Wanted, tracked down and identified: Mesozoic non-biting midges of the subfamily Chironominae (Chironomidae, Diptera)

Here we provide evidence for the presence of non-biting midges of the subfamily Chironominae in the Mesozoic, based on descriptions of exceptionally rare amber inclusions. The subfamily has already been reported, but based on a single and unspecified record from Late Cretaceous Taimyr amber (~84 Mya). That record is here revised and confirmed. Moreover, a new find in Burmese amber locates the temporal boundary of the subfamily back to the mid-Cretaceous (probably ~100 Mya). We describe two new genera assigned to the tribe Pseudochironomini: Mesoacentron gen. nov. with the species Mesoacentron kaluginae sp. nov. (Taimyr amber) and Palaeocentron gen. nov. with the species Palaeocentron krzeminskii sp. nov. (Burmese amber), the oldest known representative of the subfamily. The systematic position of the new taxa is discussed, and a key to the identification of adult males of extinct and extant Pseudochironomini genera is presented. Insights in the phylogeny and diversification tempo of the Chironominae and Pseudochironomini in the past are also provided.


INTRODUCTION
The Chironomidae (non-biting midges) are a large family of dipterans with nearly 7500 species and 550 genera described (Pape et al., 2011; plus data published thereafter).Although their larvae may be terrestrial or semi-aquatic, the majority are strictly aquatic developing in freshwater habitats, both lotic and lentic (Armitage et al., 1995).Many species have narrow tolerances for temperature, trophic state, salinity or acidity, which renders them important bioindicators of environmental changes and water quality.This allows fossil chironomids to be used in reconstructing histories of environmental changes from the near to more distant past (e.g.Velle et al., 2005;Verschuren & Eggermont, 2006;Stebner et al., 2017).
Cretaceous Chironominae have been recognized to the present only from a note by Kalugina listing a single specimen (Kalugina, 1974: 48, table 2), but with no further data.Fifty years later we describe two genera: one based on Kalugina's original material, here re-examined in detail, and the second based on a recently obtained specimen.Both are assigned to the tribe Pseudochironomini.

GeoloGical context
The specimens studied are adult male inclusions found in Cretaceous ambers from Asia, in Russia (Taimyr amber from Yantardakh; Upper Cretaceous) and in Myanmar (Burmese amber from Hukawng Valley; mid-Cretaceous).
Yantardakh is the largest and best-studied insect-bearing site of the Kheta Formation [Taimyr Peninsula, the right bank of River Maimecha (a tributary of the Kheta), 3 km upstream of its mouth, at Yantardakh Hill (71°18'26"N, 99°33'46"E)].The deposits date back to the Santonian (Zherikhin & Sukatsheva, 1973;Rasnitsyn et al., 2016).The Chironomidae described herein were found among about 3000 arthropod inclusions by the expedition of the Paleontological Institute of the Academy of Science of the USSR in 1970-71.All the Chironomidae collected in Yantardakh at that time were sorted and identified by N. S. Kalugina in subfamilies as follows: Orthocladiinae s.s.-599, Aphroteniinae -50, Tanypodinae -22, Diamesinae -three, Podonominaeone and Chironominae -one; but only Aphroteniinae and Diamesinae were described (Kalugina, 1974(Kalugina, , 1976(Kalugina, , 1980b)).Although a single Chironominae specimen was mentioned, actually two individuals of this subfamily were collected and marked by Kalugina, including one identified tentatively because of its poor state of preservation.Both specimens are described here for the first time.
Burmese amber, or Burmite, is predominantly mined in the Hukawng Valley, in the district of Myitkyina, State of Kachin in the north of Myanmar (Xing & Qui, 2020).Due to redepositions, its geological age has long been debated, but a dating of Burmite from Kachin to near the Albian-Cenomanian boundary is now welljustified (Rasnitsyn et al., 2016).The number of insect inclusions described from Burmite has increased greatly over the past 20 years.About 400 insect families have been recorded, rendering the Burmite fauna the richest Cretaceous fossil resin arthropod assemblage known to date (Rasnitsyn et al., 2016;Ross, 2019Ross, , 2020)).As usual in Cretaceous ambers, chironomids are numerous in Burmite, although only several members of the subfamilies Podonominae, Buchonomyiinae and Orthocladiinae have been described (Baranov et al., 2017;Giłka & Zakrzewska, 2017;Giłka et al., 2019Giłka et al., , 2020)).In contrast, to date no Chironominae have been found in Burmite.The specimen described here is thus the oldest known representative of the subfamily.
techniques used and repository data Amber pieces were ground and polished following the protocols of Sidorchuk (2011) and Sidorchuk & Vorontsov (2018) (Taimyr amber) or manually (Burmese amber), so that the inclusions and their diagnostic structures could be examined and photographed at high magnification.Owing to the exceptionally fragile nature of the Taimyr amber, the pieces examined were embedded in Buehler's EpoThin (an epoxy resin).The epoxy embedding of amber inclusions, routinely practiced for more than 20 years (Nascimbene & Silverstein, 2000), is believed to be the safest way to deal with brittle amber pieces and to protect them during prolonged storage.However, epoxy resin readily fills not only fissures in the amber, which would be one of the purposes of embedding, but also those parts of the inclusion that are open to the surface of the piece.The resin-filled parts of the inclusion become less visible, sometimes almost transparent, yet can be imaged with optical contrasting methods.For this reason, the Taimyr amber pieces were imaged both before and after the embedding.
Specimen dimensions are given in micrometres, except for the total length (in millimetres, rounded off to the first decimal place).The body and wing lengths were measured from the antennal pedicel to the end of the gonostylus and from the arculus to the tip, respectively.The lengths of leg segments and palpomeres were rounded to the nearest 5 and 1 μm, respectively.The antennal, leg and venarum ratios (AR, LR and VR) were calculated to the second digit after the decimal point.The morphological terminology and abbreviations follow Saether (1980), with additions for hypopygial structures proposed by Cranston (2019b).Photographs were taken using a PZO Biolar SK14 microscope with a Sony NEX-3N digital camera (Burmese amber) and a Nikon E-800 microscope, with dry (×4, ×10) and water immersion optics (×40 and ×60) and an Olympus OM-D E-M10II digital camera (Taimyr amber).Brightfield, differential interference contrast and incident illumination were used.The stacks were corrected for colour, brightness and noise with Adobe Lightroom.The images were compiled using the Helicon Focus 6 stacking software.
For the amber not embedded in epoxy resin, bought on the open market, the Fourier transform infrared spectroscopy (FT-IR) spectrum [attenuated total reflectance (ATR)] was obtained to support identification of the Myanmar inclusion examined (see Discussion: infrared (IR) spectrum).A Nicolet iS10 FT-IR spectrometer was used to record the ATR (diamond crystal) FT-IR spectrum.The resolution was 4 cm -1 ; 16 scans were performed.The spectrum with reference curve numbers obtained is registered and stored in the archives of the Laboratory of the International Amber Association (IAA), Gdańsk, Poland and at the Laboratory of Systematic Zoology, Department of Invertebrate Zoology & Parasitology (LSZ DIZP), University of Gdańsk, Poland.
The Yantardakh amber inclusions studied are housed in the collection of the Borissiak Paleontological Institute, Russian Academy of Sciences, Moscow (PIN).The specimen preserved in the Burmese amber is the property of the Natural History Museum, Institute of Systematic and Evolution of Animals, Polish Academy of Sciences, Kraków (ISEZ PAN) enriched with a collection of amber inclusions from Myanmar in 2015.

RESULTS
systematics Family: chironomidae newman, 1834 subFamily: chironominae newman, 1834 tribe: pseudochironomini saether, 1977 In adult Chironomidae, the lack of the wing vein MCu is the synapomorphy for Chironominae + Orthocladiinae, a clade comprising the two largest, most diverse and probably youngest chironomid subfamilies, considered sister to the remaining Chironomidae combined (e.g.Saether, 2000;Cranston et al., 2012).A character state common to all the Chironominae concerns the gonostylus that is backward-directed and connected rigidly with the gonocoxite in most Chironominae, although able to flex slightly in a few taxa (Cranston et al., 1989;Tang & Cranston, 2019).Within four Chironominae tribes (Chironomini, Pseudochironomini, Tanytarsini and the recently erected Xiaomyiini), the character state best separating the tribe Pseudochironomini is the dark comb on the apex of the foreleg tibia, similar to that on mid-and hindlegs (a probable tribal synapomorphy) and the pars ventralis, the structure being, however, weakly developed or absent in most genera.
Following reports on a distinct heteromorphy within the Pseudochironomini, especially with respect to the adult genital morphology, as highlighted recently by Krosch et al. (2020), the significance of diagnostic characters or character combinations of a tribe should be treated with the highest caution.The wing anal lobe usually well-developed (vs.anal lobe weak or absent in the Tanytarsini, but also in some Pseudochironomini genera), the vein RM oblique (vs.RM parallel to M and R 4 + 5 in the Tanytarsini or transverse to M and R 4 + 5 in most other Chironominae) and the foreleg ratio (LR 1 = ta 1 /ti) usually close to, or less than 1 (vs.higher than 1 in most Chironominae) are the character states in Pseudochironomini that may support the tribe definition.However, these features, and the presence of the pars ventralis, rarely appear as a full set, hence their proper interpretation usually requires a reference to taxa of other Chironominae that may display character combinations or some character states similar to those listed above.Owing to the difficulty in establishing a clear diagnosis based on a set of synapomorphies for the tribe (in the current taxa composition), its monophyly is still being tested but not confirmed (Cranston et al., 2012;Cranston 2019a, b;Tang & Cranston, 2019;Krosch et al., 2020).To date, the following six extant genera are included in the tribe Pseudochironomini (Andersen, 2016;Krosch et al., 2020): Aedokritus Roback, 1958, Manoa Fittkau, 1963, Madachironomus Andersen, 2016, Pseudochironomus Malloch, 1915, Riethia Kieffer, 1917, and Megacentron Freeman, 1961, the latter with Megacentron eocenicus Doitteau & Nel, 2007 -the sole Pseudochironomini known from the fossil record (Eocene Oise amber) (Doitteau & Nel, 2007).Descriptions of two further genera, based on unique characters found in specimens preserved in the Cretaceous ambers, are presented below.

Derivation of the name:
The specific epithet is dedicated to Nadezhda Sergeevna Kalugina (1930-90), eminent Russian palaeoentomologist and dipterologist, who first recognized the specimen to the subfamily level.
Diagnosis: Adult male.Small species, total length
Thorax (Fig. 1B, E): Scutum strongly developed, protruding; head deeply embedded between scutum and sternum.Ac at least 6; Dc at least 8 on each side; Scts at least 6, Pa not observed.

Notes
The wing with an extensive, round anal lobe, the squama bearing a fringe of setae, the RM vein oblique relative to M and R 4 + 5 and the genital apparatus with two pairs of volsellae, digitus, pseudovolsella and a well-developed pars ventralis form a set of characters testifying to the position of the new species within the Pseudochironomini, yet are unique in the tribe.The arrangement of hypopygial appendages in a set typical of Pseudochironomini, but with a pseudovolsella instead of the median volsella (see below), combined with a strongly elongated, narrow anal point have not been found in any extant/extinct chironomid and thus support our decision to erect the new genus, Mesoacentron.The antenna of previously described Pseudochironomini males consists of 13 flagellomeres, the only exception being 14 flagellomeres found in the extinct Megacentron eocenicus (Doitteau & Nel, 2007).
The species was placed in Megacentron based on the reference to the main generic characters, including the number of antennal segments, as defined by Freeman (1961) and Cranston et al. (1989).However, Freeman's diagnoses relied on the old style of antennal segment counting, whereby the pedicel was included into the total count.Therefore, the Megacentron described by Freeman have an antenna with a pedicel and 13 flagellomeres, the number (13) resulting also from the modern way of description (Cranston et al., 1989).The inconsistency in referring to this feature and a definition of the volsellae arrangement (including the lack of data on the superior volsella) may call into question the generic assignment of the species described by Doitteau & Nel (2007).However, it cannot be ruled out that the number of flagellomeres in Megacentron eocenicus, higher than that in extant congeners, follows a trend similar to that known from other Chironominae genera, in which a higher number is interpreted as plesiomorphy (Zakrzewska & Giłka, 2014, 2015;Zakrzewska et al., 2016Zakrzewska et al., , 2020;;Zakrzewska & Jankowska, 2021).This contention seems to be supported by the antenna bearing 14 flagellomeres, recorded here in Cretaceous Pseudochironomini.Interestingly, a partial fusion of flagellomeres 1-4, observed in Mesoacentron (Fig. 1D), may be regarded as a transitional phase between the oldest known Pseudochironomini (bearing 14 flagellomeres, cf.Fig. 5B) and the extant ones (13 flagellomeres).
Mesoacentron lacks a median volsella, while a trifid protrusion or aggregation of tubercles with linearly merged bases is observed in the male hypopygium (Fig. 2G, H).Recently, some hypopygial structures in the Chironominae have been redefined by Cranston (2019b).According to that approach, a structure treated so far as the median volsella in some Riethia, cannot -due to its morphology and location -be regarded as homologous with the true median volsella, as it is more a cluster of several setal tubercles that lacks a stem and is located on the inner part of the gonocoxite.It is proposed to call the structure a pseudovolsella.On account of a considerable spatial depth of the hypopygium, the position of its appendages along the vertical axis is particularly difficult to determine in fossil specimens, even in amber inclusions mounted using modern methods.However, the linearly merged aggregation of tubercles observed in Mesoacentron kaluginae corresponds well with Cranston's description, thus it is identified here as the pseudovolsella.Cranston's suggestions regarding the terminology of the hypopygium appendage structures concern also a lobe of a variable size that is either separated from the inferior volsella or fused with it.A minute finger-like lobe placed at the base of the inferior volsella, under the superior volsella, was also observed in Mesoacentron (Fig. 2G, H).Following Cranston (2019b), we call the structure the digitus.The recent key to the identification of Pseudochironomini males (Strixino & Shimabukuro, 2018) uses the presence/absence of the digitus as a character that delimits two closely related genera, Riethia and Manoa.However, the digitus was later observed in both genera (Cranston, 2019b), thus, the only characters that enable separation of these taxa appear to be those of the immature stages.Consequently, our key to adult males treats the two genera jointly (see below).
Anal point slender, tapering towards tip.Remaining structures poorly observable or not visible.

Notes
The male of the species has the gonostyli directed backward, which justifies its assignment to the subfamily Chironominae.Placement in the Pseudochironomini is tentative on the basis of several character states known from the tribe (a welldeveloped and round anal lobe of the wing, and the RM vein oblique relative to M and R 4 + 5 ), as well as several features similar to those of Mesoacentron kaluginae, i.e. the tibial spurs and comb structure/arrangement and the wing venation pattern (with slight differences in the length of Sc and the length ratio/arrangement of R 4 + 5 and M 1 + 2 ).However, the two Taimyr specimens most likely represent different species.They differ distinctly in the size/length proportions of the main body parts (thorax and abdomen) and the presence/lack(?) of the pars ventralis, well developed in M. kaluginae but not observed in the 'Pseudochironomini?incertae sedis'.Unfortunately, the severely damaged specimen precludes its identification or accurate description as a potential new species.Nevertheless, the record is a valuable confirmation of the Chironominae being present in the Cretaceous Taimyr amber.

Ty p e m a t e r i a l : H o l o t y p e, M P / 4 0 2 0 ( I S E Z PAN)
: adult male (wings missing) preserved in a 17 × 13 × 3-mm piece of the Burmese amber (Hukawng Valley, Kachin State, Myanmar; probably Albian-Cenomanian, mid-Cretaceous, Fig. 4A.IR spectrum no.17497 IAA, Fig. 4B).
Legs (Fig. 6): Spur of foreleg tibia dark brown to black, stout (~30 μm long), pointed, with narrow dark comb at base (Fig. 6A, B).Mid-and hindleg tibia with two wellseparated combs, teeth up to 32 μm long on midleg and 34 μm long on hindleg, each comb bearing a straight or slightly curved spur, 32 and 42 μm long on midleg and 38 and 44 μm long on hindleg, spur and combs dark brown to black (Fig. 6C, D); hindleg tibia subapically bearing fan composed of five to six strong thorn-like bristles arranged as in Fig. 6E; 15-19 bristles of similar shape arranged in a row at two-thirds the length of the hindleg tibia (Fig. 6F, G).Length of leg segments and leg ratios in Table 1.

H y p o p y g i u m ( F i g. 7 ) :
G o n o s t y l u s s t r a i g h t , broadened in distal half, with apex rounded, ~100 μm long, similar in length to gonocoxite (Fig. 7A, B).Anal point stout, parallel-sided, distally rounded and excavated, with narrow spike-shaped median prolongation (Fig. 7C, D).Superior volsella narrow, finger-like, with distinct roundish swelling at midlength bearing one or two setae, distal part swollen and directed posteromedially (Fig. 7E).Inferior volsella fin-shaped, narrowed at base, broadest at midlength, tapering to pointed apex, with curved swollen ventral margin, and straight semi-transparent dorsal edge, covered with setae and microtrichia (Fig. 7F, G).

Notes
As mentioned above, the dark comb on the foreleg tibia is regarded as the key character (tribal synapomorphy) placing the new species in the Pseudochironomini.The comb at the base of the spur, observable in lateral view (Fig. 6A, B) and similar to combs on mid-and hindlegs (Fig. 6C,  D) but slightly narrower, in the new species is nearly identical to that known from, for example, Manoa (cf.Qi et al., 2017: fig.5d-f).Hereby, we consider assigning the new species to the tribe Pseudochironomini legitimate.However, in the present understanding of the importance and interpretation of other morphological diagnostic characters defining the Pseudochironomini, we see no justification in assigning the new species to any of the genera of the tribe known so far.Consequently, we propose that the species be included in the new genus, Palaeocentron, peculiar in having several distinct features discussed below.Mesoacentron and the remaining extant genera that display combinations of more complex and probably more advanced structures/character states.
In Megacentron eocenicus, the sole previously described extinct Pseudochironomini species, the general structure of the hypopygium (distinctly elongated, with the gonostylus broadened and rounded apically and the well-developed but narrow anal point) resembles that of Palaeocentron (cf.Doitteau & Nel, 2007: figs 49, 50).However, the three pairs of volsellae in Megacentron vs. two pairs in Palaeocentron clearly discriminate between the two genera.
The stout, parallel-sided hypopygial anal point with a spike-like extension (Fig. 7A, C, D), observed in the adult male of Palaeocentron, is unique within the Pseudochironomini.The anal point is absent in Manoa, Pseudochironomus and Riethia.Doitteau & Nel (2007) and Andersen (2016) gave a brief overview of shapes of the structure in other Pseudochironomini genera, including Madachironomus (a strong and nearly parallel-sided anal point, but with a rounded apex), Aedokritus (a relatively short triangular or subtriangular anal point) and Megacentron (a narrow anal point, slightly similar to that described in Mesoacentron; see above).
An unusual structure found in the Palaeocentron male is the hindleg armature: a row of strong, thornlike bristles placed along the tibia and forming a fan near its apex (Fig. 6D-G).Similar leg structures are known from closely related families (mostly in females), especially the predatory/haematophagous midges (e.g.Downes, 1978;Szadziewski et al., 2007;Urbanek et al., 2015).In these flies, the mouthparts usually are elongated and/or provided with structures used in biting and intake of liquid nourishment, blood or haemolymph.However, no predatory/haematophagous species have been so far recorded in the subfamily Chironominae that as adults feed on honeydew or are considered aphagous.Thus, the function of these thorn-like bristles in Palaeocentron, which features a shortened proboscis and mouthparts reduced as in a typical Chironominae, remains an open question (cf.Sivinski, 1997).

pseudochironominitempo oF diversiFication
Studies of Chironomidae have been prominent in estimations of phylogeny and dating of the tempo of their evolution.Brundin (1966Brundin ( , 1976) demonstrated how a phylogenetic hypothesis could be melded with geological evidence, obtained from both fossil specimens and biogeography from earth history.The conclusions drawn by Brundin (1966) have found substantial support for both the evolutionary relationships and the timing (e.g.Cranston et al., 2010Cranston et al., , 2012)), although reconstruction methods differ now with more explicit analyses and additional sources of evidence.Foremost have been advances in molecular techniques to allow reconstruction of phylogenies with many more character states available than from morphology alone, even compared with data derived from all life stages -the gold standard for morphology.Allied to this has been development from early (naive) ideas concerning a 'molecular clock' from which the tempo of evolutionary events could be reconstructed and tested against earth history events.With Bayesian analyses to the fore, hypotheses of relationships could be supplemented by complex models of molecular evolution, with calibration points derived from fossils, to estimate the tempo of change across any well-sampled clade.'Well sampled' evidently is critical, as material from which molecular phylogenies are derived ought to be freshly collected for study, in contrast to the more easily available historical slide-mounted morphological material: the inclusion of Pseudochironomini in contemporary studies is impeded by shortfall in the diversity sampled.It is worth noting that molecular data provide an independent timeline on major evolutionary events, but in no way can be argued to represent the 'truth'.Inferences of stem and crown ages of clades are only as accurate as the models of molecular evolution and the appropriateness of the calibrations selected and their ages based on fossil records.
In testing Brundin's ideas on Gondwanan-induced tempo using divergence time analysis, Cranston et al. (2010) estimated a sole species of Riethia to have separated from a solitary Polypedilum (representing Chironomini) at ~62 Mya, and these two combined (representing Chironominae) to have separated from Brillia (an Orthocladiinae) at ~100 Mya.Dates typically have wide statistical variances (HPD, highest posterior density intervals) of +/-30%.Increased sampling in Cranston et al. (2012), but with Pseudochironomini still undersampled (with only Riethia included), was inconclusive on monophyly and tribal relationships.Most recently, with formal recognition of the tribe Xiaomyiini as sister to all other Chironominae (Tang & Cranston, 2019;Krosch et al., 2020), and with Pseudochironomini enhanced by inclusion of Pseudochironomus, Krosch et al. (2020) inferred stem-group origination of Chironominae in the earliest Cretaceous and tribe Pseudochironomini separation from tribes Chironomini + Tanytarsini at the end of Early Cretaceous (~108 Mya).Median dates for origin of crown diversification of Pseudochironomini (still comprising only Riethia and Pseudochironomus) ~87 Mya, and Riethia originated at either ~61 Mya or ~52 Mya, depending on the position of two problematic Neotropical species.
T h e a d d i t i o n o f m o l e c u l a r d a t a f r o m Pseudochironomus in this study showed it to be strongly supported as sister to Riethia, yet the monophyly of the broader Pseudochironomini is tested only trivially, lacking the geographically disparate and rare additional proposed members of the tribe (see above) and likely to remain unattainable.
Regarding calibrations of molecular phylogenies with dated fossils, following discussions with those studying the Kalugina fossil described here, Krosch et al. (2020) included it as calibration of Chironominae less Xiaomyiini.Confirmation here as a member of the Pseudochironomini consolidates and refines their analyses, in which the fossil date was set at a minimum age of 71.5 Mya with the crown radiation estimated as 86.8 Mya with HPD of 68-112 Mya (Krosch et al., 2020)   -Wing membrane with distinct colour spots and/or crossbands, anal lobe well-developed, strongly protruding; median volsella posteriorly directed, with stem single-lobed (Oliveira & Messias, 1989: figs 2, 7;Pinho, 2018: fig. 1c, e;Pinho et al., 2019: fig. 1a, d)  Amber (or amber-like materials) offered by private sellers on the free market may be often substituted by fakes (e.g.Eriksson & Poinar, 2015).Moreover, the provenance of amber can be mistakenly assigned to geological deposits other than those from which it actually originated.Clearly this serious problem may also affect specimens designated as types for new taxa.Recording and archiving the amber IR spectra is thus a highly recommended procedure for museum materials to confirm the validity of the newly described taxa (Zakrzewska et al., 2020).One of the specimens examined in the present study was purchased on the free market, hence the analysis of its IR spectrum was vital.The FT-IR (ATR) spectrum obtained shows features typical of Burmite from Kachin, with the location and relative intensity of absorption peaks in fingerprint regions 1226, 1155, 1031 and 974 cm -1 (Fig. 4B) (E.Wagner-Wysiecka, personal communication).

CONCLUSIONS
The riddle of the presence of the Chironominae in the Cretaceous has been revisited and is now finally solved.The subfamily had been previously recorded only once, based on a single and unspecified record from Late Cretaceous Taimyr amber, the record that is now revised and identified.Finding a further member of the subfamily in Burmese amber demonstrates the group to have occurred as early as the mid-Cretaceous.All specimens studied have been assigned to the Pseudochironomini on the basis of a set of features putatively considered diagnostic for this tribe.Information provided by the specimens examined are in line with results of current research on the rate of diversification within the family/subfamily, based on molecular markers derived from extant representatives.All the data obtained so far support the hypothesis that the Chironominae appeared in the Cretaceous, diversifying subsequently throughout this period.
Extremely rare findings of Chironominae in the ambers studied here and the lack of data on these dipterans in other Mesozoic fossiliferous deposits may suggest that their abundance/species richness in the Cretaceous was still relatively meagre, while it increased significantly in the Eocene (see Introduction).

(
FiGs 1a-e, 2a-h) Z o o b a n k r e g i s t r a t i o n : u r n : l s i d : z o o b a n k .org:act:008459CE-58FD-474F-AB20-EF63844900A9.