaching R, but never touching it, separated from Pt. SC weak, joining only C, sometimes preserved only at wing base, transformed into fold, or completely lost. Pt usually not sclerotized, veins bordering it usually thickened. R straight up to base of RS, with slight bend near point of divergence of RS, usually straight within cell 1r, less often arcuate anteriad, without gap at base of Pt. RS1 (anterior branch of RS) lost without traces. 1r-rs not longer than 2r-rs. Cell 1r along anterior margin subequal in length to 2r. 1-RS either shorter or longer than 1-M. M + Cu arcuate (very rarely slightly S-shaped). Cell 1mcu long and narrow, at least twice as long as wide (Rasnitsyn, 1969, p. 50–51; Kopylov, 2014, p. 611).
Generic composition.—The subfamily consists of three genera: Madygella Rasnitsyn, 1969; Samarkandykia Kopylov, 2014; and Chubakka Kopylov, 2014 from the Middle-Upper Triassic of Kyrgyzstan.
Comparison.—The subfamily Madygellinae differs from other xyelid groups in the absence of RS1 (anterior branch) branches with rare exceptions: Xyela alberta (Curran, 1923) and Lydoxyela excellens (Rasnitsyn, 1966). The Madygellinae differs from subfamilies Xyelinae and Macroxyelinae in having the long cell 2r, and (for the majority of representatives of these two subfamilies) in the narrow costal space. It also differs from subfamily Xyelinae in having the apically thickened C and from subfamily Macroxyelinae in the absence of characteristic sclerotization at the base of the Pt.
Remarks.—The original diagnosis (Rasnitsyn, 1969) was considerably expanded by Kopylov (2014). We add this study as a new locality to the distribution.
Genus Madygella Rasnitsyn, 1969
Madygella Rasnitsyn, 1969, emend. Kopylov, 2014
Type species.—Madygella analoga Rasnitsyn, 1969.
Revised generic diagnosis.—1-RS shorter than RS + M and usually shorter than 1-M, rarely slightly longer. Length of cell 1r plus 2r longer than that of 3r + 4r, sometimes length of cell 1r plus 2r shorter than that of 3r + 4r; height of cell 3r + 4r almost as high as that of 2r plus Pt, sometimes height of cell 3r + 4r lower than that of 2r plus Pt. RS curved at point of 2r-m joining it.
Species composition.—Madygella genus consists of five species: M. analoga Rasnitsyn, 1969; M. levivenosa Kopylov, 2014; M. kurochkini Kopylov, 2014; M. aristovi Kopylov, 2014; M. bashkuevi Kopylov, 2014.
Comparison.—The present genus differs from Samarkandykia in 1-RS shorter than 1-M, RS curved at point of 2r-m joining it and 3r-m longer (up to 1.3 times) than 2r-m. The genus also differs from Chubakka in 2r-rs diverging in apical one-third of Pt, cell 1mcu weakly dilating toward apex.
Remarks.—Most characters of Madygella humioi sp. nov. match with the previous generic diagnosis. Unlike other Madygella species, however, M. humioi sp. nov. shows cell length of 1r plus 2r shorter than that of 3r + 4r and cell height of 3r + 4r lower than 2r plus Pt as described below. Taking these features into consideration, the generic diagnosis is slightly revised and updated here.
Madygella humioi Oyama and Maeda, sp. nov.
Figure 2
Etymology.—In honor of Humio Takahashi, director of the Mine City Museum of History and Folklore.
Holotype (MMHF3-00091).—A specimen of well preserved forewing without anal area from the Carnian deposits of the Momonoki Formation at Okubata, Omine area, Mine City, Yamaguchi Prefecture, southwest Japan (Figure 1). It is stored at the Mine City Museum of History and Folklore. Impression of other insects is overprinted on the surface.
Diagnosis.—Madygella species having a cell length of 1r plus 2r shorter than that of 3r + 4r, and a cell height of 3r + 4r lower than 2r plus Pt in a forewing.
Description.—SC is developed over the entire length, joining C basal to RS base. Costal space is extremely wide. R at the base of RS has a weak curve, slightly dilates; within cell 1r, R is straight. The Pt is not sclerotized; the veins running along the periphery of the Pt are thickened. 2r-rs is inclival, shorter than 4-RS, 1.3 times longer than 1r-rs; the base of 2r-rs is 2.6 times closer to the apex of the Pt than to 1r-rs. 1-RS is much shorter than 1-M; RS + M is about 2 times longer than 1-RS. 2-RS is straight (2-RS joins RS + M). 5-RS is 3.1 times longer than 4-RS. 2r-rm is strongly inclined, 1.9 times longer than 2r-rs. 3r-m is probably longer than 2r-m. 1m-cu is subequal in length to 3-Cu. The free end of M and Cu are lost; 3r-m gradually passes into 5-M; 2m-cu gradually passes into 4-Cu. 3-M has a weak curve at the base of 3-M.
Measurements.—Forewings impression length, 4.5 mm; width, 2.3 mm.
Comparison.—The newly described species differs from subfamily Madygellinae in its extremely wide costal space.
It is similar to the type species Madygella analoga and M. levivenosa in developing SC over the entire wing length. However, M. humioi sp. nov. differs from the type species in non-sclerotized Pt, strongly inclined 2r-m, same length of both 3-Cu and 1m-cu, absence of a free end of M with bend and slightly curved 3-M. Madygella humioi sp. nov. differs from M. levivenosa in inclival 2r-rs, strongly inclined 2r-m, same length of both 3-Cu and 1m-cu and distinctly curved basal 3-M.
The newly described species is also similar to Madygella bashkuevi in strongly inclined 2r-m and curved 3-M. Strong 2r-m and curved 3-M are peculiar characteristic features to M. bashkuevi and M. humioi sp. nov. On the other hand, the new species is clearly distinct from M. bashkuevi in SC joining C basal to RS base and subequal 2r-rs and 1r-rs.
Figure 2.
Forewing of Madygella humioi sp. nov., holotype MMHF3-00091 from the Upper Triassic Mine Group. A, photograph of the horizontal view; B, line drawing based on A with abbreviated terminology of vein and cell. Scale bars, 1 mm.
Table 1.
Comparison of forewing characters in selected Madygella species. –, undescribed.
Figure 3.
Insect fossils from the Mine Group not showing remarkable distortion. A, two partly overlapped wings of a sawfly (Wing P, Q, MMHF3-00095) from the same horizon of Madygella humioi sp. nov. B, outline of A with measurement axes and crossed-axes angle between the wings P and Q (67°). C, body parts of Coleoptera (MMHF3-00092) from the same horizon of Madygella humioi sp. nov. The body was vertically collapsed and flattened by compaction, but the planar geometry is not secondarily distorted and well preserved. D, measurements of B. Note aspect ratios of both wings are not disparate (3.00–3.35) even though orientations of the two wings are different. Scale bars, 5 mm.
Discussion
The fossil records of Triassic Hymenoptera were previously found from four regions: Kyrgyz Republic (Rasnitsyn, 1964, 1969; Kopylov, 2014), Queensland (Riek, 1955; Grimaldi and Engel, 2005), South Africa (Schlüter, 2000), and Argentina (Lara et al., 2014). Although East Asia has several contemporaneous fossil insect localities (Zheng et al., 2018), there was no previous record of individuals belonging to the order Hymenoptera in the region. Madygella humioi sp. nov. is the oldest record of sawflies in East Asia, including Japan. This discovery reveals that primitive Hymenoptera (xyelid sawflies) were already widespread in five isolated realms, and further expands our knowledge regarding the early evolution and diversity of sawflies during the Triassic period.
The taxonomic study of Triassic sawflies has been largely based on wing morphology as wings possess a higher potential for fossilization than other body parts (Grimaldi and Engel, 2005; Schlüter, 2000). However, the simple wing morphology is not always adequately diagnostic (Lara et al., 2014), in part due to the potential post-burial deformation of morphological characters (Kopylov, 2014). For example, Madygella bashkuevi from the Middle–Upper Triassic in the Kyrgyz Republic has inclined 2r-m and curved 3-M, but these could have resulted from secondary distortion by shear stress of tectonism (Kopylov, 2014). Thus, Kopylov (2014) adopted a standardizing proportion formula to reconstruct the original wing geometry.
Madygella humioi sp. nov. from the Mine Group also shows inclined 2r-m and curved 3-M (see Table 1). Taking Kopylov's results into consideration, we examined the mode of secondary distortion in our samples. Many insect wings from the Mine Group are variously scattered on a bedding plane without showing any preferred orientation. Therefore, the geometric disparity would be expected depending on their orientation if secondary distortion occurred. However, a close taphonomic observation reveals that most insect fossils in the Mine Group are not remarkably distorted, unlike Kyrgyz Republic samples. For example, two partly overlapped wings of a sawfly are preserved in a bedding plane of the shale (Figure 3A). Their crossed-axes angle is quite large, i.e., 67° (Figure 3B), while remarkable geometric disparity is not recognized between the two wings. Their aspect ratios are not disparate (3.00–3.35) even though orientations are different (Figure 3D). The minimum secondary distortion is also suggested by the preservational state of an intact fossil beetle (Coleoptera) found from the same horizon as M. humioi sp. nov. (Takahashi et al., 1997, pl. 8, fig. 3C). The body of the beetle was vertically collapsed and flattened by compaction, while the planar geometry was still intact and not remarkably distorted (Figure 3C). The line of evidence suggests that the inclined 2r-m and curved 3-M shown in M. humioi sp. nov. can be interpreted as an essential morphological character, rather than a secondary distortion.
Acknowledgements
The authors are highly indebted to Humio Takahashi (Mine City Museum of History and Folklore) and Kenji Shinoda (Mine City) who allowed to borrow the material for studying, and to Koji Wakita (Yamaguchi University) for his valuable comments and help in the field work. Constructive critical comments by Aleksandr P. Rasnitsyn and Dmitry Kopylov were very helpful for improvement of the draft. The authors would like to express their sincere gratitude toward Yusuke Okazaki (Kyushu University), Munetoshi Maruyama (The Kyushu University Museum), and Toshiharu Mita (Kyushu University) for their valuable suggestions and comments. We also thank Hirokazu Yukawa (Fukui Prefectural Dinosaur Museum) for allowing us to study fossil insects from the Mine Group. In addition, we would like to thank members of the Mine Fossil Museum and Mine City Museum of History and Folklore; and Syougo Yoshinaga, Yukiko Arakawa, Fumiko Inakura, and Yuko Kazitani for their support during the field work. This study was conducted with the support of The Late Professor Tatsuro Matsumoto Scholarship, and was partly supported by a JSPS KAKENHI Grant (number JP18K03827).
