Dilophus idanos Fitzgerald and Greenwalt, 2022

Greenwalt, Dale E., Amorim, Dalton De Souza, Hauser, Martin, Kerr, Peter H., Fitzgerald, Scott J., Winterton, Shaun L., Cumming, Jeffrey M., Evenhuis, Neal L. & Sinclair, Bradley J., 2022, Diptera of the Middle Eocene Kishenehn Formation II, Palaeontologia Electronica (a 22) 25 (2), pp. 1-52 : 6-12

Dilophus idanos Fitzgerald and Greenwalt sp. n.

Figures 2-4

zoobank.org/ C3C58815-8BCE-43EB-A3D5-0547CC742A43

Type species. Tipula febrilis Linnaeus, 1758 , by subsequent designation of Latreille, 1810.

Holotype. Male, USNM 624840 ; deposited in the Paleobiology collections at the National Museum of Natural History , in Washington, D.C.

PALAEO- ELECTRONICA.ORG

Locality and horizon. Dakin site, Middle Fork of the Flathead River (Pinnacle, Montana, USA). Middle Eocene Coal Creek Member, Kishenehn Formation.

Etymology. The specific epithet is the Greek term idanos , meaning fair, comely.

Diagnosis. Dilophus idanos is distinguished from fossil congeners by a combination of its small size (wing 2.4 mm), fore tibial spines 2:2:6(8?) and patterns and numbers of the mesonotal spines (see comparison to previously described fossil Dilophus in Table 1).

Description (male). Body length (excluding antennae) 3.7 mm; black except thorax, tibiae and femora. Head black, 0.52 mm long; antenna short (0.2 mm), base hidden; approximately six flagellomeres present with terminal flagellomeres bulbous with several apical setae. Terminal palpal segment 83 μm long with several apical setae ( Figure 2A ). Thorax light brown, 0.93 mm long, 0.60 mm wide; anterior margin of mesonotum with broad semicircular arrangement of 12 spines; just posterior to anterior semicircle of spines,mesonotum with narrower semicircular pattern of 10 spines with additional two spines below apex of semicircle forming diamond-shaped pattern ( Figure 3A ). Wings 2.4 mm long, 0.91 mm wide, hyaline with microtrichia on surface; pterostigma strongly pigmented, dark brown. Costa, with marginal anterior setae, continued beyond R 4+5 to slightly less than halfway to wing apex; Sc long, reaching C just before pterostigma; r-m long, 3.2 times as long as bRs (left wing); anterior veins except Sc (C, radial veins, base of M prior to junction with r-m, r-m) thick, strongly pigmented dark brown. Sc and apical tips of M 1, M 2, M 4 and CuA faint, light brown; CuP not visible ( Figure 2 B-C). Hind legs not preserved. Tarsi black, tibiae and femora light brown. Fore tibia with two pairs of spines below mid-point, one pair above other; apex of fore tibia with circlet of six visible spines with probable total of eight (2:2:6[8?]) ( Figure 3 B-C). Abdomen black, narrow (2.3 mm x 0.45 mm) as typical for males of family. Terminal segment bulbous, details of genitalia not preserved.

Description (female). Unknown.

Synimpressions. None

Paratype. Male, USNM 768228 ; deposited in the Paleobiology collections at the National Museum of Natural History , in Washington , D.C.

Figure 4

Locality and horizon. Dakin site, Middle Fork of the Flathead River (Pinnacle, Montana, USA). Middle Eocene Coal Creek Member, Kishenehn Formation.

Synimpressions. None

Remarks. The genus Dilophus contains about 232 described extant species and is most diverse in the tropics of the Southern Hemisphere; approximately 18 and 36 species are found in the Nearctic and Palaearctic regions, respectively ( Hardy, 1965; Haenni and Bosák, 2007; GBIF, 2021; Skartveit, 2017). Including D. idanos , there are eleven valid fossil species assigned to the genus, four of which are in amber: D. crassicornis Skartveit, 2008 , D. palaeofebrilis Skartveit, 2008 , and D. succineus Skartveit, 2008 from Baltic amber and D. matilei Waller, Nel and Menier, 2000 from Dominican amber. Species described from compression fossils include: D. krantzii Heyden, 1870 and D. luteipennis Théobald, 1937 (transferred from Bibio Geoffroy, 1762 by Skartveit and Nel, 2017) from the Oligocene of Germany and France respectively ( Heyden, 1870; Skartveit and Nel, 2017), D. pumilio Skartveit and Pika, 2014 from the Miocene of Germany, D. pinguis ( Heer, 1849) from the Miocene of Croatia (transferred from Bibio by Skartveit and Krizmanic, 2020), D. andrewrossi Nel, Collomb and Waller, 2019 from Isle of Wight deposits (Krzemiński et al., 2019) and D. magnus Dürrenfeldt,1968 from Pliocene Germany ( Dürrenfeldt, 1968; Skartveit and Pika, 2014; but see treatment of this species below). Dilophus campbelli , a compression fossil of a larval specimen, was reported from Eocene New Zealand by Harris (1983). Dilophus priscus Loew, 1850 , a Baltic amber specimen which may be lost, was considered a nomen nudum ( Skartveit, 2008). However, Loew (1850) stated that “the throax has thick horns,” which is sufficient to make the name available. Dilophus deletus Heyden, 1859 was treated as nomen dubium by Skartveit and Wedmann (2021). Numerous specimens not preserved sufficiently for assignment to species have been reported from Baltic amber ( Skartveit, 2008), the Oligocene of France ( Skartveit and Nel, 2017), the Miocene of Germany ( Skartveit and Pika, 2014) and Miocene of Iceland (Skartveit et al., 2017).

Dilophus idanos is an exceptionally well-preserved specimen. It is the oldest adult representative of the genus, the first fossil of the genus to be reported from the Nearctic, and distinct from all known fossils of the genus (see comparison to previously described fossil Dilophus in Table 1, which roughly summarizes much of this Remarks section). With the exceptions of D. crassicornis , D. succineus and D. matilei , it is smaller than most fossils of the genus. However, size and color are not reliable characters for distinguishing different fossil species. For example, Collomb et al. (2008) reported that body lengths of the extant species Bibio hortulanus Linnaeus, 1758 ranged from 4.5 to 11 mm in males and from 6 to 11 mm in females. That said, Dilophus idanos is of typical size to most extant species of Dilophus and is unlikely to be mistaken with the unusually large fossil Dilophus species D . krantzii (male) and D.pinguis (male and female), which have body lengths greater than 1 cm. Dilophus idanos differs from D. luteipennis (female) in the ratio of the lengths of r-m relative to that of the basal portion of Rs (bRs); the ratio for D. idanos is 3.2 (left wing, Figure 2 B-C). Skartveit and Nel (2017) indicated that this ratio for D. luteipennis was slightly greater than one, but the specimen was figured with r-m slightly shorter than bRs and data in their table 5 showed bRs significantly longer than r-m. Although their values (their table 5) for the r-m and Rs veins of D. luteipennis varied by approximately 50%, the ratio of the average of the measurements of r-m/bRs is about 0.6, which is the value reported here in Table 1. However, it should be noted that Skartveit and Nel (2017) stated that “As usual with fossil insects ... it is not possible to calculate ratios between different measurements from the data as here presented.” Heyden (1870) figured the wing of D. krantzii with crossvein r-m approximately half the length of bRs (plate 45, figure 24). Skartveit and Pika (2014) state that “crossvein r-m (in D. pumilio ) is markedly longer than the basal part of Rs, which is a diagnostic character for Dilophus ” (the r-m/bRs ratio for D. pumilio = 2.1). However, Collomb et al. (2008) documented extreme intraspecific variation in the length of the r-m cross vein in the extant Bibio hortulanus , including an r-m/bRs ratio that would be a “diagnostic character for Dilophus ” (their figure 5c); the value of the ratio of the lengths of r-m to bRs in Dilophus would be of greater value if intraspecific variation data were available. Dilophus idanos differs from the male of D. palaeofebrilis and both the male and female of D. succenius in that the antennae of the latter two species has twice as many (12) flagellomeres and are 2.5 times as long. The costa of D. luteipennis , which did not extend beyond R 4+5, distinguishes this species from D. idanos ( Figure 2 B-C).

However, one of the more informative characters for distinguishing between different species of Dilophus is the pattern and number of fore tibial and mesonotal spines. The fore tibial spines consist of those that form a circlet at the apex of the tibia as well as one or more short rows or clusters of spines mesally. Dilophus idanos has two mesal rows of spines, one above the other, of two spines each, and six visible spines at the apex of the fore tibia (a few additional apical spines may be obscured so the number could be about 8) ( Figure 3 B-C); a shorthand description of this pattern can be given as 2:2:6(8?) with the far right-hand value always indicating the number of spines in the apical circlet. The fore tibial patterns of D. palaeofebrilis (male), D. succineus (male, female), D. matilei (male) and D. crassicornis (male, female) are 1:2:8, 2?:3:6, 2:3:6, 2:4:8, 3:6 and 3:6, respectively. Dilophus luteipennis was described as “possibly with 3 mesal spines” ( Skartveit and Nel, 2017) and D. pinguis as “possibly with two small spines mesally” ( Skartveit and Krizmanić, 2020). Heyden (1870) stated “die Vorderschienen mit einem Stachelkranz endingen” (apical end of fore tibia with a ring of spines) in his description of D. krantzii , but provided no information about mesal spines.

The description of D. andrewrossi makes no mention of mesal spines (absence of mesal spines is unknown in the genus) and states that only two apical spines are present in the fore tibia instead of the typical crown of apical spines (Krzeminski et al., 2019).

Several of the fossil species also differ in the number, size and pattern of pro- and mesonotal spines. Dilophus idanos has a broad semicircular arrangement of 12 spines at the anterior margin of the mesonotum and, more posteriorly, a narrower semicircular pattern of 10 spines with an additional two spines below the apex of the semicircle forming a diamond-shaped pattern ( Figure 3A ). The D. succineus specimens are described as having transverse rows of spines ( Skartveit, 2008); the posterior row of mesonotal spines in D. palaeofebrilis , as figured, is also transverse ( Skartveit, 2008). The female of D. crassicornis has two rows of eight mesonotal spines with “one spine on each side laterally between the two spine rows” ( Skartveit, 2008), while the male has approximately ten spines in a semicircular pattern somewhat similar to D. idanos . The two rows of mesonotal spines in D. matilei (male) are transverse (anterior) or nearly so (posterior) as originally figured ( Waller et al., 2000). The mesonotum of a Dilophus sp. from the late Miocene of Iceland (Skartveit et al., 2017) is described as having “traces of two transverse spine rows.” In an examination of the female holotype and a female paratype of D. luteipennis, Skartveit and Nel (2017) stated that mesonotal spines were not visible. Dilophus pumilio , a female from the Miocene of Öhningen, Germany, was described with “Probable traces of protibial spines preserved, their exact pattern not possible to make out but may have consisted of one basal (two spines?) and one mesal (two or three spines?) group.” and a “vague trace of pronotal spine row” ( Skartveit and Pika, 2014). The M 4 vein of D. pumilio was figured nearly touching M 1+2 just apical of r-m but this may have been an artifact of the preservation process. The venation was described as “veins brownish, no difference in pigmentation between anterior and posterior veins.” This latter character would distinguish this specimen from D. idanos .

The paratype of Dilophus idanos appears to differ from the holotype in several respects. The terminal palpal segment of the latter is much longer than wide while that of the paratype is ovoid in shape. There are also subtle differences in both the fore tibial and mesonotal spines. In the holotype of D. idanos , the rows of the latter are more strongly arched although the number of spines appears to be the same in both specimens. In addition, while the holotype appears to have four spines above the apex of the fore tibia, the paratype may have five ( Figures 3 B-C, 4C). We do not consider these differences sufficient to suggest that specimen USNM 768228 is a different species. Fossilization-related changes to minute structures often give rise to spurious shapes. For example, the more distal row of fore tibial spines of the left leg of D. idanos have distinctly crenate termini ( Figure 3 B-C), while those of the right leg are sharply pointed.

Perhaps more importantly, species descriptions of extant Dilophus often report the number of spines as a range. For example, “...the posterior has twelve to fourteen minute teeth” ( Hardy, 1953) or “Pronotum with a transverse row of 8-11 spines, anterior margin of mesonotum with a transverse row of 11-12 smaller spines” ( Haenni and Baez, 2001). In a key to Nearctic Bibionidae, Hardy (1945) stated “comb comprised of seven to ten teeth” in female D. strigilatus McAtee, 1922 and “comb ... usually twelve to sixteen teeth” in female D. oklahomensis Hardy, 1937 (currently treated as a junior synonym of D. occipitalis Coquillett [in Baker, 1904]) and D. breviceps Loew, 1869 (currently treated as a junior synonym as a subspecies of D. tibialis loew, 1869).

Figures

FIGURE 2. Dilophus idanos Fitzgerald and Greenwalt sp. n., USNM 624840, male. A, Habitus; B, Left wing; C, Wing drawing. Scale bars equal 1.0 mm (A) and 0.5 mm (B).

FIGURE 3. Dilophus idanos Fitzgerald and Greenwalt sp. n., USNM 624840. A, Mesonotal spines; B, Spines of left fore tibia. C, Antennae (left arrow), palpus (right arrow) and spines (arrowheads) of right fore tibia. Scale bars equal 200 um.

FIGURE 4. Dilophus idanos Fitzgerald and Greenwalt sp. n., paratype. USNM 768228, male. A, Habitus; B, Mesono- tal spines; C, Spines of fore tibia. Scale bars equal 1.0 mm (A) and 0.2 mm (B, C).