Extant Genus in the Mesozoic: Paleoplatyura Meunier (Diptera: Keroplatidae) Found in the Cretaceous Amber of Myanmar

Simple Summary Burmese amber is very rich in perfectly preserved insects. Consequently, it is an invaluable source of information for taxonomic and evolutionary studies. Moreover, it forms a unique connection between the Jurassic and Cretaceousfaunas and documents the first representatives of modern genera. In this paper, a primitive genus of Keroplatidae, Paleoplatyura Meunier, 1899, is recorded from Burmese amber for the first time. This represents a rather rare case of the presence of an extant insect genus in the Mesozoic. Three new species of Paleoplatyura are described, indicating that this genus was relatively diverse already in the Cretaceous. Abstract Three new species of Paleoplatyura Meunier, 1899, i.e., Paleoplatyura agnieszkae sp. nov., P. miae sp. nov., and P. magnifica sp. nov., are described and figured. The concept of the genus is briefly discussed, and its systematic position is clarified. A key to fossil species is provided. The genus Paleoplatyura is described from the Eocene Baltic amber. It is concluded that, in Baltic amber, this group is represented only by the type species, and the identity of the other two species is problematic. No additional specimens have been found so far in this amber. Therefore, the presence of as many as three new species in Burmese amber, certainly belonging to Paleoplatyura, is a confirmation of its occurrence already in the Mesozoic.

The genus Paleoplatyura currently formally includes three extant species, i.e., P. aldrichii Johannsen, 1909 [17]; P. johnsoni Johannsen, 1910 [16], and P. melanderi Fisher, 1941 [15], and three described fossil species, i.e., P. macrocera (Loew, 1850 [6]), P. loewi Meunier, 1922 [18], and P. (?) eocenica Cockerell, 1921 [19]. However, the placement of P. aldrichii, P. melanderi, P. loewi and P. eocenica in the genus Paleoplatyura is questionable and most probably wrong. The holotype of P. aldrichii is probably lost [20], and the original description by Johannsen [17] is very short, without any figure provided, only Johannsen [16] mentioned that "this species differs in several important structural characters from P. johnsoni" (he explicitly specifies only the subcostal cross-vein absent and cubital vein not reaching wing margin), indicating that P. aldrichii probably does not belong to true Paleoplatyura. Concerning P. melanderi, Mantič et al. [1] did not consider this species as true Paleoplatyura, because of R 2+3 ending in R 1 and its considerable molecular distance from P. johnsoni. Additionally, P. loewi does not belong to Paleoplatyura, because this species lacks the transverse r-m vein characteristic for this genus and possesses the more apomorphic r-m fusion, typical of most keroplatids. This can be seen exactly in the drawing of the wing given in the work of Meunier [18], p. 3, Figure 1., where the vein M 1+2 merges at some distance with the vein Rs. The identity of P. eocenica is most obscure, because nothing important can be inferred from the original description and it is even unclear if the species belongs to Keroplatidae. Cockerell [19] himself was not sure whether this species belongs to the genus Paleoplatyura, providing a question mark after the genus name in the original description.

Materials and Methods
Specimens were examined using a Nikon (Minato, Japan) SMZ25 stereomicroscope, equipped with a Nikon DS-Ri2 digital camera. Photomicrographs are focus stacks captured using this system and processed using NIS-Elements Imaging Software (Minato, Japan). Line drawings were produced by tracing photographs. The terminology follows Ševčík et al. [21], where the homology and wing vein nomenclature in Bibionomorpha are briefly explained. The holotypes are deposited in the collection of the Institute of In this paper, we aim to clarify the taxonomic concept of the genus Paleoplatyura and describe three new Cretaceous species of this remarkable keroplatid genus.

Materials and Methods
Specimens were examined using a Nikon (Minato, Japan) SMZ25 stereomicroscope, equipped with a Nikon DS-Ri2 digital camera. Photomicrographs are focus stacks captured using this system and processed using NIS-Elements Imaging Software (Minato, Japan). Line drawings were produced by tracing photographs. The terminology follows Ševčík et al. [21], where the homology and wing vein nomenclature in Bibionomorpha are briefly explained. The holotypes are deposited in the collection of the Institute of Systematic and Evolution of Animals Polish Academy of Sciences (ISEA PAS) and paratypes in the National Museum, Prague, Czech Republic (NMPC). The specimens described here come from the Hukawng Valley in Kachin State, northern Myanmar. Burmese amber was dated by Cruickshank and Ko [22] to the middle-late Albian, based on insect inclusions and a specimen of the ammonite, but Grimaldi et al. [23] estimated the age of this resin to the Turonian-Cenomanian, based on arthropod inclusions. Shi et al. [24], based on U-Pb dating of zircons from the volcaniclastic matrix of the amber, estimated the age of Burmese amber at 98.79 ± 0.62 Ma (earliest Cenomanian).
This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The LSID for this publication is: urn:lsid:zoobank.org:pub:90A627F2-9ECB-4ACE-81FC-9AC006D90656.
Diagnosis: Cross-vein r-m present; basal part of Mb clearly visible; cross-vein m-cu situated in between M 1+2 and Cu; R 2+3 oblique and ending in C, anal vein strong and reaching wing margin; gonostylus narrow and apically bifurcated.
Remarks: P. macrocera is the type species of the genus Paleoplatyura. It was described on the basis of a single female inclusion in the Baltic amber. Unfortunately, the study is made more difficult by the missing holotype and the fact that despite the analysis of numerous specimens classified to Sciaroidea in various Baltic amber collections, no specimen belonging to this species was found. Therefore, the concept of the genus is based on the figure provided in the work of Meunier (1899, Figure 9            Paleoplatyura miae sp. nov. (Figures 1B, 2B, 3B, 4B,E and 5B,C). urn:lsid:zoobank.org:pub:90A627F2-9ECB-4ACE-81FC-9AC006D90656 Diagnosis: Sc ends distinctly beyond Rb bifurcation at R1 and Rs; R4+5 3× longer than Rs and almost 2× longer than R2+3+4; M1 about 4.5× longer than M1+2; gonostylus almost equal in length to gonocoxites, strongly bifurcated at end, the upper arm of bifurcation almost 2× longer than lower one.
Etymology: The specific epithet is given after name of the granddaughter, Mia, of one of the authors (WK).
Material examined: Holotype (male), No. MP/4075, Burmese amber; deposited in the collection of the ISEA PAS.
Description: Wing length 2.9, width 1.2 ( Figure 1B). Head. Antennae with 16 segments; scapus tubular; pedicel short and oval; flagellomeres increasing in length and the last segment nearly 5× longer than its width; palpi short ( Figure 4B). Thorax: Wing about 2.7 times longer than its width; Sc ends beyond fork of Rb into R1 and Rs; approximately opposite ⅓ of length of Rs; R1 ends opposite half the length of R2+3, near tip of R2+3; R2+3 nearly straight; R4+5 more than 3× longer than Rs and almost 2× longer than R2+3+4; r-m short, equals ⅙ of length of Rs, located distinctively before middle of length of M1+2; Mb present, clearly visible; M1 more than 4× longer than M1+2; m-cu just beyond fork of Mb, located between M3+4 and Cu; Cu at end strongly waved; pseudovein (ps) clearly visible; A1 with apical half strongly bent to wing margin ( Figure 3B). Legs: foreleg and midleg with single spur that are nearly 3.5 times the width of tibia; hind leg with two spurs of unequal length ( Figure 4E). Hypopygium: Gonocoxites long and narrow (length about: 0.28 mm); gonostylus almost equal in length to gonocoxites (length about: 0.27 mm), forked at end, the upper arm of bifurcation almost 2× longer than the lower ( Figures  2B and 5B,C).
Description: Wing length 1.6 mm, width 0.6 mm ( Figure 1A). Head: Antennae with 16 segments; scapus wide and barrel-shaped; pedicel oval; flagellomeres almost 1.5× longer than its width, and the last segment almost 3× longer than its width; palpi short ( Figure 4A). Wing almost 3× longer than its width; Sc ends just beyond fork of Rb into R 1 and Rs; R 1 ends opposite half the length of R 2+3 , near tip of R 2+3 ; R 2+3 distinctly waved at mid-length; R 4+5 more than 3× longer than Rs and 2× longer than R 2+3+4 ; r-m short, equals 1 /5 of length of Rs, located in 1 / 3 length of M 1+2 ; Mb present, distinctly visible; M 1 about 2.5× longer than M 1+2 ; m-cu just beyond fork of Mb, located between M 3+4 and Cu; Cu slightly arched; pseudovein (ps) clearly visible; A 1 visible only in basal part ( Figure 3A). Legs: foreleg with a single spur almost 3.5 times the width of tibia; hind and middle leg with two spurs of unequal length ( Figure 4D). Hypopygium; Gonocoxite broad, greatly expanded in basal part (length about: 0.12 mm); gonostylus slightly longer than gonocoxites (length about: 0.14), strongly forked at end, the upper arm of bifurcation much shorter than the lower one (Figures 2A and 5A).
Etymology: The specific epithet is given after name of the granddaughter, Mia, of one of the authors (WK). Description: Wing length 2.9, width 1.2 ( Figure 1B). Head. Antennae with 16 segments; scapus tubular; pedicel short and oval; flagellomeres increasing in length and the last segment nearly 5× longer than its width; palpi short ( Figure 4B). Thorax: Wing about 2.7 times longer than its width; Sc ends beyond fork of Rb into R 1 and Rs; approximately opposite 1 / 3 of length of Rs; R 1 ends opposite half the length of R 2+3 , near tip of R 2+3 ; R 2+3 nearly straight; R 4+5 more than 3× longer than Rs and almost 2× longer than R 2+3+4 ; r-m short, equals 1 /6 of length of Rs, located distinctively before middle of length of M 1+2 ; Mb present, clearly visible; M 1 more than 4× longer than M 1+2 ; m-cu just beyond fork of Mb, located between M 3+4 and Cu; Cu at end strongly waved; pseudovein (ps) clearly visible; A 1 with apical half strongly bent to wing margin ( Figure 3B). Legs: foreleg and midleg with single spur that are nearly 3.5 times the width of tibia; hind leg with two spurs of unequal length ( Figure 4E). Hypopygium: Gonocoxites long and narrow (length about: 0.28 mm); gonostylus almost equal in length to gonocoxites (length about: 0.27 mm), forked at end, the upper arm of bifurcation almost 2× longer than the lower ( Figures 2B and 5B,C).
Etymology: The specific epithet is given to emphasize the large size of the specimen (from the Latin magnifico/feminine magnifica/, meaning magnificent or gorgeous).
Material examined: Holotype (male), No. MP/4076-Burmese amber; deposited in the collection of the ISEA PAS.
Description: Wing length 3.8 mm, width 1.8 mm ( Figure 1C). Head: antennae with 16 segments; scapus tubular; pedicel short and oval; flagellomeres nearly 2× longer than their width; palpi relatively long, segments of nearly the same length ( Figure 4C). Wing about 2 1 / 3 × longer than its width; Sc ends far beyond fork of Rb into R 1 and Rs, opposite 2 /3 of length of Rs; R 1 ends opposite half the length of R 2+3 , near the tip of R 2+3 ; R 2+3 distinctly waved at basal part; R 4+5 more than 3× longer than Rs and about 2× longer than R 2+3+4 ; r-m short, equals 1 /8 of length of Rs, located close to middle of length of M 1+2 ; Mb present, clearly visible; M 1 3× longer than M 1+2 ; m-cu more than its own length beyond fork of Mb, situated between M 3+4 and Cu; Cu at end strongly bent; pseudovein (ps) clearly visible; A 1 slightly wavy at the middle of its length ( Figure 3C). Legs: front leg with a single spur, middle and hind leg with two spurs of nearly the same length and nearly 3× longer than the width of tibia ( Figure 4F). Hypopygium: Gonocoxite broad and greatly expanded in the basal part; gonostylus almost equal in length to gonocoxite, forked at the end, processes of equal length, strongly sclerotized ( Figures 2C and 5D).

Discussion
The type species of the genus Paleoplatyura, P. macrocera, from Baltic amber, differs from all congeners in Burmese amber by the short vein Sc, which ends clearly before the bifurcation of Rb into R 1 and Rs. Surprisingly, the Burmese amber species of Paleoplatyura are thus more similar in this respect to the extant species P. johnsoni than to the Baltic amber type species. However, the relative length of Sc also differs among various species of Macrocera Meigen, 1803 [13], and similar variation is known in some genera of Mycetophilidae, so that this character appears as species-specific rather than of fundamental phylogenetic importance. We thus prefer to maintain the concept of Paleoplatyura as defined in this paper; which means including intrageneric variation of the length of Sc, considering the well-known fact that higher taxonomic categories, like genus or subfamily, are usually more or less subjective, and their concept (breadth) may differ, even within one family. The subjective concept of genera, with different breadth defined by various authors, may also be a key to the understanding why Mesozoic genera, such as Paleoplatyura, are (seemingly) present in both the Tertiary and extant fauna.
The presence of extant genera in the Mesozoic fauna is a remarkable phenomenon itself. Recent studies of Burmese amber inclusions increasingly indicate the origin of modern genera as early as in the Cretaceous. Examples of extant insect genera found in the Mesozoic are well documented, though not common. Grimaldi and Cumming [28] stated that Apalocnemis canadambris Grimaldi and Cumming [28] (family Empididae), was the only species out of 49 species studied within their paper on Cretaceous ambers Brachycera belonging to an extant genus. In Diptera, several other similar cases are known from Burmese amber, e.g., Antocha lapra Podenas and Poinar [29], and Helius lebanensis Kania, Krzemiński and Azar [30], from the family Limoniidae, or Nemopalpus quadrispiculatus Stebner et al. [31], and Phlebotomus vetus Stebner et al. [31], from the family Psychodidae. In some cases, a recent genus serves only as a "wastebasket taxon", to tentatively place a problematic species in a described genus, as is the case of some genera of Mycetophilidae in Cretaceous ambers [32]. A similar situation is reported in [33], e.g., for the click-beetle genus Elater Linnaeus [25]. Some taxa of Diptera appeared even earlier, e.g., the genus Protanyderus Handlirsch, 1909 (family Tanyderidae), in the Upper Jurassic of Mongolia [34].
The extant species Paleoplatyura johnsoni, which unambiguously belongs to Paleoplatyura, was described from North America, and recently found also in Europe (southern Italy, see [12]). It is a large species (wing length is 8 mm), significantly larger than the fossil congeners, with the wings strongly marked, and relatively short antennae, only slightly longer than the head and thorax together. Surprisingly, there are no new specimens of Paleoplatyura available from Baltic amber. In contrast, several specimens from this genus, belonging to the three species described in this paper, have been found in Burmese amber. A major problem in the study of fossil Keroplatidae, is the loss of most of the holotypes of previously described species, especially from Baltic amber, and the lack of a recent, comprehensive morphological study of Keroplatidae belonging to the tribe Orfeliini. Additionally, the concept of the tribes Keroplatini and Orfeliini appears as outdated in the light of modern molecular studies [1].
A similar wing venation and structure of the male terminalia as in Paleoplatyura are found in the genus Asynaphleba Matile, 1974 [35], containing a single South African extant species, which differs from Paleoplatyura only by the absence of the vein Mb and by shorter anal vein, not reaching the wing margin. Thus, it seems that the presence of both Mb and the cross-vein r-m are good diagnostic characters for the genus Paleoplatyura. Taking into consideration only a distinct vein Mb, as a clear plesiomorphic character state, it is not unique to Paleoplatyura within keroplatids, but it is present also in several genera of Macrocerinae, including the species-rich genus Macrocera, and also in the South African genus Schizocyttara Matile, 1974 [35], which was shown to be closely related to Paleoplatyura johnsoni by Mantič et al. [1], although it lacks cross-vein r-m. Both Mb and cross-vein r-m are well developed also in the genus Arachnocampa Edwards, 1924 [36], from the most plesiomorphic keroplatid subfamily Arachnocampinae, which, however, differs from Paleoplatyura in some other characters, such as the absence of R 2+3 and different proportion of wing veins.

Conclusions
Genus Paleoplatyura represents one of the most ancient and plesiomorphic genera of Keroplatidae, with the most complete wing venation representing a ground plan within the family (together with genus Arachnocampa). This group possesses also the most plesiomorphic structure of male terminalia, which appears to be simple with long and apically forked gonostyli. Molecular data indicate a close relationship of Paleoplatyura with Macrocerinae [1]; additionally, if we compare the wing of Paleoplatyura with that of Macrocera, we can see many similarities, confirming the evolutionary trend to the reduction of radio-medial cross-vein to the so-called radio-medial fusion, typical of most keroplatids, together with prolongation of the antennae in Macrocera and elimination of some wing veins in Lygistorrhininae. The basal part of medial vein is still retained in most Macrocerinae, while it is absent in almost all species of the more apomorphic subfamilies of Keroplatinae and Lygistorrhininae. The overall structure of male terminalia is also very similar in Paleoplatyura and most Macrocerinae.