Revision of Protonephrocerus Collin (Diptera: Pipunculidae)

. Two new species of Protonephrocerus Collin, P. flavipilus Skevington, Marques & Rafael sp. nov. and P. misionensis Skevington sp. nov. are described. The only other currently recognized species in the genus, P. chiloensis Collin, is redescribed and genetic data ascribed to this species are transferred to P. flavipilus sp. nov. All previously published genetic data refer to P. flavipilus sp. nov. The first records of Protonephrocerus in Argentina are documented.


INTRODUCTION
Protonephrocerus Collin is a key lineage of bigheaded flies (Diptera, Pipunculidae). Along with species in the extinct genus Metanephrocerus Aczél, they are currently placed in the subfamily Protonephrocerinae and considered to be sister to the diverse subfamily Pipunculinae (Collin, 1931;Aczél, 1948;Skevington & Yeates, 2000;Kehlmaier et al., 2014). Protonephrocerus chiloensis Collin is the only currently recognized extant member of the subfamily Protonephrocerinae and all previously published work on the genus, including sequences, are ascribed to this species. As it turns out, P. chiloensis is very local in its distribution and a much more widespread, undescribed species is what has been referred to in the literature. All published molecular data actually refer to this undescribed species. We describe this species below and clarify which of the two species have been cited in previous literature.
A third species for the genus was also discovered while preparing this manuscript. It is known from a single Protonephrocerus has also been considered to be sister to Nephrocerus Zetterstedt in the subfamily Nephrocerinae (Rafael, 1988;Rafael & De Meyer, 1992). They possess characters of both Pipunculinae and Nephrocerinae and decisions about which characters are pleisiomorphic and which are apomorphic have led to the differing hypotheses. Fossil data and molecular data suggest that this hypothesis is incorrect, but more data are needed to test this. The discovery of two additional extant species of Protonephrocerus will help with such future analyses.
Knowing the sister taxon of the diverse subfamily Pipunculinae may help us better understand what led to the radiation of this group. One thousand three hundred and forty-six of the 1462 currently recognized species of pipunculids are in the subfamily Pipunculinae (De Meyer, 1996;De Meyer & Skevington, 2000;Evenhuis & Pape, 2000;Skevington, 2020).

MATERIAL AND METHODS
Specimens examined in this study were obtained from the following collections (abbreviations follow Evenhuis (2020) Genitalia were excised by removing the whole abdomen, placing them in standard 1.5 mL Eppendorf tubes (Eppendorf Canada, Mississauga, Ontario, Canada) and immersing them in 85% lactic acid for no less than four hours on a 95 °C dry bath incubator (Fisher Scientific, Toronto, Ontario, Canada). Macerated abdomens were then placed on a glass depression slide and immersed in glycerin for dissection. The genitalia were then separated from the abdomen for comparison under a dissecting scope and were ultimately stored in genitalia vials with the specimens. Drawings were made of the genitalia of exemplar males and females. Genitalic terminology nomenclature follows Cumming & Wood (2017) and is discussed by Skevington (2001) with specific reference to Pipunculidae.
Exemplar specimens were photographed with a Canon 50D EOS Digital Camera with an MP-E 65mm macro lens (Canon Canada Inc., Mississauga, Ontario, Canada) and images were then stacked using Zerene Stacker Version 1.04 (Zerene Systems, Richland, Washington, United States of America). These images were further manipulated using Photoshop CS6 software (Adobe Systems Incorporated, San Jose, California, United States of America). Figure plates were prepared in Adobe Illustrator CS6 (Adobe Systems Incorporated). Illustrations of specimens were made by T.O.B. The microscope used for preparing illustrations was a Nikon SMZ1500 with a HI-150 High Intensity Illuminator (Nikon Canada Inc., Mississauga, Ontario, Canada). Measurements were made using a graticule following methods outlined in Skevington (2001).
All specimens are labeled with a unique reference number, typically in the format J. Skevington Specimen # n, CNC Diptera # n or CNCn. These have been shortened to follow the format JSSn, CNCDn, and CNCn respectively throughout the text. These numbers are used in public specimen database maintained at the CNC (Skevington, 2020) and are cited in the material examined sections and figure captions for illustrations to allow for repeatability of the research.
Molecular methods and analysis follow Motamedinia et al. (2020). Genetic work was carried out at the molecular laboratory of the CNC in Ottawa, Ontario, Canada and at the Biodiversity Institute of Ontario in Guelph, Ontario, Canada. Uncorrected pairwise genetic distances (p-distance) were calculated with Mega X (Stecher et al., 2020). Sequence accession numbers issued by GenBank (GB) are provided for each specimen in Table I.
The most recent key to genera that includes  Protonephrocerus can be found in Skevington & Yeates (2001).

RESULTS AND DISCUSSION
Protonephrocerus Collin, 1931: 52. Type species: Protonephrocerus chiloensis Collin, 1931, by original designation;Rafael, 1988: 465;Rafael & De Meyer, 1992: 652;Skevington & Yeates, 2001: 440. silvery white pubescence which are longer and denser on face. Posterior eye margin notched in middle. Posterior head margin running down straight. Occiput moderately wide, silver-pubescent laterally, sparsely brown-pubescent dorsally; covered with numerous short yellow hairs. Arista dark brown with yellow base. Postpedicel yellow, roughly ovate, slightly pointed ventrally. Pedicel and scape brown. Labellum and palpi yellow. Thorax Thorax (cf. Figs. 1, 2A-B, 4A). Proepisternum without propleural fan. Postpronotal lobe light brown to dark brown with yellow edges. Scutum dark brown to black dorsally, yellow to light brown on postalar callus. Pleuron entirely brownish black, only occasionally yellow around posterior spiracle and meron. Chaetotaxy: dorsocentral row of short bristles and terminating in 1 strong seta; intra-alar bristles small; 1 posterior supraalar; 2 postalars; 2 pairs of scutellars; 2 notopleurals, posterior one longer; proepimeron with some short proepimeral setae and anepimeron with one moderately strong bristle. Legs Legs mostly yellow with brown marks; long and slender; with dense pubescence and bristles; front and mid coxae with numerous long setae on anteroapical half; hind coxae with numerous shorter hairs along anteroapical and outer lateral margin. Front femur with posteroventral row of longer hairs in basal half, and several rows of shorter bristles; mid femur with posterior row of long hairs from base to apex (about as long as width of femur), and several rows of shorter bristly hairs; hind femur at least anterodorsally with 2 or more outstanding long bristles near apex (longer than width of femur), and posterior as well as antero-/posteroventral rows of longer bristly hairs, and several rows of shorter bristly hairs. First and second tarsal segments very long. Wings Wings (cf. Fig. 1) long, slightly more than three times as long as broad; hyaline. Pterostigma present, brownish. Third costal section about four times as long as fourth. Cross-vein r-m placed near basal third of discal medial cell (dm); vein M. present in most species (absent in P. misionensis), long but never reaching wing margin. Anal lobe absent. Halter yellow. AbdomenAbdomen (cf. Figs. 1, 4A) slightly narrower than and about twice as long as thorax; entirely dark brown to black, hairs longest along lateral and posterior margins of tergites; tergites 1-7 and syntergosternite 8 visible from dorsal; tergite 1 with a tuft of long yellow bristles anterolaterally; sternites 1-5 light brown.
Male. Male. Head Head ( Figs. 2A-C). Eyes touching on frons for about 2.4 times as long as frontal triangle. Pedicel brown with 4 short dorsal bristles and 3 ventral bristles, two of them short and one long. Thorax Thorax ( Figs. 2A-C). Postpronotal lobe dark brown with yellow edges, with 3-4 yellow hairs on posterior edge. Scutum mostly dark brown to black with black bristles. Scutellum black with 2 pairs of strong black marginal setae and numerous small black bristles over entire surface. Wing Wing with M. vein present. Legs Legs mostly yellow with brown marks medially on fore and mid femora, and distal half of hind femur; long and slender; with dense pubescence and hairs. Small hairs on trochanters, femora, tibiae and tarsi mostly black; front femur with posteroventral row of longer hairs in basal half, and several rows of shorter bristles; mid femur with posterior row of about 20 long hairs from base to apex, and several rows of shorter bristly hairs; hind femur anterodorsally with 2 outstanding long bristles near apex and several rows of shorter bristly hairs. Abdomen Abdomen (Figs. 2A-D) dark brown to black with black bristles. Tergite 1 with 13 yellow lateral hairs anterolaterally. Tergites 6 and 7 large and shining, densely covered with pubescence and bristles; tergite 7 scarcely visible from above. Sternites 1-5 light brown. Sternite 5 rectangular (Fig. 2D). Sternites 6 and 7 yellow, not visible dorsally; sternite 6 dark brown basally, markedly desclerotized in apical half, bulbous, protruding beyond sternite 7. Syntergosternite 8 dark brown, large, without membranous area. Male Male genitalia genitalia (Figs. 3A-E) large, grayish pruinose. Epandrium brownish yellow, shorter medially; symmetrical. Surstyli stubby, symmetric, with cluster of long bristles on dorsomedial surface. Hypandrium small, about half length of simple-shaped phallic guide complex. Gonopods reduced and symmetrical, with rounded apex. Phallic guide V-shaped, with two setae medially. Phallus short, bifid. Female Female. As male except: Ovipositor yellow, short, piercer distinctly curved upwards, not sinuous dorsally, in lateral view, 0.8-0.9 mm (Fig. 3F) Remarks Remarks. Only seven specimens of this species have been collected so any inferences about behavior or phenology are tentative. The flight period appears to be predominantly between late November and late December in lowland, coastal Nothofagus forests. We were unable to sequence any of the known specimens of this species (presumably due to degradation of the tissues over time).
Male. Male. Head.Head. Eyes touching on frons for about 2 times as long as frontal triangle. Pedicel brown with 4 short dorsal bristles and 3 ventral bristles, two of them short and one long. Thorax Thorax (Figs. 4A-B). Postpronotal lobe brownish yellow, with 5 yellow hairs on posterior edge. Scutum mostly dark brown to black with yellow bristles. Scutellum black with 2 pairs of strong yellow marginal setae and numerous small yellow bristles over entire surface. Wing Wing with M. vein present. Legs Legs mostly yellow with light brown on basal half of fore femur, medially on mid femur and distal half of hind femur; long and slender; with dense pubescence and bristles. Small hairs on trochanters, femora, tibiae and tarsi mostly black; front femur with posteroventral row of longer hairs Etymology Etymology. From the Latin flavus for yellow and pilus for hair, in reference to the extensive yellow pile on the thorax and abdomen. Remarks Remarks. This is the most common species of Protonephrocerus and appears throughout the literature as P. chiloensis (Kehlmaier et al., 2014, Fig. 32, page 33;Rafael, 1988, Figs 3-9, page 468 and male description on page 469 based on P. flavipilus sp. nov.; Skevington & Yeates, 2000, specimens coded in matrix and used in phylogeny based on P. flavipilus sp. nov.). All previously sequenced specimens of Protonephrocerus are of this new species (JSS4485 -GenBank numbers AF154736 and AF154811-12S and 16S ribosomal DNA sequenced and were identified as P. chiloensis in Skevington & Yeates (2000); JSS3651 -GenBank numbers AF154737 and AF154812 -12S and 16S ribosomal DNA sequenced and were identified as P. chiloensis and incorrectly listed as JSS945 in Skevington & Yeates (2000)). Specimens have been collected between mid-September and early March, mostly between 340 and 1400m elevation (with one specimen, JSS3651, as low as 90m). Specimens have been collected in Nothofagus and Fitzroyia forests as well as meadows.
The holotype specimen has been DNA barcoded (Table I). This species differs from P. misionensis by 10.1 to 10.9% (uncorrected pairwise distance). Intraspecific variation is 0-3.5%. There are two genetic clusters of specimens with 0-1.7% pairwise divergence within one cluster (n = 8 -flavipilus A) and 0-0.2% within the other (n = 3 -flavipilus B) (Table I). Pairwise divergence between these clusters is 1.7 to 3.5%. Although this is suggestive that two species may be involved, we do not find any morphological support for this. Specimens from the same location and date occur in each cluster. The only variable morphological character, the shape of the surstylus tip (long vs. short), is also suggestive of multiple species, but examination of many specimens shows that this character varies continuously and individuals with short-tipped surstyli occur in both Female. Female. Head Head (Fig. 6). Pedicel brown with 2 short dorsal bristles and 2 ventral bristles. Thorax Thorax (Fig. 6). Postpronotal lobe dark blackish brown. Scutum mostly dark brown to black with yellow bristles. Scutellum black (bristles missing, rubbed off). Wing Wing with M. vein absent. Legs Legs mostly yellow with dark brown on basal half of femora; long and slender; with dense pubescence and bristles. Small hairs on trochanters, femora, tibiae and tarsi mostly black; mid femur with posterior row of about 15 long yellow hairs from base to apex, and several rows of shorter bristly hairs; hind femur anterodorsally with 6-8 outstanding long yellow hairs near apex. Abdomen Abdomen (Fig. 6) dark brown to black with yellow bristles. Tergite 1 Etymology Etymology. Named after the Misiones region where the only known specimen was collected.
Remarks Remarks. The Misiones region where this species occurs is the southern extent of the Atlantic Forest ecozone that supports the highest biodiversity in Argentina. This species distribution is disjunct from that of other Protonephrocerus species, lying 1650 km NE of the closest collecting locality in central Chile (near Santiago).
We are generally opposed to describing pipunculids from females as the males are more character rich. However, we feel that the DNA barcode, the diagnostic missing M. wing vein, the short, narrow ovipositor, as well as the disjunct distribution makes it extremely unlikely that this species will become a nomen dubium like so many pipunculids described solely from females. We also feel that it is important to name this species to make it visible for conservation reasons. The Atlantic Forest has been decimated and much of the flora and fauna has disappeared or their distributions have become severely fragmented. This is thus the first stage of getting to know this species. Hopefully awareness of it will spur further research.
The holotype specimen has been DNA barcoded (Table I). This species differs from its closest relative (P. flavipilus sp. nov.) by 10.1-10.9% (uncorrected pairwise distance). This considerable genetic difference from its nearest neighbour along with the missing M. wing vein suggest that this species has had a long independent history apart from other Protonephrocerus species, and as such is a significant piece of the puzzle needed to unravel the relationships of the family. Hopefully more specimens will be found now that we are aware of its existence.
Protonephrocerus sp. 1988-01 (Figs 7,8) Additionally, one putative undescribed species was mentioned by Rafael (1988) based on one male in poor condition (crushed), characterized by vein r-m more basal, placed at level of the end of vein Sc (versus slightly beyond the vein Sc in P. chiloensis, P. flavipilus sp. nov. and P. misionensis sp. nov.), male sternite 5 concave on posterior border (versus rectangular in P. chiloensis and with two invaginations on posterior border in P. flavipilus sp. nov.), surstyli bigger, as long as wide in dorsal view (versus wider than long in P. chiloensis and P. flavipilus sp. nov.) and ejaculatory apodeme umbeliform (versus horn-shaped in P. chiloensis and P. flavipilus sp. nov.). The male specimen is from Chile, Ñuble, Las Trancas [-36.605707, -72.073482; CNC1766810], collected 1.ii.1983 by L.E. Peña and belongs to INPA. The undescribed male specimen does not fit the species described here and is another indication that the genus is more speciose than expected. A second specimen, also in very poor condition, appears to be this species and was collected 19-30.xii.1983 by L.E. Peña at Chile, Chillán, Shangrilá [-36.891458, -71.47309; CNC1766811] (also in INPA).