The Fossil Record of Darkling Beetles (Insecta: Coleoptera: Tenebrionidae)

The fossil record of Tenebrionidae (excluding the Quartenary) is presented. In total, 122 fossil species, clearly belonging to the family, are known; some beetles were determined only to genus; 78 genera are listed in the fossil record, including 29 extinct genera. The great diversity of tenebrionids occurs in the Lower Cretaceous Lagerstätte of China (Yixian Formation), Middle Paleocene of France (Menat), Lower Eocene deposits of Germany (Geiseltal), Upper Eocene Baltic amber (Eastern Europe), Upper Eocene deposits of Florissant Formation (USA) and Miocene (Dominican amber). Tenebrionids of the following major lineages, including seven subfamilies, are currently known in the fossil record. These include the lagrioid branch (Lagriinae, Nilioninae), pimelioid branch (Pimeliinae), and tenebrioid branch (Alleculinae, Tenebrioninae, Diaperinae, Stenochiinae). The importance of the fossil record for evolutionary reconstructions and phylogenetic patterns is discussed. The oldest Jurassic and Early Cretaceous darkling beetles of the tenebrionoid branch consist of humid-adapted groups from the extant tribes Alleculini, Ctenopodiini (Alleculinae), and Alphitobiini (Tenebrioninae). Thus, paleontological evidence suggests that differentiation of the family started at least by the Middle Jurassic but does not indicate that xerophilic darkling beetles differentiated much earlier than mesophilic groups.


Introduction
Tenebrionidae is one of the largest beetle families, comprising nearly 20,000 extant species and 2300 genera in the World [1]. The monophyly of the family was justified by many authors in the 20th and 21st centuries [1] and was also recently supported after the analysis of eight gene markers, but at least two large subfamilies, Tenebrioniae and Diaperinae, seem to be paraphyletic or polyphyletic [2]. The systematics of tenebrionid beetles was modified by many authors. The currently accepted classification of the family is based on the analysis of multiple external and internal morphological structures [1,[3][4][5][6][7][8][9][10] and includes 11 subfamilies [10][11][12][13] from lagrioid, pimelioid, and tenebrionoid branches [11]. Fossil Tenebrionidae have been used in evolutionary reconstructions [14,15]. The fossil record of darkling beetles is highly fragmented. Matthews et al. [1] briefly discussed the fossil record of Tenebrionidae, but since that time, important new data have been published. In the present work, the current data on extinct Tenebrionidae are summarized and analyzed. This includes a new catalogue of extinct darkling beetles, critical evaluation of fossil taxa, questions of systematics, and discussions of evolutionary scenarios and paleo-reconstructions.     The following abbreviations (Table 1) are used for the various fossil deposits in Table 2, the figure, and the text below.

Subfamily Alleculinae
The oldest known distinct representatives of Tenebrionoidea are from the Callovian/Oxfordian of the Middle/Late Jurassic, Wuhua jurassica (Wang et Zhang, 2011) [86] and Praemordella martynovi (Shchegoleva-Barovskaya, 1929) [87] and belong to the clade Mordellidae-Ripiphoridae of this superfamily [59]. Archaeoripiphorus nuwa (Hsiao et al., 2017) [88] from the same stage was originally described in Ripiphoridae but was later transferred to Tenebrionoidea familia incertae sedis [59]. Mordellid-ripiphorid beetles reach the largest proportion of the described taxa (32%) within Mesozoic Tenebrionoidea. Darkling beetles reach 11%, with the predominance of Alleculinae [59]. Such imbalances in the record seem to be a result of the fossilization: Ripiphoridae and Mordellidae are small beetles, common in Burmese amber, while much larger Mesozoic tenebrionids are known from Burmese amber by a single undescribed specimen; the majority of the Mesozoic tenebrionids were described from compression prints in a stone matrix. This is due to the different fossilization of bio-inclusions entering fossil resins and various types of compression in burials.
The oldest tenebrionid beetle Cistelites insignis (formally Alleculinae) from the Early Jurassic has an unclear position within Coleoptera, and the holotype of this species (print of single elytron) must be re-examined. The Middle/Late Jurassic Jurallecula grossa clearly belongs to the subfamily Alleculinae and, judging from its characters (preapical tarsomeres not bilobed), to the gonoderine branch of Alleculini [31]. Alleculinae are known by two described species from the Yixian Formation of the Lower Cretaceous [42,43], herewith Calcarocistela kirejtshuki, which belongs to gonoderine branch, and Platycteniopus diversoculatus has characters intermediate between those known in the alleculine and cteniopodine branches of the subfamily. All extant Cteniopodini associated with generative organs of Angiosperms and the age of Platycteniopus corresponds to the age of the early stages of the Angiosperm evolution. Probably, the origin of the tribe Cneniopodini should be dated to the Jurassic-Cretaceous border.
Many species of the formal genus Cistelites Heer, 1865, described from the Lower Jurassic of Switzerland (C. insignis), the Upper Cretaceous of Sakhalin (C. sachalinensis), the Lower Paleocene of Greenland (C. minor and C. punctulatus), the Middle Miocene of China (C. longipes) and the Upper Miocene of Germany (C. spectabilis) cannot be interpreted as possible close relatives, and this genus cannot be placed in the current classification due to its very sparse description. These taxa are not included in this review.
Numerous and diverse material on Alleculinae appears in the Cenozoic (Paleogene), especially in the Eocene Baltic and the Rovno amber, where mainly extant and one extinct (Mycetocharoides Schaufuss, 1888) genera are represented. The most diverse group among Baltic Alleculinae is the subtribe Gonoderina (tribe Alleculini). Two genera Isomira Mulsant, 1856 and Gonodera Mulsant, 1856 presented in the Recent fauna are registered in Baltic amber [52,54,65], herewith some extinct species of Isomira belong to the subgenera distributed in recent East Asia. This suggests that the most archaic representatives of this genus survived in the Palaearctic Asia and dispersed to the East Palaearctic, that is consistent with data on other families of beetles and plants [89,90]. The subtribe Alleculina, represented in Baltic amber by the genera Allecula Fabricius, 1801, Mycetochara Guérin-Méneville, 1827, and Hymenalia Mulsant, 1856, occur in the recent fauna and as well as the extinct genus Mycetocharoides, which only slightly differs from Mycetochara. The tribe Ctenipodini is known from Baltic amber by two extant genera, Cteniopus Solier, 1835, and Cteniopinus Seidlitz, 1896. The gonoderine taxa from the genera Gonodera and Capnochroa LeConte, 1862 [27,28], and the alleculine Hymenophorus Mulsant, 1851 [27], were described from the volcanic deposits of Florissant with a comparable age. It should be added that Doyen and Poinar [38] doubted the correctness of the Wickham's identifications. Only extant genera are known from the Neogene deposits [28], excluding the extinct genus Sinocistela Zhang, 1989, from the Miocene of China. Sinocistela clearly belongs to the tribe Cteniopodini because it has six exposed abdominal ventrites.
Based on the fossil record, the oldest tenebrionids are gonoderine comb-clawed beetles (tribe Alleculini, subtribe Gonoderina) with non-lamellar and non-bifurcated preapical tarsomeres. They are known from the Late Jurassic. However, taken into consideration their "modern" body, it can be assumed that this group probably originated earlier, perhaps in the Middle Jurassic.
Mesozoic gonoderine comb-clawed beetles are characterized by the morphological conservatism and slightly differ from the extant taxa of the subtribe Gonoderina (tribe Alleculini), with some particular adaptations, such as very long metatarsal spurs in Calcarocistela. Unlike Cteniopodini, many Alleculini feed on Cyanophita, Bryophita, or lichens growing on trunks of trees; only adults of some groups (for example, Isomira) switched to feeding on the generative organs of Dicotyledones, but their larvae develop in rotten wood, as do the larvae in most Alleculini [91,92].
The alleculine branch (subtribe Alleculina) appears in the fossil record only in the late Eocene, but that seems to be caused by a lack of material. Diverse taxa of this subtribe currently are widely distributed on all continents, and they most likely originated much earlier than indicated in the fossil record.
Separation of the alleculine and cteniopodine branches could have occurred at the Jurassic-Cretaceous border. Extant Cteniopodini feed on flowering plants: therefore, it is assumed that the appearance of the angiosperms promoted diversification of the subfamily.

Subfamily Tenebrioninae
Tenebrioninae is the largest and probably most diverse subfamily of darkling beetles in the Recent fauna. The oldest representative of this subfamily, Alphotopsis initialis, was found in the Lower Cretaceous Lagerstätte of Yixian, China [41]. This species, described from a well-preserved print, is very similar to extant taxa of the tribe Alphitobiini (in particular to Alphitobius Stephens, 1829) and has all characters typical for the subfamily, including intersegmental abdominal membranes, defensive glands, and open mesocoxal cavities with well-expressed trochantin. Only this particular species is known from the Mesozoic. Other tenebrionines appear in the fossil record with a long break only in the Middle Paleocene (the Thanetian in Menat, France). Taxa of Helopini and Opatrini are known from this deposit [44,45]. Representatives of these two tribes belong to extinct genera but have features of specialization as in the extant tenebrionids from these groups. For example, Cryptohelops menaticus is very similar to the extant genus Stenohelops sensu str., by having widened male protarsi, coarsely punctured prohypomera, cordiform pronotum, lacking wings, and epipleura not reaching the apex. Furthermore, the emargination on the inner side of the protibiae of Cryptohelops menaticus is characteristic also of some extant genera from the subtribe Helopina.
Paleosclerum pohli belongs to the specialized myrmecophilous group of the subtribe Sclerina, as some extant genera of the tribe Opatrini. One extinct opatrine genus, Eupachypterus eocenicus from the subtribe Neopachypterina is known from Lower Eocene Oise amber [46]. The extant taxa of this subtribe occur in the Indo-Malayan and Afrotropic regions, while others are distributed in transitional biogeographical areas of the Palaearctic.
All these findings (including taxa of Opatrini from distinct subtribes) indicate a much earlier Mesozoic diversification of various tribes of the subfamily Tenebrioninae.
The next Paleogene time period with diverse and multiple samples of Tenebrioninae is the late Eocene-early Oligocene, especially Baltic amber and Upper Eocene deposits of Florissant in the USA. Mainly extant genera from the tribes Bolithophagini, Pedinini, Triboliini, Ulomini, Helopini, Palorini are known in Baltic amber. Amber specimens listed by Klebs [29] are lost, and those taxa cannot be compared with extant ones. Helopini are represented by one abundant species Stenohelops klebsi [49]. This species is very similar to Nalassus spp. (subtribe Cylindrinotina) and was initially described in this genus. After the study of additional materials, it was transferred to the genus Stenohelops (subtribe Helopina) on the basis of the catomoid type of aedeagus [93,94], absence of grooves near the lower surface of eyes, double bead of abdominal ventrite 5, and widened male protarsi [52]. Palorini are represented in Baltic amber by two genera, the extinct Vabole Alekseev et Nabozhenko, 2015, and extant Palorus Mulsant, 1854. Eocene species of both taxa belong to the life form with flattened bodies, similar to those in the Gondvanian generic group Platycotylus-Australopalorus [47,48].
Four taxa of the subfamily Tenebrioninae were described from the Florissant deposits: two species of Opatrini, one from Bolithophagini and one from Amarygmini [24,28]. Both fossil species of Opatrini, Ephalus adumbratus and Ulus minutus (subtribe Opatrina) belong to the extant xeromorphic psammophilous genera with species occurring in inner-continental and marine supralittoral landscapes. Proteleates centralis, described in the tribe Bolithophagini, is very similar to the extant Neotropic genus Eleates Casey, 1886 (this genus was erroneously listed as Proteleratus and included to the tribe Blaptini by Kirejtshuk et al. [39]). Heyden and Heyden [18] erroneously placed Bolitophagus vetustus in this genus on the basis of two thin horns on the head. This species should be transferred to another genus of Bolitophagini, because this head character is typical for the extant Eastern Palaearctic and Indo-Malayan genus Byrsax Pascoe, 1860; some Neotropic Megeleates Casey, 1895, also have horns on the head. Wickham described Meracantha lacustris in the tribe Amarygmini; the second known species (extant) of this genus is distributed in North America and feeds on arboreal lichens.
Within the Palaeogene Tenebrionidae are some taxa with unclear positions (including members of the genera Tenebrio Linnaeus, 1758, Bolitophagus Illiger, 1798, and Uloma Dejean, 1821), from the Lower Eocene deposits of Canada (Whipsaw Creek) [23] and the Upper Oligocene deposits of Rott in Germany [16][17][18], which were only briefly described and poorly illustrated. Only the recent genera occur in the Neogene. The most reliable information relates to taxa from the genera Wattius Kaszab, 1982 (Toxicini), Rhipidandrus LeConte, 1862 (Bolithophagini), Cymatothes Dejean, 1834 (Amarygmini), and Hypogena Dejean, 1834 (Triboliini), described from the Miocene Dominican amber [38]. The majority of mentioned genera are spread in the Neotropics and partly in the Nearctic regions, with the exception of Rhipidandrus, species of which are known from all tropical areas of the World.
Within the described Neogene darkling beetles, the Miocene Helopini are the most diverse, but the tribal position of these taxa is very doubtful. All these species were described in the genus Helops sensu lato, which in the 19th century contained the most current helopin genera and many unclear genera from different subfamilies of tenebrionid beetles. It is impossible to clarify the taxonomic position of these taxa without examining the type specimens. Comments on these species are given below: [Helops] atticus Redtenbacher in Ungern, 1867 (Ng(Mi-Kumi) 20.4-16). This species was described on the basis of the print of the right elytron and compared with extant Entomogonus peyronis (Reiche, 1861) and Erionura gigantea (Kraatz, 1862). The high-quality image in the original description does not give any reason to clearly attribute this print to any subfamily of Tenebrionidae. As a result, this species is regarded as Coleoptera, family incertae sedis.
[Helops] meissneri Heer, 1847 (Tafel V, Figure 9) (Ng(Mi)12.7-11.6). This print, as described in detail, can belong to Helops sensu str. Its most peculiar character is the distinct deflected lateral margin of elytra. Its head is very poorly preserved and cannot be used for diagnostics. O. Heer mentioned in the original description that its elytra are similar to some Carabus Linnaeus, 1758, but the clearly-expressed full elytral striae indicate the possible position of this species within the genus Helops. Unfortunately, the image in the original description is not informative.

Subfamily Stenochiinae
The oldest representative of the subfamily Stenochiinae is known from the early Paleocene of Greenland (Pg(Pc) 61.7-58.7): Pseudohelops groenlandicus (originally Helops molassicus Heer, 1883). Haupt [30] compared this genus and species with the Eocene species of Anthracohelops on the basis of a similarity in the metallic shine of its elytra but contraposed both genera in the structure of elytral striae. Other Stenochiinae are known from the Middle Eocene of Germany (Pg(Eo-Geis) 48.6-40.4), where they were hugely diverse [30]. Haupt described several taxa in 1950 [30]: Pyrochalcaspis (one species), Eodromus (five species), Caryosoma (one species), Paraceleusticus (one species), Anthracohelops (four species), Mimohelops (one species) and Pseudohelops (one species). He divided all these genera by the structure of pronotum and the elytral sculpture. The first two can be compared with the extant species of the Neotropic genus Hegemona Laporte, 1840 (Cnodalonini), which also have fusiform convex elytra with very deeply impressed strial furrows. The genus Eodromus was compared by Haupt with extant representatives of the Neotropic genus Camaria Lepeletier and Audinet-Serville, 1828. However, Eodromus distinctly differs from Camaria and allied Neotropic genera (Mylaris Pallas, 1781, Taphrosoma Kirsch, 1866, etc.) in the absence of humeral angles (while the mentioned extant genera have clear humeral angles and callosity indicating the ability to fly). Eodromus can be compared with recent Hegemona, which are also wingless and have very similar elytral sculpture. Another genus, Cariosoma Haupt, 1950, "close" (according to Haupt) to Eodromus, has strongly sculptured elytra with elongate tubercles, similar to those in the Indo-Malayan taxa of the tribe Cnodalonini (Morphostenophanes Pic, 1925, and allied genera). Parakeleusticus postumulus is very similar to the recent Hegemona resplendens Laporte, 1840, from Southern Mexico. The taxonomic position of other mentioned genera of the subfamily Stenochiinae remain unclear, although Haupt compared them with the recent Camaria (Cnodalonini).
Thus, in the early Eocene of Geiseltal (Pg(Eo-Geis)48.6-40.4) the thermophilic and humid adapted species of the subfamily Stenochiinae, similar to those from Recent Neotropic and Indo-Malayan taxa, were well represented. This, as in the case of the Coleoptera composition in the earliest Eocene amber Oise, could be just at the beginning of the transition from the climatic thermo-era to cryo-era, from the thermal maximum in the Paleocene-Eocene to the global trend of climate cooling in the Middle Eocene-late Oligocene [95].

Subfamily Diaperinae
Diaperinae are poorly represented in the fossil record. This feature probably relates to their secretive lifestyle in natural landscapes, such as inside tunnels of bark beetles (Hypophlaeini), in the litter of trees and shrubs (many Gnathidini), in tree mushrooms, and under the bark (Diaperini, Scaphidemini).
Ceropria messelense was described from the early Eocene of Germany (Pg(Eo-Mess)48.6-40.4), although this genus was determined by the author with doubts [80]. Extant representatives of this genus are widespread in the Indo-Malayan Region.
Some Diaperinae were listed from the late Eocene to early Oligocene: Pg(Eo-BA)37.2-33.9 and Pg(Eo-Flor) 37.2-33.9. All of them belong to extant genera of the tribe Diaperini: two species of Platydema Laporte and Brullé, 1831, from Florissant [27], and one species of Pentaphyllus Dejean, 1821, [39] from Baltic Amber. The recent representatives of the first genus are widely distributed in the Palaearctic, Nearctic, Indo-Malayan, Afrotropic, and Neotropic regions. Pentaphyllus occurs in the same biogeographical regions, except Neotropic.

Subfamily Pimeliinae
Taxa of pimeliine branches are poorly represented in the fossil record because they probably occurred, as did the Recent members of this tribe, mostly in arid biomes (deserts and semideserts), and they would rarely be deposited in tree resin or compression lacustine burials. However, even single records are of great interest because they can help to analyze the diversification of pimeliine tribes, as well as to reconstruct paleo landscapes.
In the paleontological record, only extant tribes of Pimeliinae are presented (Asidini and Stenosini). Fossil species of tribe Asidini were found in the deposits of Florissant (Pg(Eo-Flor) 37.2-33.9) (Pelecyphorus (Stenosides) primus) [27] and in Baltic amber (Asida (Planasida) groehni) [53]. Both species belong to extant subgenera and undoubtedly demonstrate the earlier (possibly, early Eocene or late Paleocene) diversification of these groups. The finding of fossil Asida groehni in Baltic amber permits a more critical look at various reconstructions of the Baltic amber forest, in which the presence of xerophytic landscapes was not assumed in previous interpretations, and also the taphonomic reasons and selectivity of the amber fauna were not always taken into account [96,97]. Meanwhile, the Palaearctic genera of this tribe are xeromorphic (closed mouthparts without membranous surfaces, developed subelytral cavity, etc.) and include species of Asida Latreille, 1802, and allied genera that inhabit dry landscapes in arid Mediterranean woodlands, steppes, and semideserts. At least arid woodlands undoubtedly existed in the late Eocene landscapes of the Paleo-Eurasia.
Another fossil representative of the subfamily Pimeliinae belongs to the tribe Stenosini, which includes the myrmecophilous species, sporadically occurring in the Recent New World and widespread in the Recent Old World, mainly in the Palaearctic and the dry subtropics of the Indo-Malayan Region. Miostenosis lacordairei, described from the Eocene Florissant deposits on the basis of well-preserved print [27], is externally similar to the Palaearctic Stenosis Herbst, 1799. A distinctive feature of Miostenosis Wickham, 1913 is large round eyes, not characteristic of Stenosini, which have narrow, partly, or completely divided eyes, sometimes oval dorsally (some Stenosis). The taxonomic position of this extinct species is still unclear.
A single Neogene fossil representative of the extant genus of the tribe Edrotini, Trientoma hascens, was described from Miocene Dominican amber [38]. The extant species of this genus are distributed in the West-Indies and often occur in coastal dry woodlands or dry subtropic forests.
The genus Yantaroxenos is most similar to the Neotropic Rhypasma Pascoe, 1862, which indicates ancient connections of the Early and Middle Paleogene faunas of Coleoptera of different paleo continents [29,32]. Records of many taxa of the mainly tropic subfamily Stenochiinae in early Eocene deposits of Geiseltal (Pg(Eo-Geis) 48.6-40.4) [30] also support this assumption.
The separate tribe Gonialaenini was erected for the genus Gonialaena because it combines some characters of the tribes Laenini, Goniaderini, Lupropini, and Lagriini and, possibly, represents a Mesozoic relic that survived until the late Eocene [51].
In Neogene Dominican amber are the only known fossil species of Lorelus Sharp, 1876, and Statira [38]. The extant taxa of the first genus are widely distributed in the New World; however, extant species of Statira occur in the Indo-Malayan, Nearctic, and Neotropic regions.

Subfamily Nilioninae
These unusual darkling beetles were interpreted within the Lagrioid branch of Tenebrionidae [1]. All extant taxa of the subfamily are Neotropic. The only extinct species Nilio dominicanum was described from the Miocene Dominican amber [98]. The species was originally described as dominicana but should be corrected to dominicanum according to the neuter generic gender. The extant species of Nilioninae feed and develop on lichens [1,99].

The Fossil Record of Tenebrionidae in Evolutionary Reconstructions
The conclusions regarding fossil Tenebrionidae are primarily assumptions based on incomplete material rather than postulates.
In the current evolutionary scenario using molecular dating and diversification analyses, Kergoat et al. [14] suggested that Tenebrionidae originated after the Triassic-Jurassic mass extinction, and diversification of the major branches of Tenebrionidae proceeded in the period of Pangaean and Gondwanan breakups. These authors established that arid-adapted darkling beetles (mostly Pimeliinae) diversified earlier (in the Middle-Late Jurassic), while the humid-adapted tenebrionids (Tenebrioninae, Diaperinae, Stenochiinae) diversified much later, in the early and late Cretaceous (Alleculinae).
Generally, the fossil record supports this view (Figure 1), but some discrepancies can be found. Based on the record of Middle/Late Jurassic Jurallecula grossa (Mz(Ju) 164.7-155.7), L. Medvedev [31] was the first who assumed (long before modern studies), that Tenebrionidae originated in the early Jurassic or slightly earlier. So, his estimated time of the origin of darkling beetles does not contradict Kergoat et al. [14]. Despite the lack of paleontological evidence, the presumably Jurassic diversification of the pimelioid branch as assumed by Matthews et al. [1] and Kergoat et al. [14] can be supported. It can also be assumed that the pimeliine tenebrionids during the Jurassic were diversified and produced main separated phyletic branches because, among all extant representatives of the largest subfamily Pimeliinae and the small Kuhitangiinae, mesophilic humid-adapted groups are absent. Generally, the fossil record supports this view (Figure 1), but some discrepancies can be found. Based on the record of Middle/Late Jurassic Jurallecula grossa (Mz(Ju) 164.7-155.7), L. Medvedev [31] was the first who assumed (long before modern studies), that Tenebrionidae originated in the early Jurassic or slightly earlier. So, his estimated time of the origin of darkling beetles does not contradict Kergoat et al. [14]. Despite the lack of paleontological evidence, the presumably Jurassic diversification of the pimelioid branch as assumed by Matthews et al. [1] and Kergoat et al. [14] can be supported. It can also be assumed that the pimeliine tenebrionids during the Jurassic were diversified and produced main separated phyletic branches because, among all extant representatives of the largest subfamily Pimeliinae and the small Kuhitangiinae, mesophilic humidadapted groups are absent.
The oldest Jurassic and Jurassic-Cretaceous darkling beetles of the tenebrionine lineage belong to humid-adapted groups from the extant tribes Alleculini, Ctenopodiini (Alleculinae), and Alphitobiini (Tenebrioninae) (Figure 1). Thus, Paleontological evidence supports a hypothesis that the differentiation of the family started at least in the Middle Jurassic, regardless of whether the ancestral forms of Tenebrionidae were xerophilic or mesophilic (Figure 1). In this case, paleontological evidence contradicts the scenario suggested by Kergoat with coauthors [14].
A more or less distinct time interval can be established only for the origin of the tribe Cteniopodini. The oldest genus of this tribe, Platycteniopus (Mz(Cr) 125.5-122.5), combines both the features of gonoderine ancestors and the "new" cteniopodine characters, which are explained by the The oldest Jurassic and Jurassic-Cretaceous darkling beetles of the tenebrionine lineage belong to humid-adapted groups from the extant tribes Alleculini, Ctenopodiini (Alleculinae), and Alphitobiini (Tenebrioninae) (Figure 1). Thus, Paleontological evidence supports a hypothesis that the differentiation of the family started at least in the Middle Jurassic, regardless of whether the ancestral forms of Tenebrionidae were xerophilic or mesophilic (Figure 1). In this case, paleontological evidence contradicts the scenario suggested by Kergoat with coauthors [14].
A more or less distinct time interval can be established only for the origin of the tribe Cteniopodini. The oldest genus of this tribe, Platycteniopus (Mz(Cr) 125.5-122.5), combines both the features of gonoderine ancestors and the "new" cteniopodine characters, which are explained by the change of trophic specialization of all representatives of this tribe to feeding on generative organs of angiosperms. This is the only known case of the direct impact of early angiosperms on the evolution of darkling beetles. In this case, we support the opinion of Kergoat with co-authors [14] about the insignificant impact of angiosperms on the diversification of tenebrionids, contrary to Yu et al. [64], who indicated the origin of Tenebrionidae in the Cretaceous (near 120 Ma) and an important role of angiosperms in their evolution.
The main contradiction is a discrepancy between the fossil record, morphological conclusions and evolutionary scenarios.
According to many morphological works and phylogenetic reconstructions [1,3,[6][7][8][9][10][11], the hypothetical ancestor of tenebrionid beetles had the characters typical of the lagrioid branch and archaic Zolodininae (tentatively pimelioid branch). However, the oldest Tenebrionidae belong to the tenebrionoid lineage and only partly fit the 'lagriine' features (Table 3). * The ancestor is most closely related to the extant Belopini of the lagrioid branch and Zolodininae, tentatively in the Pimeloid branch [10] and probably Kuhitangiinae of the pimelioid branch [13] Thus, only two characters fit (character eight only partly fits) the hypothetical ancestor of darkling beetles. One additional dual character in the hypothetical ancestor is the open or closed mesocoxal cavity. Zolodininae, the most primitive Tenebrionidae that is similar to the hypothetical ancestor, have open mesocoxal cavities, but Belopini and Kuhitangiinae have closed ones. Matthews and Bouchard [10] (p. 35) assumed that "tenebrionoids would have branched off very soon thereafter" because Palorini of the tenebrionoid branch have many of the mentioned "ancestral" characters. In general, this hypothesis is supported by the fossil record, and also, it can be assumed that three main lineages of Tenebrionidae were branched during a short time period.
On the other hand, the oldest fossil Tenebrionidae has many characters (1)(2)(3)(4)(5)7,10), which are currently interpreted as derived [1,10]. This discrepancy can be explained by gaps in the fossil record (especially in Jurassic deposits) and the earlier origin of Tenebrionidae, in Early Jurassic, as assumed by L. Medvedev [31] and Kergoat et al. [14]. In any case, taking into account the grounded monophyly of Tenebrionidae, the basal diversification (branching of the main lineages) of this family started at the earliest stages of the evolution of the darkling beetles.

Conclusions
The family Tenebrionidae is represented in the fossil record with 122 species, 78 genera (including 29 extinct genera) from seven subfamilies of three main branches in the time span of 201-11.6 Ma. This includes the lagrioid branch (Lagriinae, Nilioninae), pimelioid branch (Pimeliinae), and tenebrioid branch (Alleculinae, Tenebrioninae, Diaperinae, Stenochiinae). The most diverse records are reported in the Paleogene, Eocene (Geiseltal Halle, Baltic Amber, Florissant), and Miocene Dominican Amber. The oldest known distinct Mesozoic darkling beetles are from the Middle/Late Jurassic of Karatau (Alleculinae) and early Cretaceous of Yixian (Alleculinae and Tenebrioninae).
Darkling beetles of the tenebrionoid branch are the most diverse in the fossil record: Alleculinae: 30 species (since Middle Jurassic in the fossil record); Tenebrioninae: 28 species (since early Cretaceous in the fossil record); Stenochiinae: 21 species (since Middle Paleocene in the fossil record); Diaperinae: 15 species (since Middle Paleocene in the fossil record); lagrioid branch: 14 species: Lagriinae: 13 species, including an undescribed one (since late Cretaceous in the fossil record); Nilioninae: one species (since early Miocene in the fossil record).
Paleontological evidence partly suggests that the following evolutionary scenarios of Tenebrionidae occurred as indicated by Kergoat with co-authors [14]: (1) angiosperms had an insignificant impact on the diversification of tenebrionids, with the exception of diversification of the pimelioid branch during the Late Jurassic; (2) paleontological evidence supports the differentiation of the family at least by the Middle Jurassic, whether the ancestral forms of Tenebrionidae were xerophilic or mesophilic. Discrepancies between the hypothetical lagrioid ancestor of Tenebrionidae and the oldest known tenebrionoid fossils are possibly related to gaps in the fossil record.