Recent discoveries of new Elephantomyia (Diptera, Limoniidae) fossils in Baltic amber

New data on the genus Elephantomyia (Diptera: Limoniidae) from Baltic amber are presented. A new subgenus Hoffeinsonia subgen. nov. is established with one new species: Elephantomyia (Hoffeinsonia) prima sp. nov. The new subgenus is characterized by a wing at most 2.5 × as long as it is wide without a darker pattern along the veins Sc and R1, elongate Sc, straight vein R1, sharp half of vein R2+3+4 sharply arched to the upper edge of the wing, short, wide, trapezoidal d-cell and oval pterostigma. The fossil subgenus Hoffeinsonia subgen. nov. shares features with the extant subgenera Elephantomyodes and Elephantomyia. One other extinct species of Elephantomyia was discovered and described herein as E. (s. str.) christelae sp. nov. Such features as a very elongate vein R2+3+4, 2.5 × as long as the Rs easily allowing this new species to be distinguished from the other fossil representatives of the genus Elephantomyia. The taxonomic decision on Elephantomyia grata as a species placed in nominative subgenus is provided. A list and key of fossil species of Elephantomyia are given. The morphological pattern of the genus is discussed in relation to the adaptation to a specific food spectrum, coevolution with Angiospermae of the representative genus Helius known since Cretaceous and closely related to this genus representatives of the much younger genus Elephantomyia.

From fossil record seven species of Elephantomyia are known, six of them are known to have come from Eocene Baltic amber and belong to nominative subgenus Elephantomyia [6][7][8] . We only know of one species of Elephantomyia from Miocene to this day unclassified to any subgenus-Elephantomyia grata Podenas & Poinar 9 (Table 1; Fig. 1). None of the other three subgenera of Elephantomyia: Elephantomyina, Elephantomyodes, Xenoelephantomyia have so far been found in the fossil record.
Baltic amber forms the largest amber deposit in the world and is a relevant source for fossil insects. Diptera are largely dominant and diverse among animal inclusions in this kind of resin what may be related to the fact that the deposits of the Eocene Baltic amber were formed over a relatively long time and under various environmental conditions 10 .
The new materials under investigation made it possible to discover a representative of a new subgenus of Elephantomyia in the fossil material. This new, peculiar discovery from Baltic amber described herein, provides evidence of the existence of craneflies belonging to a new subgenus in the Eocene and this is the first case to confirm the existence of a subgenus other than nominative the Elephantomyia in the past.  www.nature.com/scientificreports/ Diagnosis Antennae 14-segmented; rostrum shorter than wing, only slightly longer than 1/2 of wing, longer than abdomen; palpus longer than glossal lobes; R 2+3+4 very elongate, 2.5 × as long as Rs; d-cell short and wide, 1.5 × as long as wide; m-cu just before half the length of d-cell; M 3 2 × longer than d-cell; vein m-m very short, almost completely reduced; vein Rs relatively short, length of vein Rs arranges only about twice the length of the basal deflection of R 5 ; shorter than R 2+3+4 .
Etymology The new species is dedicated to Christel Hoffeins from Hamburg, Germany, the amber collection owner and expert of the Baltic amber inclusions.
Horizon and locality The age of Baltic amber has been a matter of debate for many years 9,16-23 . But, the most current state of knowledge is that it is of Priabonian age 17 . This means it is between 38 and 34 million years old (based on pollen, spores and phytoplankton of the amber embedding layer, the Blue Earth). The age of Baltic amber has also been estimated to approximately 47-41 Ma, which is mainly based on a study by Ritzkowski 24 ; however, the reliability of the methods used in his study have been questioned due to contaminations that can lead to older age estimations 16,17 . The age range of all Baltic amber bearing strata possibly cover 48 to 23 million years, and it is still debatable [18][19][20][21][22][23].
Description. Body (Fig. 3A): brown with dark distal part of abdomen darker than rest of body, body 3.28 mm long (without rostrum).
Head ( Figs  ) brevipalpa is as long as, or longer than vein R 2+3+4 , in contrast to E. (E.) christelae sp. nov. where Rs is distinctly shorter than R 2+3+4 . E. (E.) christelae sp. nov. differs also from the other species of Elephantomyia in the ratio between wing, rostrum, and abdomen length. In E. (E.) christelae sp. nov. rostrum is slightly longer than the abdomen, being longer than half wing length, but shorter than wing.
Hoffeinsonia subgen. nov. Etymology The new subgenus is dedicated to Christel Hoffeins from Hamburg, Germany, the amber collection owner and expert of the Baltic amber inclusions.
Description As for species.
Comparison What occurs in Elephantomyina is a strong supernumerary cross-vein connecting vein R 2+3+4 shortly before tip of latter, cross-vein r-m connecting with Rs a short distance before its fork 2 , while in other subgenera of the genus Elephantomyia (including Hoffeinsonia subgen. nov.) the supernumerary cross-vein does not occur and cross-vein r-m connecting with Rs beyond of Rs. Moreover, as mentioned by Osten Sacken 25 : "anal field of wing reduced in area, with a single vein", while in three other known subgenera of Elephantomyia and Hoffeinsonia subgen. nov. two well developed annal veins are observed.
Only the hind leg with part of the tarsus of the specimen of Hoffeinsonia subgen. nov. is preserved, but is clearly visible that basal and middle part of femur and part of tibia are pale. This feature and others such as: vein R 1 straight, basal half of vein R 2+3+4 sharply arched to the upper edge of wing, veins R 2+3+4 and R 1 running closer together than veins R 2+3+4 and R 5 , reduced palpi are similar to these which occur in subgenus Elephantomyodes.   (Figs. 5, 6, and 7). Diagnosis Antennae 15-segmented; rostrum shorter than wing, shorter than 1/2 of wing, shorter than abdomen; palpus shorter than glossal lobes; R 2+3+4 1.5 × as long as Rs; d-cell approximately 1.5 × as long as wide; m-cu Horizon and locality as for E. (s.str.) christelae sp. nov. Description Body 3.61 mm long (Fig. 6A). Head (Fig. 6B, C) width 0.42 mm, 0.38 mm high; rostrum 1.36 mm long, approximately as long as half the body length, shorter than half wing; length of antenna 0.72 mm (Fig. 5A); scape elongate, cylindrical, pedicel oval, wider than scape and other antennal segments, flagellomeres 1 and 2 only slightly elongate, longer than wide, flagellomeres 3-15 elongate, longer than twice of its width with very elongate setae, approximately twice Abdomen (Fig. 6A) 2.04 mm long; hypopygium 0.37 mm long with lobe-short, small, comparable length gonostyles, outer gonostylus (branch II = clasper of gonostylus; dorsal gonostylus) narrow, tappered at the end, strongly sclerotized, inner gonostylus (branch II = lobe of gonostylus; dorsal gonostylus) widened at the base and in its middle, narrowed at the end (Fig. 6D), less sclerotized than the outer gonostylus.
Remarks Well preserved holotype specimen, but only partially preserved legs. Key to fossil species of the genus Elephantomyia.

Discussion
Craneflies from the genus Elephantomyia to present were represented in the fossil record only by the subgenus Elephantomyia. The most well known species from the fossil record were described on the basis of inclusions in Eocene Baltic amber. We do not know the older representatives of the genus Elephantomyia. One species was described from a younger period (the Miocene) 9 (Fig. 8) and to this day it has not been classified under any subgenus of the genus Elephantomyia. After careful analysis, it was possible to indicate herein that this species, like the other species described based on fossil material, belongs to the nominative subgenus Elephantomyia. So, of the four known subgenera of the genus Elephantomyia, only one is represented in the fossil record. The subgenus Elephantomyia is also the most diverse in modern fauna 5 .
There are over 100 species belonging to the subgenus Elephantomyia occurring in modern fauna. Subgenera like Elephantomyina and Xenoelephantomyia are represented by only single species 5 .
The characteristic morphological features of the newly discovered specimens mentioned in this paper allows the indication and description of a new subgenus of Elephantomyia.
The general features which characterize the genus Elephantomyia are: very elongate rostrum, sometimes longer than the body length, maxillary palpus four segmented, first palpomere reduced, atrophy of cross-vein r-r (R 2 ) and comparatively short and wide gonocoxites with small gonostyles 25,[29][30][31] . The combination of features such as legs at least partially pale, these parts almost white, vein R 1 is straight, basal half of vein R 2+3+4 is sharply arched to the upper edge of wing, veins R 2+3+4 and R 1 are positioned closer together than veins R 2+3+4 and R 5 and pale brown oval pterostigma, wing without colour pattern distinguish the new subgenus from other subgenera within the genus (Fig. 8).
It is also worth noting that the fossil subgenus Hoffeinsonia subgen. nov. shares features with the extant subgenera Elephantomyodes [such as basal half of vein R 2+3+4 sharpy arched to the upper edge of wing, veins R 2+3+4 and R 1 running closer together than veins R 2+3+4 and R 5 and straight R 1 , reduced palpi, legs with very light (almost white) parts)] and Elephantomyia (such us oval pterostigma and wing distinctly wider than in Elephantomyodes) (Fig. 9). These similarities can indicate a phylogenetic relationship of Hoffeinsonia subgen. nov. and Elephantomyodes or Elephantomyia. The colour patterns of legs are rather rare in craneflies belonging   (Fig. 10). Fossil records prove the existence of these insects as early as the Eocene. The discovery of the new subgenus sheds new light on the diversity of the genus Elephantomyia, the evolution of the Limoniidae and introduces important information that can be used in further research on the phylogenetic relationships of this group of insects.
The presence in Baltic amber of inclusions of flies of the genus Elephantomyia with a very elongate rostrum adapted to a specific food spectrum (most likely the nectar of Angiospermae flowers) can provide evidence that in "Baltic amber forests" were plants pollinated by these insects. This co-evolution of Angiospermae and Limoniinae began much earlier, in Cretaceous. From Cretaceous period are known species of genus Helius which are characterized by an elongate rostrum -flies adapted to feed on the nectar of Angiospermae flowers. This genus is closely related to Elephantomyia 33, 34 .
It has been also proven that the Baltic amber flora comprises elements of both extant northern American and East Asian warm-temperature flora in "Baltic amber forests" and humid climate 20 . Recent species of subgenus Elephantomyia occur mainly in Neotropical and Afrotropical regions 3 and are represented as an inclusions in Baltic amber while Elephantomyodes occur in Holarctic region. But, its absence in Baltic amber may be due to the fact that these insects were very rare in the Eocene, as in modern fauna 5 .

Material and methods
The study was based on four inclusions in Eocene Baltic amber from the collection of Christel and Hans Werner Hoffeins. The holotypes of new described species herein are deposited in Senckenberg Deutsches Entomologisches Institut (SDEI) Müncheberg, Germany. The specimens were examined using a Nikon SMZ 1500 stereomicroscope equipped with a Nikon DS-Fi1 camera. The measurements were taken with NIS-Elements D 3.0 software. The length of head was measured as length of head capsule excluding rostrum. The length of the discal cell-measurements were given from its posterior edge to the point of connection of vein m-m with vein M 3 , the length of the vein M 3 -measurements were given from the point of connection of vein m-m with vein M 3 to the margin of wing. The length of hypopygium was measured from the posterior margin of tergite IX to the apex of gonocoxite. The measurements and the relationship between the length of rostrum, wing and abdomen were given only in case when relevant structures were not distorted. Drawings were made by tracing the specimens and the photographs. Drawings and photographs (Figs. 2, 3, 4, 5, and 6, graphics 8-10 partially) were made by Iwona Kania-Kłosok. The map was built using the map Maps-For-Free (https:// maps-for-free. com) and modified with the software programs Corel Draw and Corel Photopaint X7. The stratigraphic chart was used according to International Stratigraphic Chart, International Commission of Stratigraphy (v. 2021/05) https:// strat igrap hy. org/ chart. The wing venation follows that of 38 , terminology applied to the male genitalia nomenclature, is in accordance 8,30,39 .