The revision of fossil big-eyed bugs suggests a peculiar evolutionary history of a peculiar true bug family (Heteroptera: Lygaeoidea: Geocoridae)

The fossil record of the lygaeoid true bug family, Geocoridae so far consisted of a compression fossil from the Florissant Formation, Geocoris infernorum, and a mould fossil from the Izarra Formation, G. monserrati. Here, we report the third taxon, Protogeocoris arcanus gen. et sp. nov. from Cenomanian Burmese amber along with revision of the two previously mentioned species. The study of the specimens resulted in erecting a new genus, Eocenogeocoris gen. nov. to accommodate G. infernorum [ = Eocenogeocoris infernorum (Sudder, 1890) comb. nov.], because of its unique combination of characters compared to extant geocorine true bugs. Furthermore, the interpretation of the novel morphological data acquired from the study of the fossil geocorids in the context of the knowledge on extant representatives of the family suggests that characteristic groups of the family may have been evolved until the mid-Cretaceous, whilst the species-level diversity of the nominotypical genus, Geocoris Fallén, 1814, and closely allied genera is possibly resulted by quick radiation event after a climatic changes at the Eocene-Oligocene boundary.


Introduction
The family Geocoridae (Heteroptera: Pentatomomopha: Lygaeoidea), or commonly known as big-eyed bugs, is a peculiar lygaeoid true bug family in terms both morphological and ecological aspects. Representatives of the family display a series of highly specialised exoskeletal morphological features that are unique among lygaeoid true bugs, e.g. enlarged, reniform, often stylate eyes and medially curved sutures of abdominal tergites 4/6 and 5/6 (Henry, 1997). Furthermore, omnivorous feeding with a preference on predaceous behaviour is also unusual within the superfamily Lygaeoidea (lygaeoids are mostly seed and sap feeding except the hematophagous tribe Cleradini), which has consequently made some of the species promising candidates for biocontrol research (Sweet 2000;Kóbor 2020). However, taxonomy, phylogenetic relationships between and within taxon groups, and evolution of the representatives of the family are not well understood and virtually unstudied.
Controversially, these deficiencies are resulted by the peculiar appearance of these insects. Diagnoses of taxa were mostly based on colouration patterns and easy-toobserve exoskeletal characters, e.g. arrangement of head and eyes or shape of pronotum. This resulted in taxonomic confusions and misconceptions, especially in the nominotypical genus, Geocoris Fallén, 1814 which is considered as non-monophyletic in its current sense as suggested by more recent revisionary works (Readio and Sweet, 1982;Malipatil, 1994).
The examination of fossils provides useful information regarding the phylogenies and evolution of the studied groups and the interpretation of the results in context of recent taxa (e.g. including fossil taxa in phylogenetic reconstructions of extant groups) is momentous in solving systematic problems. As for geocorid true bugs, such data is relatively scarce. In total two fossil representatives of the family are known so far: Geocoris infernorum Scudder, 1890 from the Eocene (Florissant Formation,approximately 34 Ma.); and Geocoris monserrati Ortuño and Arillo, 1997 from the Oligocene (Izarra Formation, approximately 33.9 to 23.03 Ma.). Furthermore, the taxonomic placement of G. infernorum is precarious as stated by the author himself in the original description (Scudder 1890).
In this study, we present the description of the oldest known taxon belonging to this family from mid-Cretaceous Burmese amber. Furthermore, we review and revise the knowledge on the previously described fossil representatives of big-eyed bugs, discuss their placement within the family based on the morphological data available on recent taxa and form hypotheses on their evolutionary significance.

Materials and methods
The amber piece containing the studied fossil originated from Noije Bum, near Tanai (Grimaldi et al. 2002: fig. 1, Cruickshank andKo 2003: fig. 1;Bai et al. 2016). The studied piece of amber was purchased by and deposited in the collection of Paleoentomological Collection, Instituto de la Patagonia, Universidad de Magallanes in good faith for educational and research purposes and the authors of present study have no knowledge on unethical origin.
Observations, photomicrographs, and measurement for Protogeocoris arcanus gen. et sp. nov. were done with the use of Japan Optical Co. XLT-2310 stereoscopic microscope with Ricoh WG-50 digital camera adapted. Incident and transmitted lighting were used, mostly simultaneously.
Genus Protogeocoris nov. (Figures 1-3) Type species: Protogeocoris arcanus sp. nov.; by present designation and monotypy; gender: masculine. Etymology: Generic name is derived from the name of genus Geocoris Fallén, 1814 (nominotypical genus of family Geocoridae) adding the prefix "proto-" (Latin, meaning: "first of its kind") which refers to the fact that this genus is the oldest known representative of the geocorid true bugs. Differential diagnosis: Pronotum bilobate with narrow, but distinct collar and posterior margin concave (bilobate pronotum and distinct prothoracic collar can be observed in some representatives of Pamphantinae, but posterior margin is never concave); R-M of hemelytron basally fused, branching at 2/3 of length (basally fused, branching R-M can be observed in e.g. Engistus Fieber, 1864, Umbrageocoris Kóbor, 2019, but in these taxa pronotum always unilobed with straight or convex posterior margin); sutures between abdominal tergites 4/5-5/6 slightly and evenly curved posteriad at entire length (sutures of abdominal tergites 4/5-5/6 curved moderately to strongly medially in other geocorine genera; see Fig. 6a-b).

Description
Head pentagonal with vertex broad; eyes large, reniform, moderately stylate. Ocelli present, situated at middle of vertex, near base of eye stalks. Clypeus slightly pointed, appearing to exceed the mandibular plates. Antenniferous tubercles developed, but not visible is dorsal view. Antennomere I shortest, granuliform; antennomere II longest, cylindrical; antennomeres III and IV subequal in length; antennomere III cylindrical, antennomere IV fusiform. Labial trough non-observable. Labiomere I shortest, labiomeres II-IV subequal in length; apex of labium reaching mesocoxae. Thorax: Pronotum bilobate, trapeziform with narrow, but distinct collar; lateral margins moderately carinate and slightly constricted medially; posterior margin concave; pronotal callosities furrow-like, separated. Scutellum subequilateral, triangular with lateral margins slightly curved. Hemelytron appears to be membranous with clavus and exocorium more sclerotized than the rest of the hemelytron. Margins of clavus subparallel; claval commissure short, but distinct. Integument of corium appears to be wrinkled with simplified venation; Cu running near claval furrow; R-M basally fused, branching at approximately 2/3 of length; exocorium widening apically. Membrane with simplified, slightly visible venation lacking closed cells. Peritreme of metathoracic scent efferent apparatus bulbous without dorsal supportive process; orifice appears to be rounded; evaporatorium fine, matte, covering most of the metapleurite. Femora incrassate, fusiform; tibiae subcylindrical, slightly widened at apex. Tarsomere I longest, tarsomere II and III subequal in length; tarsomeres I and II cylindrical, tarsomere III fusiform. Claws simple, thin, evenly curved with thin parempodia-like appendages. Abomen: Abdominal dorsum mostly obscured by wings. Sutures between abdominal sternites 4/5-5/6 appear to be slightly, evenly curved posteriad. Outer genital characters, e.g. posterior opening of pygophore, parameres in situ, not observable.  Etymology: Species epithet "arcanus" (Latin, meaning: mysterious) refers to the evolutionary importance of the species in terms of the family Geocoridae. Differential diagnosis: Small insect (> 2mm) with characteristic pentagonal head, and slightly stylate reniform eyes. Hemelytra with clavus and exocorium more sclerotized and darker than rest of the hemelytron. Abdominal dorsum with a pair of dark longitudinal bands sublaterally. Other characteristics as in generic diagnosis. Description: Almost complete male individual in amber piece, left hind leg missing; characters obscured by syninclusions and air bubbles denoted in description respectively. Colouration: Vertex, eye stalks and thoracic dorsum with irregular darker markings, partly obscured by air bubbles. Antennomere I with irregular brownish spots and brownish annulation apically, antennomeres II-III with darker annulation apically, antennomere IV entirely darker (Fig. 1a). Hemelytra with clavus and exocorium entirely darker than rest of the hemelytron. Femora with irregular brownish spots. Abdominal dorsum with a pair of longitudinal, subparallel darker bands sublaterally. Abdominal sternites darker dorsolaterally. Structure: General habitus moderately elongate, slightly flattened dorsoventrally. Head: Eye stalks slightly projected, posterior edge of eyes not encompassing, nor touching anterior edges pronotum; ocular sulcus unobservable. Ratio of antennomeres: 1: 5.5: 5.0: 4.0. Ratio of labiomeres: 1: 6.5: 5.25: 0.30. Thorax: Integument of thoracic dorsum densely punctate except at pronotal callosities. Anterior edges of pronotum angulate, slightly rounded; humeral angles acute anteriorly and rounded posteriorly. Median trifurcate carina not observable, obscured by air bubbles. Thoracic pleurites and sternites densely punctate. Femora appear to be maculate.   Remarks on palaeohabitat: The amber piece containing multiple syninclusions, mostly soil and plant debris, remains of other insects and a piece of skin suspected to be originating from an avian leg. The most notable syninclusions are three complete insects: two barklice (Psocoptera gen. sp.) and a snakefly (Nanoraphidiini gen.sp.). Considering the results of Christiansen and Nascimbene (2006), Lis et al. (2018Lis et al. ( , 2020 and Jiang et al. (2019) we suggest that the palaeohabitat of the insect was part of an abundant geophilous-arboreal environment in an amber forest. The preference of geophilousarboreal habitats is similar to those of the extant representatives of the family Geocoridae (Slater 1977).

Redescription
Head lunular with vertex strongly widened; posterior edge of eyes encompassing anterior edges of pronotum. Eyes appear to Thorax: Pronotum conspicuously widened with anterior edges strongly rounded. Scutellum appears to be elongate triangular. Clavus, surroundings of venation and exocorium appear to be more sclerotized than the rest of the hemelytron. Clavus with margins subparallel; claval commissure complete, developed. Femora fusiform, fore femora appear to be strongly incrassated.
Abdomen. Abdominal tergites with sutures appear to be straight.
Remarks: The original description (Scudder, 1890) denotes that the identity and generic assignment of the species is problematic. This uncertainty stems from the nature of compression fossil (Azar et al., 2011;Chen et al., 2016) which can be distorted and mostly lacking detail. However, in case of G. infernorum there are characteristics to be observed based on which a diagnosis can be formulated. The convex general habitus, conspicuously lunular head and widened pronotum resembles to representatives of subgenus Piocoris Stål, 1872 and Geocoris s. str. species e.g. G. flaviceps (Burmeister, 1834) or G. willey Kirkaldy, 1905. However, these taxa are characterised by moderately incrassate fore femora, clavus with margins converging and claval commissure reduced, and sutures of abdominal tergites 4/5-5/6 strongly curved posteriad. In G. infernorum the fore femora are strongly incrassate, margins of clavus subparallel with a complete claval commissure and sutures of abdominal tergites 4/5-5/6 appear to be at most moderately curved. Similar arrangement of clavus and abdominal tergites can be found in Germalus and allies. Summarising, this species displays an amalgamation of characters present in various geocorine genera combined with the unusually strongly incrassate fore femora which is to be considered as autapomorphy in terms of the subfamily, thus its placement in Geocoris s. str. is not justified. Considering the above it is to be concluded that the species represents an extinct lineage of geocorine true bugs which lived during the Eocene epoch thus, the genus Eocenogeocoris Kóbor gen. nov. is suggested to be erected to accommodate the species.

Redescription
Head pentagonal, eyes reniform, slightly stylate; ocular sulcus appears to be complete, but slightly defined; posterior edge of eyes touching anterior edges of pronotum. Ocelli situated near ocular sulcus, at base of vertex. Clypeus rounded, appearing to be exceeding the mandibular plates; margins of clypeus converging basally. Vertex with median longitudinal furrow extending from the middle of vertex to apex of clypeus. Antenniferous tubercle appears to be well-developed. Antennae tetramerous with antennomere I granuliform, shortest; antennomere II cylindrical, longest (but incomplete); antennomeres III and IV subequal; antennomere III cylindrical, antennomere IV fusiform. Labial furrow appears to be closed; labiomere I present but appears to be incomplete.  Ortuño and Arillo (1997) assigned the taxon to genus Geocoris and suggested a close relationship to G. ater (Fabricius, 1787), a species of subgenus Geocoris. According to our findings the genus assignment and the suspected relationship is established, thus G. monserrati is the first known representative of "modern geocorids", i.e. Geocoris sensu lato and allied mono-and oligotypic genera.

Discussion
Superfamilial placement of Protogeocoris gen. nov. (Fig. 5) The studied fossil possesses four-segmented piercing-sucking mouthpart i.e. rostrum and heteronomous forewing i.e. hemelytron consisting of clavus + corium and membrane, such characteristics are placing this insect in suborder Heteroptera (Schuh and Weirauch, 2020). The general appearance of the insect, especially the large and stylate compound eyes resemble to the representatives of Yuripopovinidae Azar et al., 2011 e. g. Caulisoculus Zhang andChen, 2020 or Pseudocaulisoculus Kóbor and Roca-Cusachs, 2021. However, the venation of hind consists of three veins from which M-R are basally fused and venation of membrane is simplified, i.e. lacking closed cells. Contrastingly, hemelytron of the representatives of Yuripopovinidae possess three separate, longitudinal veins forming two closed cells, venation of membrane with closed cells. Furthermore, eyes unequivocally appear to be reniform, not rounded as in Yuripopovinidae. The presence of ocelli, incrassate fore femora and simplified venation of hemelytron with membrane lacking closed cells are shared synapomorphies of superfamily Lygaeoidea (Henry 1997). The studied fossil possesses all three of these characteristics, thus we ascribe this taxon to superfamily Lygaeoidea.

Placement of Protogeocoris gen. nov. within Lygaeoidea
This lygaeoid bug has pentagonal head with wide vertex and large, reniform, moderately stylate compound eyes which are characteristic to the family Geocoridae. Besides the arrangement of head and eyes, this peculiar true bug family is characterised by the dorsally situated abdominal spiracles II-IV and medially curved sutures between abdominal tergites 4/5-5/6 (Henry 1997) Abdominal sutures 4/5-5/6 appear to be slightly curved in Protogeocoris gen. nov. The degree of curvature of abdominal sutures considerably varies among geocorid taxa as demonstrated in Fig. 6a-b. Though abdominal spiracles are obscured in case of the specimen, supportive characters are to be observed which are displayed by various representatives of Geocoridae, e.g. placement of ocelli or trapeziform pronotum with furrow-like callosities (similar as in Unicageocoris Malipatil, 2013 and Nannogermalus Kóbor and Kondorosy, 2020). The proportion of diameter of compound eyes to head length is similar as in subfamily Geocorinae Baerensprung, 1860. This similarity is supported by the arrangement of head which resembles to genus Germalus Stål, 1862 and allied genera. Furthermore, these geocorine taxa are characterised with subequilateral triangular scutellum and clavus with margins parallel and claval commissure present. Prothoracic collar as separate structure can be observed in genus Epipolops Herrich-Schaeffer, 1850 of subfamily Pamphantinae Barber andBruner, 1933 (Rengifo-Correa et al. 2014). However, traces of such structure can be found in form of slightly elevated, impunctate region near the anterior margin of pronotum and prosternum in several species of subfamily Geocorinae (senior author's unpublished notes, Fig. 6c-d). Based on the above discussed evidence we suggest ascribing this new true bug taxon in subfamily Geocorinae.
Relationship of Protogeocoris gen. nov. to other representatives of Geocorinae The tribal classification of Geocorinae was proposed and later elaborated by Montandon (1907Montandon ( , 1913. Further improvement of the classification was later suggested by Parshley (1921); however, no revised classification was proposed subsequently. Though Montandon's concept gained no further acceptance by the community, the additional evidence provided by the study of Slater and Hurlbutt (1957) on the morphology of the metathoracic wing of Lygaeoidea, along with the senior author's unpublished findings suggests that two distinct major lineages are to be recognised within Geocorinae which are partly consonant with Montandon's tribes. However, there are genera which display transitions or unique states of characters, thus cannot be classified in these two groups e.g. Apennocoris Montandon, 1907;Stylogeocoris Montandon, 1913 or Nannogermalus Kóbor andKondorosy, 2020. Based on the comparison of diagnostic characters of these groups (Table 1) and the unique character state of prothorax with distinct collar it is to be assumed that Protogeocoris represents a separate, extinct lineage within the subfamily. The basally fused and apically branching R-M of hemelytron can be observed in multiple geocorid genera e.g. Engistus Fieber, 1864 (Henestarinae) or Umbrageocoris Kóbor, 2019 (Geocorinae) and the exact systematic importance of the characteristics of the hemelytron is subject of further comparative study.
Evolutionary significance of fossil big-eyed bugs According to Ross (2019Ross ( , 2022, Protogeocoris is the first representative of the superfamily Lygaeoidea reported from mid-Creatceous Burmese amber and the oldest known taxon of family Geocoridae. Until now, only two extinct representatives of the family were known from compression and mould fossils: Eocenogeocoris infernorum (Scudder, 1890) from the Eocene (Florissant formation, approximately 34 Ma.); and Geocoris monserrati Ortuño and Arillo, 1997 from the Oligocene (Izarra formation, approximately 33.9 to 23.03 Ma.).
The original description of E. infernorum is cursory and the illustrations are lacking detail. It must be noted that characterisation and diagnosis of compression fossils is problematic in most cases (Azar et al. 2011;Chen et al. 2016). The examination of the available photo documentation of the fossils suggest that the species clearly identifies as a representative of family Geocorinae and can be considered as a representative of an extinct of lineage displaying a combination of characteristics of various geocorine genera and possessing a suspected autapomorphy, strongly incrassate fore femora, thus Eocenogeocoris gen. nov. is erected to accommodate the species. G. monserrati is described and illustrated accurately and it is to be confirmed that the placement suggested by Ortuño and Arillo (1997), i.e. the species is closely related to the extant species, Geocoris ater (Fabricius, 1787) and its Mediterranean and Palaearctic relatives, is justified.
The fossil record of big-eyed bugs currently consists of three distinct taxa from three different geographic era: Protogeocoris arcanus from the mid-Cretaceous, Eocenogeocoris infernorum from the late Eocene and Geocoris monserrati from the Oligocene. The two oldest of them, P. arcanus and E. infernorum, represent extinct lineages whilst G. monserrati is unambiguously displaying characteristics of the "modern" representatives of the subfamily Geocorinae, more specifically the genus Geocoris sensu stricto. This suggests that the major synapomorphies characterising the family Geocoridae, e.g. the pentagonal head with variably stylate, reniform eyes have been evolved until the lowermost-Cenomanian age of the mid-Cretaceous, thus the origin of geocorid subfamilies and characteristic lineages can be dated back to this epoch. Other character states e.g. the reduction of claval commissure or the variability in the degree of curvature of the sutures of abdominal sutures 4/5-5/ 6 ( Fig. 6a-b), have been developed more recently and the taxa displaying character states which are suspectedly derived, i.e. the "geocorine lineage" (Geocoris sensu lato and closely allied mono-and oligotypic genera) appear latest in the Oligocene. The majority of geocorid species are representing this lineage. The species-level delimitation of these taxa is often problematic because the species can be diagnosed by a set of minor characters, which often display intraspecific variability of various degree (senior author's unpublished findings). The Eocene-Oligocene boundary witnessed a cooling event, which resulted significant changes in the taxonomic composition, richness, and abundance distribution of insect groups (Smith et al. 2014). Taking the above into account it is to be assumed that the diversity of the "geocorine lineage" is resulted by a quick radiation event that can be dated back to the Oligocene epoch and the "uniform heterogeneity" can be attributed to the relatively young age (< 33.9 Ma) of these taxa. However, it must be noted that this hypothesis is based on indirect evidence and its proof requires further investigations. A promising method of proof is performing morphology-based phylogenetic reconstructions involving both extant and fossil taxa. Including fossil records in morphological phylogenies of extant taxa is considered to improve the accuracy of reconstructions and to test evolutionary hypotheses even if the data acquired from the study of fossils is incomplete or partly incorrect (Edgecombe 2010;Mongiardino Koch et al. 2021).
Acknowledgements The authors would like to express their gratitude towards Jesus Alonso Ramirez, curator at MCNA for locating and documenting holotype of G. monserrati and Professor Brian D. Farrell at MCZN for providing access to the photodocumentation of E. infernorum. Furthermore, we would like to thank the work of two anonymous reviewers for their help in the improvement of the manuscript.
Author contribution All authors contributed to the study conception and design. The specimens studied were examined by Péter Kóbor and Eudardo I. Faúndez. Photomicrographs were done by Eudardo I. Faúndez, line drawings were done by Marcos Roca-Cusachs. The first draft of the manuscript was written by Péter Kóbor and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding Open access funding provided by ELKH Centre for Agricultural Research. Partial financial support was received from Project ANID-SA77210055 (Convocaria Nacional Subvención Instalación en la Academia).
Data availability Data sharing not applicable to this article as no datasets were generated or analysed during the current study.

Conflict of interest The authors declare that they have no conflict of interest
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.