Today, Chironomidae have a worldwide distribution, with more than 7,000 species (Pape et al. Reference Pape, Blagoderov, Mostovski and Zhang2011). This family also has a rich fossil record, with the oldest finding, Aenne triassica Krzemiński & Jarzembowski, Reference Krzemiński and Jarzembowski1999 dating back to the Upper Triassic (∼202 Ma; Benton & Donoghue Reference Benton and Donoghue2007). Non-biting midges are commonly recorded in ambers of various origin: from the Cretaceous (e.g., in Lebanese amber; Veltz et al. Reference Veltz, Azar and Nel2007), through the Eocene (e.g., in Sakhalin amber; Baranov et al. Reference Baranov, Andersen and Perkovsky2015) to the Miocene (e.g., in Amazonian amber; Antoine et al. Reference Antoine, De Franceschi, Flynn, Nel, Baby, Benammi, Calderón, Espurt, Goswami and Salas-Gismondi2006). The vast majority of species of the chironomid tribe Tanytarsini have so far been described from extensively examined Baltic amber (∼43–48 Ma; Seredszus & Wichard Reference Seredszus and Wichard2007; Giłka Reference Giłka2010, 2011a; Giłka et al. Reference Giłka, Zakrzewska, Dominiak and Urbanek2013; Zakrzewska & Giłka, Reference Zakrzewska and Giłka2014, Reference Zakrzewska and Giłka2015a, Reference Zakrzewska and Giłkab; Zakrzewska et al. Reference Zakrzewska, Krzemiński and Giłka2016), whereas the oldest known member of this tribe is Nandeva pudens Giłka, Zakrzewska, Baranov, Wang & Stebner, Reference Giłka, Zakrzewska, Baranov, Wang and Stebner2016, found in Chinese Fushun amber (∼50–53 Ma; e.g., Wang et al. Reference Wang, Rust, Engel, Szwedo, Dutta, Nel, Fan, Meng, Shi, Jarzembowski, Wappler, Stebner, Fang, Mao, Zheng and Zhang2014); however, this fossil belongs to the genus that is presumably either a member of the Tanytarsini or is defined as the tribe's sister group (Giłka et al. Reference Giłka, Zakrzewska, Baranov, Wang and Stebner2016).
Cambay amber is found in several active opencast lignite mines located in the state of Gujarat, India, about 30 km northeast of Surat. The amber occurs in lignitic and mud-dominated sediments, which were deposited in a near-shore chenier plain system where the sediment input was seasonal (e.g., McCann Reference McCann2010). Cambay amber has been dated to the early Eocene (54 Ma; cf. Smith et al. Reference Smith, Kumar, Rana, Folie, Solé, Noiret, Steeman, Sahni and Rose2016, Stebner et al. Reference Stebner, Szadziewski, Singh, Gunkel and Rust2017a); it is thus contemporaneous with Fushun amber from China and slightly older than amber from the Baltic region. Furthermore, Cambay amber was formed at a time of expanding modern diversity, at the beginning of the Early Eocene Climatic Optimum (EECO) and just after the Late Palaeocene Thermal Maximum (LPTM) (e.g., Zachos et al. Reference Zachos, Dickens and Zeebe2008).
As in many other amber deposits, Chironomidae are the most common dipteran inclusions in Cambay amber, in which five subfamilies have been found so far. With nearly 50 % of all chironomids recorded, Chironominae (including the tribe Tanytarsini) is the most abundant, followed by Orthocladiinae, Tanypodinae, Podonominae and Prodiamesinae. The Tanytarsini is represented by three extant genera: Stempellina Thienemann & Bause in Bause, Reference Bause1913; Stempellinella Brundin, Reference Brundin1947; and Tanytarsus van der Wulp, Reference Wulp van der1874 (Stebner et al. Reference Stebner, Baranov, Zakrzewska, Singh and Giłka2017b); however, no specific description has been published so far. Therefore, a species of an unknown extinct genus presented below is the first tanytarsine and non-biting midge described from early Eocene Cambay amber.
1. Material and methods
The specimens examined are adult male chironomids preserved in early Eocene Cambay amber from India (54 Ma), derived from the Tadkeshwar lignite mine. The inclusions are part of the collection of the Birbal Sahni Institute of Palaeosciences (BSIP), Lucknow, India. The amber was ground using a Buehler Phoenix Beta grinding machine and polished manually. Measurements of specimens are in micrometres, except for the total length (in millimetres, rounded off to the first decimal digit). The body length was measured from the antennal pedicel to the end of the gonostylus, and the wing from the arculus to the tip. Lengths of leg segments and palpomeres were rounded off to the nearest 5 μm and 1 μm, respectively. The antennal, leg and venarum ratios (AR, LR, VR) were calculated to the second decimal place. The morphological terminology and abbreviations follow Sæther (Reference Sæther1980). The photographs were taken using the microscope PZO Biolar SK14 and Helicon Focus 6 image stacking software.
2. Systematics
Family Chironomidae Newman, Reference Newman1834 Subfamily Chironominae Newman, Reference Newman1834 Tribe Tanytarsini Zavřel, Reference Zavřel1917 Subtribe(?) Tanytarsina Zavřel, Reference Zavřel1917 Gujaratomyia Giłka & Zakrzewska, gen. nov.
Type species. Gujaratomyia miripes Giłka & Zakrzewska, sp. nov. (by present designation and monotypy).
Etymology. After Gujarat, the state in Western India, where the Cambay amber deposits are located.
Diagnosis. Eyes bare, reniform. Antenna with 13 flagellomeres. Tibiae of all legs similarly short, much shorter than their femora; tarsus of mid leg extraordinarily long, with basitarsus much longer than tibia. Fore leg tibia with pair of stout apical bristles; mid and hind leg tibia with strongly elongated apical lobe, evenly tapering to bristle-like distal part, with apex bent, filiform. Gonostylus longer than gonocoxite; digitus absent; stem of median volsella short, apically branched into several lamellae.
Gujaratomyia miripes Giłka & Zakrzewska, sp. nov.
Holotype. Adult male (tarsi of left mid and both fore and hind legs missing) preserved in 10×6×2.5 mm piece of amber (early Eocene, 52–53 Ma, Tadkeshwar mine, Gujarat State, India; BSIP Tad-607 a; Fig. 1A).
Figure 1 Gujaratomyia miripes gen. et sp. nov., adult male, holotype, BSIP Tad-607 a, early Eocene, 52–53 Ma, Tadkeshwar mine, Gujarat, India: (A) inclusion in amber; (B) habitus; (C) antenna (arrows indicate borders between flagellomeres, fm1–13); (D) wing.
Paratype. Adult male, incomplete specimen (only distal part of abdomen, including hypopygium, preserved) as syninclusion with the holotype (BSIP Tad-607 b). Other syninclusion: Chironominae, one female (BSIP Tad-607 c).
Etymology. The specific epithet is referred to the unusual leg structure described below.
Diagnosis. As for the genus.
Description. Adult male. Total length c.1.5 mm; wing length c.820 μm.
Head (Fig. 1B, C). Eyes bare, reniform, without dorsomedian extensions. Frontal tubercles conical, c.20 μm long. Antenna with 13 well discernible flagellomeres, AR 0.77, plume fully developed (Fig. 1C). Length of palpomeres 2–5: c.30 μm, 103 μm, 117 μm, 117 μm (ultimate palpomere slightly deformed). At least 16 clypeals.
Thorax chaetotaxy. Ac at least 16, Dc at least 12, Pa at least 3, Scts at least 8.
Wing (Fig. 1D). Slender, broadest at two-thirds of length, width: 220 μm, length/width ratio 3.73. Sc and R2+3 not observed. RM as parallel continuation of R4+5, slightly oblique relative to M. An long, ending well distally of cubital fork. FCu placed distinctly distally of RM, VRCu c.1.65. Veins ending as follows (in order from base to tip): An, Cu1, R1, M3+4 and R4+5, M1+2. Anal lobe weak. Squama bare. Wing, including membrane covered with macrotrichia in distal half at least.
Legs (Figs 1B, 2A–E). Coxae of mid and hind legs bearing several strong bristles (Fig. 1B). Tibiae of all legs similarly short, much shorter than their femora; tarsus of mid leg (the only preserved in the specimen examined) extraordinarily long, with basitarsus much longer than tibia (Fig. 2A); for length of leg segments and leg ratios see Table 1. Fore leg tibia armed with pair of stout apical bristles c.140–180 μm long (Fig. 2B). Mid and hind leg tibia with strongly elongated apical lobe, evenly tapering to bristle-like distal part, apex bent, filiform; each lobe bearing pair of subapical setae; length of whole structure c.75 μm (Fig. 2C–E). Sensilla chaetica on ta1 of p2 not observed. Pulvilli absent.
Figure 2 Gujaratomyia miripes gen. et sp. nov., adult male, holotype, BSIP Tad-607 a, early Eocene, 52–53 Ma, Tadkeshwar mine, Gujarat, India: (A) mid and hind leg (fe=femur; ti=tibia; ta1–5=tarsomeres 1–5 of mid leg; arrows indicate borders between leg segments); (B–E) tibiae and tibial armature of fore (B) and hind leg (C–E).
Table 1 Leg segment lengths (μm) and leg ratios of male Gujaratomyia miripes gen. et sp. nov. Abbreviations: p1–p3=legs of pair 1–3; fe=femur; ti=tibia; ta1–ta5=tarsomeres 1–5; LR=leg ratio.
Hypopygium (Fig. 3). Gonostylus 85 μm long, longer than gonocoxite, straight, with several long setae on apex. Anal tergite semicircular, with dense setae surrounding base of anal point. Anal point long, tapering to bevelled apex with small subapical pit (Fig. 3A, B), upturned in lateral view (Fig. 3C). Superior volsella stout, parallel-sided in proximal part, oblique relative to main body axis, with distal part slightly swollen and medially directed, bearing three long setae placed on distinct protuberances on median margin (Fig. 3A–E). Digitus absent. Stem of median volsella straight and short (c.12 μm), apically branched into one short and three longer lamellae (Fig. 3D, F, G). Inferior volsella arcuate, with enlarged apical part armed with dense setae.
Figure 3 Gujaratomyia miripes gen. et sp. nov., adult male, early Eocene, 52–53 Ma, Tadkeshwar mine, Gujarat, India: (A–C) paratype, BSIP Tad-607 b, hypopygium and its structures in dorsal (A–B) and lateral (C) aspects. (D–G) holotype, BSIP Tad-607 a, hypopygium and its structures in ventral aspect: (E) superior volsella, magnified ca.6 times relative to (D); (F) median volsella, magnified ca.3 times relative to (D); (G) median volsella, magnified ca.6 times relative to (D).
3. Discussion
3.1. Systematic position of Gujaratomyia
Several unique characters of the legs and hypopygium have not been found in any extant/fossil chironomid and support our concept of erecting the new genus Gujaratomyia (see diagnosis). The slender wing with the reduced anal lobe, the RM vein placed as a parallel continuation of R4+5, the bare squama and the membrane covered with macrotrichia (Fig. 1D) form a set of characters typical of the Tanytarsini and testify to the position of Gujaratomyia within this tribe (cf. Giłka et al. Reference Giłka, Zakrzewska, Baranov, Wang and Stebner2016). Although the subtribal placement of the new genus remains open, a combination of characters (the reniform eyes without dorsomedian extensions; the antenna with 13 flagellomeres (Fig. 1C); and, above all, the stem of the hypopygial median volsella branched apically into lamellae (Fig. 3F, G)) suggests that Gujaratomyia and Cladotanytarsus Kieffer, Reference Kieffer1921a may be members of a common group within the subtribe Tanytarsina. Apart from the characters listed in the diagnosis and defined as apomorphies for Gujaratomyia (see also below), these two genera differ from each other in the absence/presence of the hypopygial digitus and in the gonocoxite–gonostylus length proportions.
3.2. Legs in the adult male of Gujaratomyia
3.2.1. Length proportions
In adult males of the Chironomidae, the fore legs are the longest and the mid legs the shortest, since the five-segmented tarsus of the fore leg is elongated the farthest, whereas the tarsomeres of the mid leg are the shortest. The rule of mid leg tarsomeres being strongly shortened applies, in particular, to species that are flightless or have only limited flying ability (Sæther Reference Sæther1971; Giłka Reference Giłka2011b, Reference Giłkac) (see below). In the Tanytarsini, the leg ratios LR1–3, calculated as the ratio of the length of the basitarsus to that of the tibia of the same leg, are as follows: LR1 > 1, LR2<1, LR3<1. In all the Tanytarsini, as in many chironomids, the fore leg femur is much longer than the tibia (fe1≫ti1), whereas the femora and tibiae of the mid and hind legs are approximately of the same length (fe2≈ti2, fe3≈ti3). The male of Gujaratomyia examined here diverges distinctly from these rules: the tarsomeres of the mid leg (especially the basitarsus) are strongly elongated and LR2 is more than 2.5. Interestingly, the femora of the mid and hind legs are more than twice the length of the tibiae (fe2≫ti2, fe3≫ti3), the proportions being much the same as those of the fore leg (Figs 1B, 2A; Table 1). The total length of the mid legs in male Gujaratomyia (unfortunately, the only one preserved intact in the specimens examined) is also relatively high, when compared with that in males of Eocene species of a similar body and wing lengths (Table 2; cf. Zakrzewska & Giłka Reference Zakrzewska and Giłka2015a, Reference Zakrzewska and Giłkab).
Table 2 Lengths of mid leg (μm), body (mm) and wing (μm) of male Gujaratomyia miripes gen. et sp. nov. and several comparable Eocene species of the tribe Tanytarsini.
ain Zakrzewska & Giłka Reference Zakrzewska and Giłka2015a; bin Zakrzewska & Giłka Reference Zakrzewska and Giłka2015b
The strong elongation of the legs may be an expression of adaptations to non-typical mode of life and/or habitat, known in, for example, the related subfamily Orthocladiinae. The brachypterous Bryophaenocladius chrissichuckorum Epler, 2012, known from its ground mating behaviour, has long but stocky legs (Epler 2012). The fully-winged Troglocladius hajdi Andersen, Baranov & Hagenlund, Reference Andersen, Baranov, Hagenlund, Ivković, Kvifte and Pavlek2016 (in Anderson et al. Reference Andersen, Baranov, Hagenlund, Ivković, Kvifte and Pavlek2016), found in Croatian caves, features exceptionally long legs, the proportions of which (including the LR ratios), however, being within the range diagnostic for the subfamily; such legs, called “feelers”, are considered to be a part of a well-developed sensory system and an adaptation to live in caves (Andersen et al. Reference Andersen, Baranov, Hagenlund, Ivković, Kvifte and Pavlek2016). The leg and/or wing structure observed in the Orthocladiinae is not, however, fully comparable with those observed in Gujaratomyia, and cannot underpin a hypothesis of adaptations to a similar mode of life and/or habitat (see below).
In contrast to the long-legged chironomids, strongly shortened mid legs are observed in some of the Tanytarsini, for example Corynocera Zetterstedt, Reference Zetterstedt1838, Thienemanniola Kieffer, Reference Kieffer1921b and Tanytarsus tika (Tourenq, Reference Tourenq1975), in whose legs all the segments of the mid leg, and the tarsus in particular, are strongly shortened; nevertheless, the LR2<1, typical of the tribe, has been retained (Lehmann Reference Lehmann1973; Cranston Reference Cranston1980). The first two genera include brachypterous species, lacking the ability to fly; on the other hand, the non-typical leg length proportions in the winged male of T. tika are most likely an adaptation to being able to additionally inhabit the surface of the water or the ground. A parallel adaptation in the Tanytarsini mentioned above is the naked wing membrane, devoid of macrotrichia (cf. Cranston Reference Cranston1980). It is worth reiterating, however, that neither the rule of strong mid leg shortening, nor that of a reduced naked-wing membrane, applies to the male Gujaratomyia miripes.
3.2.2. Tibial armature
The tibiae of Tanytarsini imagines are armed with spurs and combs, the structure of which is a key generic character (i.a. Giłka Reference Giłka2011b). The tibia of the fore leg usually has a single spur but no combs. The anomaly of a double spur or comb occurring on the fore leg is known from both extant and fossil Tanytarsini (cf. Giłka & Paasivirta Reference Giłka and Paasivirta2008; Giłka et al. Reference Giłka, Zakrzewska, Dominiak and Urbanek2013; Zakrzewska & Giłka Reference Zakrzewska and Giłka2015a), whereas the presence of one or more stout bristles in place of the fore leg tibial spur is a diagnostic character recorded above all in fossil species of the genera Archistempellina Giłka & Zakrzewska, Reference Giłka, Zakrzewska, Dominiak and Urbanek2013, Corneliola Giłka & Zakrzewska, Reference Giłka, Zakrzewska, Dominiak and Urbanek2013 and Tanytarsus van der Wulp, Reference Wulp van der1874 (cf. Giłka Reference Giłka2010; Giłka et al. Reference Giłka, Zakrzewska, Dominiak and Urbanek2013; Zakrzewska et al. Reference Zakrzewska, Krzemiński and Giłka2016). This character appears also in Gujaratomyia (Fig. 2B).
The tibiae of the mid and hind legs of Tanytarsini are usually armed with two combs, consisting of teeth arranged in the shape of a fan. The combs may be separate or may form a uniform crown; only rarely they are absent (Giłka Reference Giłka2011b). There are usually two spurs on the mid and hind tibiae – one on each comb. They can be completely absent, or be present only on one of the two combs; sometimes they have a unique structure (Giłka Reference Giłka2011b, Reference Giłkac). In contrast to most Tanytarsini, combs and spurs do not occur in Gujaratomyia. In the male of this genus, the tibiae bear strongly elongated apical lobes tapering to a bristle-like distal part with a filiform apex (Fig. 2C–E). Somewhat similar structures, but never as these observed in Gujaratomyia, occur in some Tanytarsini lacking the (or with a limited) ability to fly, or showing a ground/water mating behaviour. Relatively small tibial lobes or pads can be observed in Cladotanytarsus in the subgenus Lenziella Kieffer, Reference Kieffer1922. However, the mid and hind legs in Lenziella are stocky, and the apical lobes of the tibiae are armed with dense setae, exceptionally with a single stout apical bristle (Giłka Reference Giłka2011c, fig. 8C). Nevertheless, all these are typical setae/bristles arising from tubercles, and are apparently not homologous with the bristle-like structures found in Gujaratomyia. Tanytarsus tika shows setose pads (Cranston Reference Cranston1980), similar to those found in Lenziella. The monotypic Thienemanniola also bears extensions on its mid and hind tibiae, along with a brush composed of the long setae, but their spurs are short, stout and strongly bent apically – similar to those known from Corynocera (Lehmann Reference Lehmann1973). However, the structures mentioned above, observed in the extant Tanytarsini, differ distinctly from those found in Gujaratomyia, the oldest extinct genus of this tribe.
3.2.3. Function
Despite the distinct leg structure suggestive of non-typical mode of life in the adult male of Gujaratomyia miripes, we see no reason to contend that the species' imago was limited in its flying ability, or occurred in a habitat not typical of the Chironomidae. Our opinion is based on the observation that: (1) the wing of Gujaratomyia is fully developed, and its membrane is covered by macrotrichia, a character typical of the flying Chironomidae; (2) the long and slender legs featured by Gujaratomyia are characteristic of flying species, as opposed to shortened and/or stout legs (and also palps and antennae) shown by brachypterous species; (3) Gujaratomyia lacks other characters typical of the chironomids with non-typical behaviour/habitat (e.g., caves) – Gujaratomyia has well-developed eyes, palps and antennae bearing fully plumose flagellum. We interpret the elongated legs in Gujaratomyia as a gliding-enhancement adaptation; enhancement of in-flight stimulus reception cannot be ruled out, either (cf. Andersen et al. Reference Andersen, Baranov, Hagenlund, Ivković, Kvifte and Pavlek2016). The non-typical tibial structures of Gujaratomyia most likely reflect adaptations somewhat similar to those found in the non-flying Tanytarsini, and may adapt the bearers to short-term presence on the water surface. However, it seems hardly plausible that they would function similarly on the ground surface.
In conclusion, we interpret the leg ratios and the tibial structures found in Gujaratomyia miripes as unique characters (these can be perceived as autapomorphies until the genus remains monotypic), which – in combination with the reniform broadly separated eyes, fully plumose 13-segmented antennal flagellum and several diagnostic hypopygial features – form a set not known from any other non-biting midge, and justify erection of the new genus.
3.3. Biogeography
Today, India harbours many endemic taxa, especially in the Western Ghats area (Myers et al. Reference Myers, Mittermeier, Mittermeier, Da Fonseca and Kent2000). In this context, Indian amber inclusions have proven to provide an important contribution to the discussion about India's complex geological history and the processes that shaped India's modern flora and fauna (summarised in Rust et al. Reference Rust, Singh, Rana, McCann, Singh, Anderson, Sarkar, Nascimbene, Stebner, Thomas, Solórzano-Kraemer, Williams, Engel, Sahni and Grimaldi2010). It has been shown that Indian amber inclusions show worldwide connections, whereas only minimal evidence for faunal isolation has been reported so far (Engel et al. Reference Engel, Grimaldi, Nascimbene and Singh2011, Reference Engel, Ortega-Blanco, Nascimbene and Singh2013; Grimaldi & Singh Reference Grimaldi and Singh2012; Grimaldi et al. Reference Grimaldi, Engel, Nascimbene and Singh2013a, Reference Grimaldi, Engel and Singhb; Rust et al. Reference Rust, Singh, Rana, McCann, Singh, Anderson, Sarkar, Nascimbene, Stebner, Thomas, Solórzano-Kraemer, Williams, Engel, Sahni and Grimaldi2010). Various biogeographic patterns of the Indian amber inclusions have been observed within the order Diptera (Rust et al. Reference Rust, Singh, Rana, McCann, Singh, Anderson, Sarkar, Nascimbene, Stebner, Thomas, Solórzano-Kraemer, Williams, Engel, Sahni and Grimaldi2010; Stebner et al. Reference Stebner, Szadziewski, Singh, Gunkel and Rust2017a, Reference Stebner, Singh, Rust and Grimaldic). At this stage of research, taxa that are endemic to the Indian subcontinent appear to be rather rare, and have been reported so far only from the families Keroplatidae (Vastaplatyura electrica Solórzano, Kraemer & Evenhuis, Reference Solórzano Kraemer and Evenhuis2008) and Psychodidae (Phlebotoiella eoindianensis Solórzano, Kraemer & Wagner, Reference Solórzano Kraemer and Wagner2009). Gujaratomyia gen. nov. now adds further evidence for the presence of endemic elements in Eocene India. The genus seems to be restricted to the Indian subcontinent, since neither fossil nor recent relatives are known from elsewhere. It must be considered, however, that the Tanytarsini are probably a rather young group, with the oldest members known from the early Eocene's Fushun amber (Giłka et al. Reference Giłka, Zakrzewska, Baranov, Wang and Stebner2016) and, as reported herein, from early Eocene Indian amber; a fact which doubtless has substantial implications for biogeographic patterns. In conclusion, many more studies on the Indian amber inclusions and their distributional patterns are needed to get a more complete picture of the origins and ranges of fossil as well as modern Indian insect faunas.
