MCNA 12068.4, from Peñacerrada I amber; fore- and hind wing distal fragments. Three associated hymenopterans are preserved as syninclusions.

Fore- and hind wing with a relatively long pterostigma with a strongly oblique and slightly sinusoid pterostigmal crossvein placed beyond pterostigmal midlength; fore- and hind wing with one closed radial cell distal to pterostigma; forewing Rs with six branches; forewing with at least eight closed subradial cells.

Sex unknown. Veins with some strong, very short setae preserved, membrane hyaline. Forewing.Length of preserved fragment 6.4, maximum width 2.8;wing apex relatively rounded; pterostigma relatively long (2.4 long, length ca. eight times basal pterostigmal width), slightly widening distally, not conspicuously infumate as preserved; pterostigma with a strongly oblique and slightly sinusoid pterostigmal crossvein placed beyond pterostigmal midlength, basally closed by a crossvein; pterostigma longer than any radial cell; R with two branches beyond pterostigma;at least three radial cells present, one closed radial cell partly distal to pterostigma; Rs with six branches and at least eight closed subradial cells;MA at least with two branches; gradate seriesveryregular, almost following a staircase-like pattern. Hind wing.Length of preserved fragment 5.9, maximum width as preserved 2.7;wing apex more pointed than in forewing;costal field distinctly narrower than in forewing; one c-sc crossvein preserved;pterostigma relatively long (2.5 long, length ca. 10 times basal pterostigmal width), slightly widening distally, not conspicuously infumate as preserved, starting 0.5 (twice pterostigmal basal width) beyond termination of Sc; pterostigma with a strongly oblique and slightly sinusoid pterostigmal crossvein placed beyond pterostigmal midlength, basally closed by a crossvein;pterostigma longer than any radial cell;R with two branches beyond pterostigma;at least five radial cells present, one small closed radial cell distal to pterostigma;Rs with five branches and at least seven closed subradial cells; MA at least with two branches; gradate seriesveryregular, almost following a staircase-like pattern.

The specific epithet is a combination of the Greek words cretaceus (taken from the period name, although specifically meaning “chalky”) and elektron, meaning “amber”.

Within the current taxonomic framework, the numerous crossveins of MCNA 12068.4 are indicative of placement in the Baissopteridae. Unfortunately, neither base of the wing is preserved, thus important characters such as the maximum width of the costal field, the pattern of distribution of c-sc crossveins (= costal crossveins), the separation between M and CuA in the forewing, and the shape of the basal piece of MA, are unknown. Also, the infumation of the pterostigma is not evident in the holotype, but it is uncertain if this could have been caused by taphonomical processes and is, therefore, not used as a diagnostic character although if the absence of infumation is true of the species in life, then it would represent a remarkable difference from all other described baissopterids. Fortunately, the pterostigma can be delimited thanks to the relative parallelness of C and R (R tends to conspicuously change its slope beyond the pterostigma in the other baissopterids) and also the greater thickness of both veins.

The specimen is tentatively classified within the genus Baissoptera as it has the pterostigmal crossvein most similar to the diversity found within this genus. Today just two of the 12 species currently classified within the genus Baissoptera, i.e., Baissoptera brasiliensis Oswald, 1990 and Baissoptera lisae Jepson, Ansorge & Jarzembowski, 2011, lack a pterostigmal crossvein in both fore- and hind wings (Oswald 1990: p. 156, figs 3, 4; Jepson et al. 2011: p. 393, text-figs 6A, B). Some genera can even show an additional, straight pterostigmal crossvein in a more basal position at least in the hind wing, i.e., Baissoptera kolosnitsynae Martynova, 1961 and Baissoptera pulchra (Martins-Neto and Nel, 1992) (Martynova 1961: p. 81, fig. 7; Martins-Neto and Nel 1992: p. 428, figs 2, 3). Regarding the other taxa currently classified within the Baissopteridae, the genera Lugala and Cretoraphidia lack a pterostigmal crossvein, at least in the hind wing (Ponomarenko 1988: p. 75, fig. 4; 1993: p. 70, figs 7, 9, 10); whereas the genera Cretoraphidiopsis and Austroraphidia, althoughshowing a pterostigmal crossvein situated beyond pterostigmal midlength, have it not as strongly oblique as in Baissoptera? cretaceoelectra sp. n., both showing irregular gradate series in both wings and a much lesser number of Rs branches in the forewing (three in Austroraphidia, four in Cretoraphidiopsis). Cretinocellia cellulosa Ponomarenko, 1988 has been recently transferred from the Baissopteridae to the Mesoraphidiidae by Bechly and Wolf-Schwenninger (2011) based on its lack of pterostigmal crossvein(s) and a Sc ending about midwing length. Although these two characters are also present in Cretoraphidia certa Ponomarenko, 1993 and Cretoraphidia magna Ponomarenko, 1993 (Ponomarenko 1993: p. 70, figs 7, 10), in both, the crossvenation is relatively higher than in Cretoraphidia cellulosa and should therefore remain in Baissopteridae for the moment. Consequently, Bechly and Wolf-Schwenninger (2011) also noted that the genus Cretinocellia might occupy a basalmost position within Mesoraphidiidae according to its relatively high crossvenation compared to the other mesoraphidiids. On the other hand, we still consider Arariperaphidia rochai Martins-Neto and Vulcano, 1989 as incertae sedis rather than as a baissopterid (contra Bechly and Wolf-Schwenninger 2011), owing to its lack of preserved characters indicating a more conclusive assignment. The shape and location of the pterostigmal crossvein is quite diagnostic for Baissoptera? cretaceoelectra. Only Baissoptera minima Ponomarenko, 1993 shows such a strongly oblique pterostigmal crossvein within the family, even slightly sinusoid as in the new species, in a relatively elongate pterostigma (length ca. eight times basal pterostigmal width) (Ponomarenko 1993: p. 64, fig. 2). However, the pterostigmal crossvein is located before pterostigmal midlength and Rs is poorly branched in Baissoptera minima. Baissoptera? cretaceoelectra has Rs in the forewing with more branches currently described within the genus. The remaining Baissoptera species always show a lesser number of branches of Rs in the forewing, i.e., five (Baissoptera brasiliensis, Baissoptera grandis Ren in Ren et al. 1995, Baissoptera liaoningensis Ren, 1994, Baissoptera lisae, and Baissoptera sibirica Ponomarenko, 1993), four (Baissoptera elongata Ponomarenko, 1993, Baissoptera euneura Ren, 1997, Baissoptera kolosnitsynae, and Baissoptera martinsoni Martynova, 1961), or three Rs branches (Baissoptera pulchra and Baissoptera minima) (Martynova 1961; Oswald 1990; Martins-Neto and Nel 1992; Ponomarenko 1993; Ren 1994, 1997; Ren et al. 1995; Jepson et al. 2011). Although Baissoptera cellulosa Ponomarenko, 1993 (based on a forewing lacking the apex) could also possess six branches of Rs and does have a sinuate pterostigmal crossvein presumably beyond pterostigmal midlength (Ponomarenko 1993: p. 65, fig. 3), it differs from Baissoptera? cretaceoelectra in that the pterostigmal crossvein is just slightly oblique and the more abundant crossvenation. The only other described baissopterid with such an elevated number of branches of Rs is Cretoraphidia certa, but it lacks a pterostigmal crossvein as has been discussed, and in addition Sc ends in a more basal position. Furthermore, the pterostigmal length of the new species is elongate when compared to the other lengths shown by the other species within the genus Baissoptera. Only Baissoptera grandis has a longer pterostigma, its length about 11 times its basal pterostigmal width (Ren et al. 1995: p. 175, fig. 2). The shortest pterostigmata within the genus are found in Baissoptera martinsoni and Baissoptera elongata, their lengths ca. four and six times their basal pterostigmal widths, respectively (Martynova 1961: p. 80, fig. 6; Ponomarenko 1993: p. 67, fig. 5). Naturally, our placement of this species in Baissoptera is a conservative position based on the lack of complete material. More completely-preserved specimens, in which the wing base characters noted above could be assessed, may force a reconsideration of the generic assignment.

Necroraphidia arcuata sp. n.

Small size; costal field very broad; pterostigma with a single, subdistal, strongly oblique, slightly sinuose to arcuate crossvein; pterostigma with a diffuse base; forewing with Rs and MA each forked twice; forewing with second radial cell proximally broad.

The new genus-group name is a combination of the Greek word nekros, meaning, “dead”, and Raphidia, common generic stem for snakeflies. The name is feminine.

Necroraphidia gen. n. is most similar to Ororaphidia and Styporaphidia from the Late Jurassic of Inner Mongolia, China (Engel and Ren 2008). These three genera show the very diagnostic character of a diffuse base to the pterostigma, lacking a crossvein as the proximal boundary of this wing region (figs 4D–E). Also, they share the presence of, at least, one pterostigmal crossvein subdistally, three discoidal cells posterior to MP in the forewing (forming a triangle), and a larger size if compared with other minute mesoraphidiids. Caloraphidia glossophylla Ren, 1997 (combination restored by Bechly and Wolf-Schwenninger 2011) shares these three characters, but it lacks the diffuse pterostigmal base according to Ren (1997: p. 184, fig. 11). Hence, to consider this species as closely related (i.e., as an ororaphidiine in their system) as suggested by Bechly and Wolf-Schwenninger (2011) seems dubious. Necroraphidia can be separated from Styporaphidia by the presence of a single pterostigmal crossvein (two in Styporaphidia) and the three branches of both Rs and MA (two branches in Styporaphidia), while it can be distinguished most readily from Ororaphidia by the strongly oblique and arcuate shape of the pterostigmal crossvein (less oblique and straight in Ororaphidia), the proximally broader second radial cell (narrowly triangular proximad in Ororaphidia), the smaller second discoidal cell, and also by the smaller size (forewing length 11.4 in Ororaphidia) (Engel and Ren 2008).

CES 391.1, from El Soplao amber; incomplete specimen, almost complete left fore- and hind wings (lacking their basalmost part), distal half of right hind wing and apex of right forewing, partial abdomen, and two leg fragments. Dense fungal hyphae infestate the abdomen and wings. The specimen is preserved together with the following syninclusions: two coleopterans, two hymenopterans (one of them, CES 391.2, belonging to the Megalyridae; Pérez-de la Fuente et al. 2012), one immature aphid, a cluster of trichomes, a fewcharcoalified plant fibers (Fig. 7A), a few timber debris, as well as other indeterminate organic remains.

As for the genus (vide supra).

Sex unknown.Legs patterned as follows (at least in the preserved fragments): femur with three dark areas, tibia with proximal area darkened and a dark area beyond midlength.Wing veins brown; veins with strong, very short setae, especially abundant on C; membrane hyaline. Forewing.Length of preserved fragment 6.9 (estimated total wing length > 9), maximum width 2.7; costal field very broad (costal field about two times wider than pterostigmal base at widest preserved point; Sc terminating into C around two-thirds of estimated wing length; three c-sc crossveins preserved; single, proximal sc-r crossvein; pterostigma elongate (3.2 long, longer than either radial cell), widening distally (maximum width almost twice basal width), and faintly infumate, starting at termination of Sc; pterostigma with a single, subdistal, strongly oblique, slightly sinuose crossvein; pterostigma with a diffuse base (i.e., lacking a crossvein as proximal boundary of this wing region); Rs with three branches, distalmost fork very short; two large radial cells present; first radial cell about 1.3 times longer than second radial cell; second radial cell proximally broad; MA arising slightly distad midpoint of first radial cell, with three branches; three discoidal cells posterior to MP; 1cua-cup crossvein not preserved; anal veins not preserved; jugal lobe not visible. Hind wing.Length of preserved fragment 6.6 (estimated total wing length 8–9), maximum width 2.7;costal field distinctly narrower than in forewing;four c-sc crossveins preserved;pterostigma elongate (2.9 long, longer than either radial cell), widening distally (maximum width almost twice basal width), and faintly infumate, starting at termination of Sc; pterostigma with a single, subdistal, strongly oblique, arcuate crossvein; pterostigma with a diffuse base; Rs with two branches; two radial cells present; MA with three branches; two discoidal cells posterior to MP, the first one trianguloid, not much smaller than second one; 1ma-mp crossvein not especially close to fork between Rs and MA; anal area not preserved. Abdomen. Length 3.8. Genitalia degraded, with dorsal part missing, and badly seen due to presence of dense fungal hyphae.

The specific epithet is the Latin term arcuatus, meaning “bent”, and refers to the arcuate form of the pterostigmal crossvein.

MCNA 9343, from Peñacerrada I amber; hind wing fragment and abdominal apex, including the genitalia.

The new species is similar to Styporaphidia magia Engel and Ren, 2008 from the Late Jurassic of Inner Mongolia, China, in the presence of two pterostigmal crossveins. Styporaphidia? hispanica sp. n.differs in that the distance between 1ptc and 2ptc is three times the distance between 2ptc and the end of the pterostigma (two times in Styporaphidia magia), the forking of Rs at the apicalmost r-rs crossvein (rather than prior to it in Styporaphidia magia), and R meeting the apicalmost r-rs crossvein beyond the pterostigma (within in Styporaphidia magia).

Male. Hind wing. Length as preserved 3.5, maximum width as preserved 2.5; wing apex relatively rounded; C especially thick when compared to other veins; pterostigma almost with constant width along its entire length, infumate; pterostigma with two crossveins, distalmost crossvein oblique and slightly arcuate, proximal crossvein apparently straight, distance between 1ptc and 2ptc three times distance between 2ptc and end of pterostigma; Rs with two branches, forking at r-rs crossvein; R meeting apicalmost r-rs crossvein beyond pterostigma; rs-ma crossvein meeting MA after its distalmost fork; MA with two branches. Abdomen. Gonocoxites 9 with a few long setae;gonostyli 9 segment rather short, rounded (not acute), slightly upcurved;tergite 10 (+11?) with distal setae;paired, contiguous, acute genital structures located dorsad to gonostyli 9, interpreted as distalmost part of parameres (Fig. 6A).

The specific epithet refers to the occurrence of this species in ancient Spain (Hispania in Latin).

Although the base of the pterostigma is not preserved and it is accordingly impossible to ascertain if it was diffuse (i.e., lacking a crossvein), this species is tentatively placed in the genus Styporaphidia owing to the presence of two pterostigmal crossveins. The presence of two pterostigmal crossveins is a rare feature among the Raphidioptera and otherwise known in a few other raphidiopterans, i.e, namely the baissopterids Baissoptera kolosnitsynae and Baissoptera pulchra (Martynova 1961: p. 81, fig. 7; Martins-Neto and Nel 1992: p. 428, figs 2, 3), but the present fossil is clearly not a baissopterid. The fragment is considered as corresponding to a hind wing due to the more distal disposition of both crossveins as occurs in Styporaphidia magia (the distance between them being greater than the distance between 2ptc and the end of the pterostigma), and by the relative position of the abdomen. The veins C and R appear to be especially thickened close to the wing apex as occurs also in Styporaphidia magia (Engel and Ren 2008: fig. 2), though this effect could be enhanced by the darkening of the margins of the pterostigma. Although the species is represented by a wing fragment and the genitalia, it is distinctive enough that it should be easy to associate with more complete material that may become available in the future.

Amarantoraphidia ventolina sp. n.

Minute size. Head ovoid, with the portion posterior to the compound eyes longer than the eye diameter and tapering caudad; three large ocelli present, situated between anterior half of compound eyes; antennae with a low number of flagellomeres (i.e., ≤ 26). Pronotum slightly longer than head, with a constant height along its entire length. Mesotibiae especially swollen; process at midlength of the metatibiae absent. Forewing with costal field moderately broad; pterostigma elongate, without crossveins; Sc terminating into C slightly distad wing midlength; six c-sc crossveins present; two discoidal cells posterior to MP; apicalmost branch of CuA simple; 1cua-cup crossvein located at the M-CuA separation.

The new genus-group name is a combination of the Greek term amarantos, meaning “that never fades, ageless”, and Raphidia, common generic stem for snakeflies. The name is feminine.

Amarantoraphidia gen. n.is first compared with the other described minute mesoraphidiid genera, which have a forewing length around 6 or less. They include amber inclusions but also some compression fossils. Apart from the minute size, in all of these taxa the pterostigma is very elongate, without crossveins, and basally closed by a crossvein. Regarding those as amber inclusions, Amarantoraphidia is readily distinct from the other two currently described Spanish mesoraphidiid genera, Cantabroraphidia and Alavaraphidia gen. n., as well as Lebanoraphidia, by its ovoid head shape (subquadrate in Cantabroraphidia, and rhomboidal in the other two genera) and by its lesser number of antennal flagellomeres (26 flagellomeres in Cantabroraphidia, 38 flagellomeres or more in Lebanoraphidia, and 44 in Alavaraphidia). The two genera Nanoraphidia (type species Nanoraphidia electroburmica Engel, 2002, Burmese amber, latest Albian in age) and Grimaldiraphidia (type species Grimaldiraphidia luzzi (Grimaldi, 2000), New Jersey amber, Turonian in age)share with Amarantoraphidia the ovoid head shape. However, in both of these genera the ocelli are placed between the posterior half of the compound eyes, not between the anterior part as in Amarantoraphidia. Additionally, whereas Amarantoraphidia has two discoidal cells posterior to MP in the forewing, Grimaldiraphidia has three and Nanoraphidia just a single cell. Also, in Nanoraphidia the M-CuA separation is located near midpoint of the first and second cua-cup crossveins (Engel 2002; Jepson et al. 2011), not at the 1cua-cup crossvein as in Amarantoraphidia.

In addition to the aforementioned taxa, five mesoraphidiids with minute size have been hitherto described from compression fossils: Grimaldiraphidia parvula (Martynov, 1925), from Karatau (South Kazakhstan), Late Jurassic in age; Grimaldiraphidia mitchelli (Jepson, Coram and Jarzembowski, 2009), Grimaldiraphidia purbeckensis Jepson, Coram and Jarzembowski, 2009, and Mesoraphidia websteri Jepson, Coram and Jarzembowski, 2009, the three species from the Purbeck Limestone Group, Dorset (UK), Berriasian in age; and Nanoraphidia lithographica Jepson, Ansorge and Jarzembowski, 2011 (tentatively assigned to this genus), from El Montsec (Spain), Early Barremian in age. Grimaldiraphidia parvula (described from a complete specimen but its wings unresolved dorsoproximally) has more proximal positions of the fork of Rs and the 2r-rs and rs-ma crossveins than Amarantoraphidia (Martynov 1925: p. 242, figs 7–9). Grimaldiraphidia mitchelli (described from a wing, most likely a forewing, with not preserved base and pterostigma), although possessing two discoidal cells posterior to MP as in the forewing of Amarantoraphidia, has a venation somewhat different, with the second radial cell relatively wide compared to its length, and more proximal positions of the fork of Rs and the 2r-rs and rs-ma crossveins (Jepson et al. 2009). Grimaldiraphidia purbeckensis possesses three discoidal cells posterior to MP in the forewing, the M-CuA separation is located near midpoint of the first and second cua-cup crossveins, and has more proximal positions of the fork of Rs and the 2r-rs and rs-ma crossveins (ibid.). Mesoraphidia websteri (based on a hind wing) has the Sc ending beyond the first radial cell, relatively shorter second radial and discal cells, and a single discoidal cell (two discoidal cells posterior to MP in Amarantoraphidia) (ibid.). Lastly, Nanoraphidia lithographica shows only one discoidal cell posterior to MP in the forewing as in the type species for the genus, Nanoraphidia electroburmica (Jepson et al. 2011).

CES 364.1, from El Soplao amber; almost complete female, just lacking the distalmost portion of both forewings beyond the end of the pterostigma and the distal third of the right hind wing. The first left leg is disarticulated. The specimen is preserved together with the following syninclusions: one evaniid (a new Cretevania species; CES 364.2, Pérez-de la Fuente et al. 2012, in prep.) and three other indeterminate hymenopterans, four dipterans (one chimeromyiid among them), one thysanopteran, a few charcoalified plant fibers, and a few timber debris, as well as other indeterminate organic remains.

As for the genus (vide supra).

Female.Integument dark brown; legs patterned as follows: femora darkened from just before their midlength to their end; three dark areas on tibiae, proximally, medially and distally; tarsomere 1 not darkened, distal tarsomeres darkened. Head.Ovoid, about 0.7–0.8 long, with portion posterior to compound eyes longer than eye diameter and tapering caudad; three large ocelli present, situated between anterior half of compound eyes; mandibles with teeth not visible;palps short;compound eyes large and exopthalmic, separated by distance slightly greater than compound eye length; antennae inserted around anterior tangent of compound eyes (exact insertion not visible); scape and pedicel gracile, both measuring about length of four flagellomeres and being subequal in thickness to them; 24 flagellomeres present, slightly longer than wide, with sparse, minute setae; coronal ecdysial cleavage line not evident; posterior border of head with a distinct collar-like lip.Thorax.Prothorax about 1.1 long, meso- plus metathorax 1.4 long; pronotum slightly longer than head, with a constant height along its entire length (i.e., without a distinct change of slope in lateral view); a few spines visible on prothorax, dorsoanterior mesothorax apparently with a few small spines; all tibiae with apical spines; mesotibiae especially swollen; metatibiae significantly more elongate and thinner than the other tibiae; process at midlength of metatibiae absent; five tarsomeres, third with bilobed extensions lacking digitiform processes (Fig. 7B); pretarsal claws simple, with a basal enlargement; arolium large.Wing veins brown, meeting wing margins without bifurcating; veins with strong, very short setae, especially abundant on C; membrane hyaline. Forewing.Length about 5.6 (tip not preserved), maximum width 1.9; costal field moderately broad (at widest point costal field about 1.4 wider than pterostigma); six c-sc crossveins present, two basalmost c-sc crossveins particularly close to each other; Sc terminating into C slightly distad wing midlength; single, proximal sc-r crossvein; pterostigma elongate (1.5 long), slightly longer than either radial cell; pterostigma with constant width along its entire length, faintly infumate, starting 0.4 (about three times pterostigmal width) beyond termination of Sc; pterostigma without crossveins, basally closed by a crossvein; Rs with two branches; two large radial cells present; first radial cell nearly as long as second radial cell, with MA arising slightly distad its midpoint; MA with two branches; two discoidal cells posterior to MP; apicalmost branch of CuA unforked; 1cua-cup crossvein located at M-CuA separation; 2A arcuate; jugal lobe not visible. Hind wing. Length about 4.1, maximum width 1.3; costal field distinctly narrower than in forewing; four c-sc crossveins present; sc-r crossvein absent; pterostigma elongate (1.5 long), slightly longer than second radial cell; pterostigma with about a constant width along its entire length, faintly infumate, starting 0.3 (slightly more than two times pterostigmal width) beyond termination of Sc; pterostigma without crossveins, basally closed by a crossvein; Rs with two branches; two radials cells present; MA with two branches; two discoidal cells posterior to MP, first one smaller and trianguloid; 1ma-mp crossvein close to fork between Rs and MA; anal area folded. Abdomen. Length 2.4; ovipositor robust, 1.7 long as preserved, 0.1 thick (about 15 times as long as wide); ovipositor showing dense annulations (Fig. 7C); ovipositor with faint, stiff, short sensory setae along its entire length; ovipositor gonostyli most likely club-shaped.

In the Cantabrian mythology, the “ventolines” are tenacious and always cheerful fairy-like air beings that dwell in the depths of the sky and, when summoned, help defenseless fishermen by placidly steering their boats to the shore while embracing them with their warm green wings. The term has been singularized and feminized for combination.

In extant snakeflies, the dense annulations of the ovipositor (cf. Fig. 7C) provide the flexibility necessary for introduction into irregular cavities, similar to a flexible metallic hose (Mickoleit 1973). It has been noted how mesoraphidiids would have had a shorter and thicker ovipositor than Recent Raphidioptera (Bechly and Wolf-Schwenninger 2011). The shape of the ovipositor in this specimen and also in Alavaraphidia gen. n. supports such a conclusion.

Alavaraphidia imperterrita sp. n.

Minute size. Head rhomboidal, with clypeus especially elongate and the portion posterior to the compound eyes shorter than the eye diameter and strongly tapering caudad; three large ocelli present, situated near the posterior tangent of compound eyes; antennae extremely elongate, with a high number of flagellomeres (i.e., ≥ 38). Pronotum shorter than head, with a constant height along its entire length. All tibiae especially swollen medio-apically; process at midlength of the metatibia indistinct. Bilobed extensions of third tarsomeres with distal digitiform processes.

The new genus-group name is a combination of Álava, from Álava amber (the name of the group for Peñacerrada I and Peñacerrada II amber localities), and Raphidia, common generic stem for snakeflies. The name is feminine.

Although most of the wings are not preserved, the other features of Alavaraphidia gen. n. are distinctive enough to justify the creation of a distinct taxon. The minute size of Alavaraphidia mainly limits its comparison to other minute taxa mainly described from amber inclusions (refer to comments for Amarantoraphidia gen. n.). Although a few other taxa based on wings from compressions are minute in size it is not possible to compare them with the new genus and species due to the absence of most of its wings. The genus Lebanoraphidia shares with Alavaraphidia the rhomboidal shape of the head with the compound eye length greater than that of the head posterior to the eyes (Bechly and Wolf-Schwenninger 2011); the head shape is ovoid in the genera Grimaldiraphidia, Nanoraphidia, and Amarantoraphidia, and subquadrangular in the genus Cantabroraphidia. Moreover, the genus Lebanoraphidia differs from Alavaraphidia in the lesser number of flagellomeres (note how the number of flagellomeres depicted for Lebanoraphidia nana Bechly and Wolf-Schwenninger, 2011, type and only species of the genus, seems to not match the description and photographs), the shorter clypeus, the shorter portion posterior to the compound eyes, the longer relative length of the pronotum, and the assumed absence of distal digitiform processes on the third tarsomere’s bilobed extensions. The high number of flagellomeres of Alavaraphidia (44) is most similar to that of Lebanoraphidia, itself with around 38 flagellomeres (Bechly and Wolf-Schwenninger 2011). Otherwise, 20 flagellomeres are present in the genus Nanoraphidia (Engel 2002), 23 in Grimaldiraphidia (Grimaldi 2000), 24 in Amarantoraphidia, and 26 in Cantabroraphidia (Pérez-de la Fuente et al. 2010). The distal digitiform processes from the third tarsomere’s bilobed extensions (fig. 10C) have not been reported from any other taxon and thus it is considered a unique character. Regarding the leg patterning, the medial dark patch found on the tibia in Cantabroraphidia (ibid.) and Amarantoraphidia is absent in Alavaraphidia. However, this character is not considered relevant at the generic level and hence has been discarded from the present diagnosis.

MCNA 13608, from Peñacerrada I amber; partial specimen showing head and ventral parts of thorax and abdomen, including the ovipositor. Only the basalmost part of the wings is preserved.

As for the genus (vide supra).

Female.Body length excluding ovipositor 5.7. Integument dark brown; legs patterned as follows: femora darkened from just before their midlength to their end; two dark regions on tibiae, proximally and distally (medial dark region absent); tarsomere 1 not darkened, distal tarsomeres darkened. Head. Rhomboidal, about 1.2 long, with portion posterior to compound eyes shorter than eye diameter (about 0.7 times) and strongly tapering caudad; three large ocelli present, situated near posterior tangent of compound eyes; mandible with teeth not visible; palps short; clypeus especially elongate; compound eyes separated by about compound eye length; three large ocelli present, situated near posterior tangent of compound eyes; antennae inserted posterior to clypealfrons sulcus, basad anterior tangent of compound eyes; antennae extremely elongate, with 44 flagellomeres; flagellomeres elongate, about 1.5 times longer than wide; scape and pedicel thicker than flagellomeres, scape measuring about two flagellomeres, pedicel measuring slightly more than a flagellomere; coronal ecdysial cleavage line not evident; posterior border of head not visible. Thorax. Prothorax about 0.8 long; meso- and metathorax about 1.1 long; pronotum shorter than head, with a constant height along its length (i.e., without a distinct change of slope in lateral view); thoracic dorsal spines not visible, if present; all tibiae especially swollen medio-apically, with apical spines, spines also visible on metatarsomeres; process at midlength of metatibia indistinct; five tarsomeres, third with bilobed extensions having six to eight distal digitiform processes (Fig. 10C), different in shape than regular leg setae (not tapering apically); pretarsal claws simple, with a basal enlargement; arolium large.Preserved wing veins brown, with strong, very short setae; preserved membrane hyaline.Forewing. Costal field not especially broad. Three c-sc crossveins preserved.Hind wing. Costal field distinctly narrower than in forewing. Three c-sc crossveins preserved.Abdomen.Length 2.7; ovipositor robust but rather elongate, about 2.7 long as preserved, 0.2 thick (about 15 times as long as wide);ovipositor with dense annulations; ovipositor with conspicuous faint, stiff, small sensory setae along its entire length; ovipositor gonostyli club-shaped; tergite 10 (+11?) with a distalmost stripe of stiff trichobothria (some probably from tergite 9).

The specific epithet is the Latin term imperterritus, meaning “fearless”, and symbolizes the unalterable condition of an organism entrapped in amber.

According to the original diagnosis, this monospecific genus was characterized by the following combination of characters: minute size (forewing length 5.5). Head more or less quadrangular; with portion posterior to the compound eyes slightly shorter than the eye diameter and not tapering caudad; three large ocelli present, situated between anterior half of the compound eyes; posterior border of head with a distinct collar-like lip. Pronotum subequal in length to head length, with anterior half narrowed dorsoventrally relative to posterior half (i.e., with slight downward curve in lateral view). Process present at midlength of the metatibiae in a posterior position. Forewing with costal field relatively broad (at widest point costal field as broad as pterostigma); pterostigma elongate, without crossveins; Sc terminating into C in distal two-thirds of wing length; four c-sc crossveins; two discoidal cells posterior to MP; 1cua-cup crossvein strongly basad M-CuA separation; posterior branch of MP forked. Refer to Pérez-de la Fuente et al. (2010) for a complete description of the type and only known species from El Soplao amber, based on an adult of unknown sex.

MCNA 9218, from Peñacerrada I amber; wing apex plus two minute wing fragments lacking formal descriptive significance. The sample consists of part and counterpart after the amber piece broke following the plane of the inclusion.

Sex unknown. Hind wing(?). Length of preserved fragment 2.4, maximum width of preserved fragment 2.4;wing apex rounded; pterostigma slightly increasing in width distally, infumate; pterostigma with an almost straight subdistal crossvein; all apical branches relatively short; Rs with three branches, distalmost fork very short; apicalmost r-rs crossvein (2r-rs?) meeting R distal to the pterostigma; rs-ma crossvein situated at distalmost fork of MA; MA with two branches.

The present material is distinct from other Spanish amber snakeflies but does not preserve enough detail to permit formal designation as a taxonomic entity. The presence of a pterostigmal crossvein immediately discounts Cantabroraphidia and Amarantoraphidia gen. n. MCNA 9218 highly resembles the hind wing of Styporaphidia in the shape and location of the distalmost pterostigmal crossvein. In fact, its venation is very similar to Styporaphidia? hispanica sp. n., though in the present material Rs is forked very close to the wing margin and the shape of its distalmost radial cell (most likely the second one) is somewhat different. Furthermore, the less pointed wing apex, shorter apical branches, and apical shapes of the radial, discal, and presumed second medial cells immediately distinguish MCNA 9218 from Necroraphidia gen. n. and genus and species indet. 2 (vide infra), and the almost straight pterostigmal crossvein and positions of the apicalmost r-rs (2r-rs?), rs-ma, and apicalmost ma-mp (2ma-mp?) crossveins further differentiate MCNA 9218 from Necroraphidia.

CES 376, from El Soplao amber; forewing apex and the area surrounding pterostigma from an additional wing. Some additional dorsoproximal parts of the wing are also present but with a very poor preservation, so just a few more characters can be elucidated. An indeterminate hymenopteran is present as a syninclusion.

Length as preserved ca. 5.0 (wing well preserved only in 3.4 of that length), maximum width as preserved 2.7; wing apex pointed, positioned within Rs series; wing veins brown, meeting wing margins without bifurcating; veins with strong, short setae, especially abundant on C; Sc ending and proximal r-rs crossvein (1r-rs?; very faintly preserved) situated at about same wing length, pterostigma slightly widening distally, infumate; pterostigma with a very faint subdistal, rather straight (not conspicuously arcuate), strongly oblique crossvein; uncertain if pterostigmal division present; apical branches of Rs, MA and MP subparallel; two apical branches of R distal to pterostigma; Rs with four branches, posteriormost originating before distalmost r-rs crossvein (2r-rs?), separated from it by much more than its length; rs-ma and distalmost ma-mp crossveins lacking in preserved wing fragment, so most likely with a proximal position; MA most likely with three branches.

Despite the fact that CES 376 is distinct from the other taxa in Spanish amber, it is not named as its preserved parts are not enough to resolve its affinities. The wing fragments show a high resemblance with some mesoraphidiids such as Mesoraphidia obliquivenatica (Ren, 1994) and Caloraphidia glossophylla, both from the Cretaceous compression deposit of Liaoning (China), as long as all of them share the presence of a strongly oblique, rather straight pterostigmal crossvein in a rather distal position and a distal portion of the wing with long apical branches and without crossveins other than the 2r-rs crossvein (Ren 1994: p. 134, fig. 4; 1997: p. 184, fig. 11). Caloraphidia glossophylla possesses subparallel apical branches of Rs, MA and MP, the Sc ending and the 1r-rs crossvein situated at about the same wing length, the 2r-rs crossvein closer to the end of the pterostigma than to the pterostigmal crossvein, and a pointed apex positioned within the Rs series, but in this species Rs has only three branches. By contrast, although Mesoraphidia obliquivenatica has Rs with four branches as in CES 376, the apical branches of Rs, MA and MP are not subparallel, the Sc ends in a more basal position than the 1r-rs crossvein, the 2r-rs crossvein is closer to the pterostigmal crossvein than to the end of the pterostigma, and the apex is more rounded and positioned between R and Rs. CES 376 is also quite similar to the wing apex of the compression fossil Iberoraphidia dividua Jepson, Ansorge and Jarzembowski, 2011, from El Montsec (Spain), Early Barremian in age, with its relatively simple venation, a Sc ending and a 1r-rs crossvein situated at about the same wing length, the 2r-rs, rs-ma and 2ma-mp crossveins not apically placed, the four branches of Rs, with the posteriormost branch of Rs originating before the 2r-rs crossvein, and the relatively simple apical fork of MA and MP (Jepson et al. 2011). CES 376 differs, however, in the presence of a pterostigmal crossvein (although in Iberoraphidia dividua the distal portion of the divided pterostigma could be basally closed by a crossvein), only two apical branches to R distal to the pterostigma (not three as in Iberoraphidia dividua), the posteriormost branch of Rs more proximally placed (separated from the distalmost r-rs by much more than its length, versus much shorter in Iberoraphidia dividua), and the more pointed wing apex which is positioned within the Rs series (rather than between R and Rs in Iberoraphidia dividua).

MCNA 9316, from Peñacerrada I amber; fore- and hind wing fragments from the same side of the body and a poorly preserved part of the abdomen, although the genitalia is somewhat visible. The amber piece contains abundant organic remains. The specimen is preserved together with legs of a spider as syninclusions.

Male.Small size (inferred from preserved wing fragments). Forewing. Length as preserved about 4.5, maximum width not measurable. Two discoidal cells posterior to MP; posterior branch of MP unforked; M-CuA separation not preserved; 2A arcuate. Hind wing.Length as preservedabout 4.7, maximum width not measurable. Costal field relatively narrow; four c-sc crossveins preserved; Sc ending into C at length of first radial cell’s midlength; pterostigma not evidently infumate, not closed basally by a crossvein, at least proximally; first discoidal cell not especially small (or compact) (small in Cantabroraphidia marcanoi); rs+ma-mp crossvein present. Abdomen. Length as preserved (including genitalia) 3.8. Gonocoxites 9 with paired basal inner tubercles with small dark teeth; gonocoxites 9 elongate, with very long, stiff setae distally; gonostyli 9 very elongate, slightly upcurved; parameres not conspicuous; tergite 10 (+11?) with distalmost two or three stripes of trichobothria.

Comments. The preserved characters of MCNA 9316 are not enough to create a new taxon. The combined presence of two discoidal cells posterior to MP as in Cantabroraphidia and Amarantoraphidia gen. n., and the lack of a crossvein closing the pterostigma basally in the preserved wing length as in Necroraphidia gen. n., Ororaphidia, and Styporaphidia (showing three discoidal cells), precludes assignment to any of these genera. The lack of a crossvein closing the pterostigma basally can mean that the base of the pterostigma may have been diffuse or that the pterostigma may have been placed in a more distal position, as occurs in some mesoraphidiids. However, the inferred size of MCNA 9316 would have not been as reduced as in the minute mesoraphidiids but would have better fit with that of those mesoraphidiids with a diffuse pterostigmal base (refer to comments for Necroraphidia gen. n.). Furthermore, the presence of a rs+ma-mp crossvein in the hind wing as occurs in MCNA 9316 is a very rare character among described Mesoraphidiidae but more common in modern snakeflies, although it is also present in Styporaphidia magia. The morphology of the genitalia appear distinctive enough as to be recognizable in future findings. It is interesting to note that Aspöck and Aspöck (2004: figs 9, 10) used the presence of wartlike tubercles (covered with dark teeth) on the inner side of gonocoxite 9, larger but similar to those shown by MCNA 9316, as a diagnostic character for the inocelliid genus Succinofibla Aspöck & Aspöck, 2004 from Baltic amber. These authors also suggested that these tubercles could be related with the closing mechanism of the genitalia.