Protanthomyza grimaldii sp. nov., a further member of the extinct subfamily Protanthomyzinae (Diptera, Anthomyzidae) from Baltic amber

Abstract A new fossil species, Protanthomyza grimaldiisp. nov. (Diptera, Anthomyzidae), is described from Baltic amber (Eocene, 48–34 Ma) based on two (male and female) inclusions. It is the ninth species of the †genus Protanthomyza Hennig, 1965 and †subfamily Protanthomyzinae Roháček, 1998. Adult morphology of P. grimaldiisp. nov. revealed that the rich chaetotaxy of the thoracic pleuron, two anal veins and presence of the anteroventral process of the epandrium are plausibly shared by all species of Protanthomyza. Relationships of the new species, which belongs to a group lacking the ctenidial spine on the fore femur, are discussed.


Introduction
Fossil taxa of Anthomyzidae (Diptera) were reviewed by Roháček (2013a), with one species subsequently added (Roháček 2014). A total of 12 ancient (all Tertiary) valid and named species belonging to four genera are currently recognized. The majority of fossil species of Anthomyzidae were described from Baltic amber (48-34 Ma) and only one, viz. Grimalantha vulnerata Roháček, 1998, originates from Dominican amber (Miocene,. Only four of the ancient species belong to the "modern" subfamily Anthomyzinae Czerny, 1903, viz. the above Grimalantha vulnerata from the Miocene (see Roháček, 1998) and Lacrimyza lacrimosa Roháček, 2013, L. christelae Roháček, 2013 and Reliquantha eocena Roháček, 2014 from the Eocene. The latter species belongs to a genus originally described by Roháček (2013b) for an extant species, R. variipes Roháček, 2013 from Great Britain. All remaining ancient Anthomyzidae are members of the exclusively fossil subfamily Protanthomyzinae Roháček, 1998 and belong to its only genus Protanthomyza Hennig, 1965. Eight named species of this genus are currently recognized, viz. P. collarti Hennig, 1965, P. hennigi Roháček, 2013, P. hoffeinsorum Roháček, 2013, P. krylovi Roháček, 2013, P. loewi Roháček, 2013, P. meunieri Roháček, 2013, P. presli Roháček, 2013and P. tschirnhausi Roháček, 2013, all of which were treated in detail by Roháček (2013a). However, Roháček (2013a: 451) also presented the diagnosis of one more new species of Protanthomyza which was left unnamed due to insufficient material (a single female with some parts of the body obscured). This unnamed species was considered distinctly different from all other known members of the genus and, therefore, it was also counted among them for the estimation of species diversity of Anthomyzidae in the Eocene Baltic amber forest ecosystem (Roháček 2013a: 470). Recently, Mrs. Christel Hoffeins purchased a nicely preserved Protanthomyza inclusion and provided it for study. Detailed examination of this specimen revealed it to be the formerly unknown male conspecific with the unnamed "Protanthomyza sp. nov." female dealt with in Roháček (2013a). To supplement the latter monographic treatment, this new species is described in detail below, including the re-examination of the above female specimen. This is only the third fossil species of Protanthomyza (and the fourth of all ancient Anthomyzidae) in which both sexes are known.

Preparation of amber specimens
The methods of preparation of amber stones with fly inclusions were described in detail by von Tschirnhaus and Hoffeins (2009). The amber specimens examined had already been cut out of the original stones, ground and polished as close and as parallel as possible to the frontal, dorsal and lateral sides of the fly and one of them had been subsequently embedded in artificial resin (also ground and polished) (Hoffeins 2001) to facilitate its stereoscopic investigation. This process left a significant cut portion of one stone bearing syninclusions, which remains deposited in CCHH.

Techniques of investigation
The amber inclusions were examined, drawn and measured using two types of binocular stereoscopic microscopes (Reichert, Olympus). Legs were drawn on squared paper using a Reichert binocular microscope with an ocular screen. The specimens were either photographed by a Canon EOS 60D digital camera with Canon MP-E 65 mm 1-5× macro lens or by a Canon EOS 5D Mark III digital camera with a Nikon CFI Plan 10×/0.25NA 10.5mm WD objective attached to a Canon EF 70-200 mm f/4L USM zoom lens. The specimen photographed by means of the latter equipment was repositioned upwards between each exposure using a Cognisys StackShot Macro Rail and the final photograph was compiled from multiple layers (35) using Helicon Focus Pro 7.0.2. The final images were edited in Adobe Photoshop CS6. Some illustrations were drawn from these and some other macrophotographs and details were inked based on direct observation at higher magnification using a binocular microscope. Measurements: Six characteristics were measured -body length (measured from anterior margin of head to end of cercus, thus excluding the antenna), wing length (from wing base to wing tip), wing width (maximum width), index Cs 3 : Cs 4 (= ratio of length of 3 rd costal sector : length of 4 th costal sector), index r-m\dm-cu : dm-cu (= ratio of length of section between r-m and dm-cu on cell dm : length of dm-cu) and index r-m\ dm-cu : CuA 1 (= ratio of length of section between r-m and dm-cu on cell dm: length of apical portion of CuA 1 ).
Morphological terminology follows that used in Roháček (2006Roháček ( , 2009) and Roháček and Barber (2016), including terms of the male hypopygium to be in continuation with Roháček (2013a), except where "orbit" is replaced with "orbital plate". Male terminalia terminology is largely based on the "hinge" hypothesis of the origin of the eremoneuran hypopygium, re-discovered and documented by Zatwarnicki (1996) and, therefore, the following alterations of terms of the male genitalia (against those used by other hypotheses) need to be listed (terms used here first): epandrium = periandrium, gonostylus = surstylus. Morphological terms of the male abdomen and terminalia are depicted in Figs 8, 9, those of the female abdomen in Fig. 13. The synonymous morphological terms of adult structures and their abbreviations as used in the recent manual of Afrotropical Diptera (Cumming and Wood 2017) are given in parentheses in the list of abbreviations below.
Description. Male (Figs 1, 2). Total body length ca 2.6 mm; general colour brown to blackish brown; only head and some extremities partly ochreous to yellow.
Head  higher than long, anteriorly somewhat angularly protruding in front of eye margin (Fig. 5). Occiput very slightly concave, blackish brown. Frons relatively narrow, blackish brown only posteriorly around ocellar triangle, pale brown in anterior half to ochreous yellow at anterior marginal area surrounding frontal lunule. Frontal triangle not developed; ocellar triangle blackish brown, distinctly protruding and delimited by marginal groove. Ocelli large (Figs 3, 4). Orbital plate lighter (ochreous yellow) anteriorly, becoming pale brown posteriorly where contrasting with blackish-brown vicinity of ocellar triangle. Frontal lunule long, ochreous yellow. Face relatively narrow, ochreous, medially lighter and somewhat depressed; parafacialia and anterior half of gena dirty yellow and narrowly brown bordered (gena ventrally); posterior half of gena and postgena brown; mouthparts yellow to ochreous, clypeus pale brown, palpus dirty yellow. Cephalic chaetotaxy (Figs 4,5): pvt (only left one visible) relatively short, convergent but not crossed; vti longest of cephalic setae, slightly inclinate; vte strongly exclinate and only slightly shorter than vti; 3 distinct ors, all slightly reclinate, posterior ors longest (about as long as vte), others becoming slightly shorter anteriorly; oc relatively thin (not longer than middle ors), proclinate (and unnaturally crossed in holotype), arising inside ocellar triangle; anterior half of frons with about 5 or 6 pairs of microsetae, mostly medially in front of ocellar triangle but a few (1 or 2) also between anterior and middle ors; vi distinct ( Fig. 5), about 3 times as long as foremost peristomal setula; no subvibrissa; 4 or 5 weak proclinate peristomals; postocular setulae in two rows as usual but with only 3 setulae in inner row (Fig. 4); outer row of postoculars long, reaching ventral eye margin; postgena with 2 setae, anterior short, posterior longer. Palpus slender, elongate, with a few (3 visible) minute setulae distally (Fig. 5). Mouthparts relatively short. Eye bare, relatively large and strongly convex, suboval, anteriorly regularly rounded, with only posterior margin somewhat straighter; its longest diameter almost vertical and 1.16 times as long as shortest diameter. Gena low, its shortest height about 0.08 times as long as shortest eye diameter. Antenna medium-sized, generally porrect ( Fig. 5) but directed anteroventrally to ventrally, with dark-brown basal segments and pale-brown to ochreous 1 st flagellomere. Pedicel with 1 longer seta and several microsetae; 1 st flagellomere oval, laterally compressed, with very short, dense and dark pilosity; arista about 1.8 times as long as antenna, entirely bare (Fig. 5), 2 basal segments slightly widened.
Legs brown to ochreous, femora darkest, fore coxa pale ochreous. f 1 lacking ctenidial spine; 5 (2 longer) distinct widely-spaced setae in posterodorsal row; setae in posteroventral row more numerous but short and weak. f 2 with 2 or 3 anterior setae near middle (cf. Fig. 14), otherwise shortly setulose as is f 3 . t 2 with distinct ventroapical seta (about as long as maximum width of t 2 ) and 2 or 3 small setae adjacent to the latter; t 1 , t 3 and all tarsi simply setulose but basitarsi of all legs with ventrobasal setulae somewhat longer than others.
Wing (Figs 1, 6) moderately long and narrow, widest at distal third; veins brown to pale brown, membrane unicolourous, pale-brown tinged; C with more or less distinct subcostal break and somewhat attenuated at humeral cross-vein. C extended to apex of M, densely uniformly setulose on Cs 2 (from subcostal break to apex of R 2+3 ), finely short-pilose more distally (on Cs 3 and Cs 4 ); Sc distinct, separate almost along its entire length, only apically fused with R 1 to form preapical kink (see Fig. 6, pk); R 1 short, dilated distally due to fusion with Sc; R 2+3 long, very slightly sinuate to almost straight and only its extreme apex slightly upcurved to C; R 4+5 slightly recurved in distal half, divergent from R 2+3 and apically slightly convergent with M; M almost straight. Cell dm of moderate length, narrow proximally and much widened distally, with angle of anterior outer corner obtuse while that of posterior outer corner distinctly acute (cf. Fig. 12); r-m situated in basal third of cell dm; dm-cu straight; apical portion of CuA 1 much shorter than distance between r-m and dm-cu, distinctly longer than dm-cu and almost reaching wing margin; A 1 relatively long but ending far from wing margin; A 2 well developed, slightly shorter than A 1 (Fig. 6); alula distinct but narrow (Fig. 6). Wing measurements: length ca 2.4 mm, width ca 0.8 mm, Cs 3 : Cs 4 = 1.38, r-m\dmcu : dm-cu = 3.50, r-m\dm-cu : CuA 1 = 2.50. Haltere (Figs 4, 8) pale ochreous, knob relatively large, darker dorsally.
Postabdomen. T6 distinctly shorter than T5, transverse; T7 yet shorter and also narrower than T6. S6 strikingly broad and transverse, wider but shorter than S5 and apparently larger than adjacent T6 and laterally almost reaching the latter; S7 only half length of S6, strongly transverse and laterally meeting with sides of T7 (Fig. 13); both S6 and S7 finely setulose. Apex of postabdomen obscured, only T8 discernible as small bare(?) sclerite (Fig. 13), less than half length of T7. Cercus not visible.

Discussion
Protanthomyza grimaldii sp. nov. was previously recognized as a new species by Roháček (2013a: 451) but left unnamed because of insufficient material (a single female having a number of characters not visible). Thanks to the efforts of Christel Hoffeins, a male specimen conspecific with this female was recently obtained for examination and enabled the description and naming of this species. This is the third species of Protanthomyza where both sexes are known; formerly, the male and female were described only in P. krylovi (1 male and 2 females found in one piece of Baltic amber) and P. tschirnhausi (1 male and 1 female in separate pieces of Bitterfeld amber), see Roháček (2013a).
Protanthomyza grimaldii belongs to a group of species lacking a ctenidial spine on the fore femur and both sexes are correctly keyed by Roháček (2013a: 442, as P. sp. nov.). With its bare arista and generally similar chaetotaxies of the head and thorax, it most closely resembles P. loewi (known only from the female), which could be its nearest relative. However, P. grimaldii can be easily distinguished from P. loewi by the distinctly smaller 1 st flagellomere of the antenna (Fig. 5, cf. Roháček 2013a, fig. 8B), the ocellar triangle delimited by a groove (Fig. 4), the less elongate wing with more divergent R 2+3 and R 4+5 (Fig. 12, cf. Roháček 2013a, fig. 7D) and the shorter and wider female abdomen with short T7 and broadly transverse S6 and S7 (Fig. 13, cf. Roháček 2013a, fig. 8A), apart from other smaller dissimilarities in head colouration, length of ppl setae, f 1 and t 2 chaetotaxy, etc. In the male, P. grimaldii differs from all five other species where the male is known (cf. Roháček 2013a) by the elongate and posteriorly tapered epandrium with a robust, hook-like, curved anteroventral process (Fig. 9, avp). Only in P. tschirnhausi is the epandrial process similarly robust and flat but it is simply triangular (not hooked) and the epandrium is short, almost globose (not elongate) (cf. Roháček 2013a, fig. 12C). Moreover, P. tschirnhausi differs markedly from P. grimaldii in a number of other characters including the ciliate arista, very large eyes, only 2 dc and a strong ctenidial spine on f 1 , see Roháček (2013a). Unfortunately, the male of P. loewi remains unknown and, consequently, the previously suggested relationship of that species with P. grimaldii cannot be confirmed by examination of characters of the male terminalia.