Flower flies (Diptera, Syrphidae) of French Polynesia, with the description of two new species

The flower flies (Diptera, Syrphidae) of French Polynesia are revised. A total of nine syrphid species were recorded from the five archipelagos of French Polynesia. Among them are two species new to science, Allograpta jacqi Mengual & Ramage sp. nov. and Melanostoma polynesiotes Mengual & Ramage sp. nov., and a new record for this country, Syritta aenigmatopatria Hardy, 1964. We provide DNA barcodes for all flower fly species of French Polynesia, making the syrphid fauna of this country the first one in the world to be entirely barcoded. New data on biology, flowers visited and some taxonomic notes are provided. An identification key for the species of Syrphidae in French Polynesia is given, as well as an identification key for the species of Melanostoma Schiner, 1860 in the Australasian and Oceanian Regions.


Introduction
The terrestrial arthropods of French Polynesia form a peculiar fauna, with several missing orders among the Hexapoda, a phenomenon called taxonomic disharmony (Roderick & Gillespie 2016).The absence of major taxonomic groups is to some degree counterbalanced by a high endemism rate resulting from numerous local speciation events.Most of the Pacific Basin was colonized by species from New Guinea and adjacent areas via over-water dispersal (Miller 1996;Ramage 2017).Munroe (1996) showed that there is a progressive decrease in the number of founding stocks and an increase in the proportion of radiating speciation with distance from the Papuan source areas, also known as the 'radiation zone' (MacArthur & Wilson 1967).
The pollinator entomofauna is unfortunately poorly known in French Polynesia.The only published records of flower / insect relationships concern a species of Megachilidae, Megachile diligens Smith, 1879 "taken on the blossoms of a trailing bean near the shore", most likely a species of Vigna Savi (Fabaceae) (Cheesman 1928), and the highly specialized pollination mutualism between the seedfeeding Epicephala Meyrick, 1880 moths (Lepidoptera, Gracillariidae) and species of Phyllanthus L. s. lat.(Phyllanthaceae) (Hembry et al. 2012(Hembry et al. , 2013a(Hembry et al. , 2013b)).Studies on the pollinator entomofauna of French Polynesia are ongoing and the present work is a first contribution regarding pollinator insects.
Insects are represented in French Polynesia by 2497 species, with 67% of them being endemic, and there are 342 species of Diptera recorded for French Polynesia, of which 210 are endemic (70%) (Ramage 2017).World-wide, flower flies (Diptera: Syrphidae) are one of the most species-rich dipteran families with over 6000 valid species (Thompson 2013).Also known as hoverflies, syrphid adults feed on pollen and nectar and use flowers as mating sites.Their larvae, on the other hand, have unusually diverse natural histories and include mycophages of fungal fruiting bodies, phytophages of numerous plant families, pollen feeders, saprophages in media as diverse as dung, nests of social Hymenoptera, decaying wood and water bodies of numerous types, and predators of a range of other arthropods, mostly soft-bodied Hemiptera, caterpillars and immatures of ants and beetles (Rotheray & Gilbert 1999;Rojo et al. 2003;Weng & Rotheray 2008;Reemer & Rotheray 2009;Zuijen & Nishida 2011;Dumbardon-Martial 2016), but also on adult flies (Ureña & Hanson 2010).
The aims of the present study are: 1) to describe this unknown taxon reported by Ramage et al. (2017) as Melanostoma polynesiotes Mengual & Ramage sp. nov.,2) to describe another new species collected during the latest expedition to Tahiti, Allograpta jacqi Mengual & Ramage sp. nov.,3) to report a new record for French Polynesia, i.e., Syritta aenigmatopatria Hardy, 1964, and to critically review the doubtful species records of Syrphidae, 4) to present the records of the flowers visited by syrphids in French Polynesia during the field expeditions and 5) to provide new DNA barcodes for all known flower flies of French Polynesia, making the syrphid fauna of this country the first one in the world to be entirely barcoded.An identification key for the species of Syrphidae in French Polynesia is provided, as well as an identification key for the species of Melanostoma Schiner, 1860 in the Australasian and Oceanian Regions.

Identification and format
Original descriptions and inspection of type material were used for syrphid identification, together with a few existing identification keys such as Lyneborg & Barkemeyer (2005) for the genus Syritta Lepeletier & Serville, 1828 and Carvalho Filho & Esposito (2009) for the genus Ornidia Lepeletier & Serville, 1828.Differential diagnoses, synonymies, references and distributions are given for all species included in the study.New species are described following the terminology of Thompson (1999) and Mengual (2012).The abbreviations used for collections follow the standard of the Systema Dipterorum (Thompson 2013) In the description of type labels, the contents of each label are enclosed within double quotation (" "), italics denote handwriting and the individual lines of data are separated by a double forward slash ( // ).The holding institution is indicated at the end of each record between parentheses.
All measurements are in millimetres and were taken using a reticule in a Leica M165 C microscope.Photographs were composed using the Zerene Stacker program ver.1.04 (Richland, WA, USA), based on images of pinned specimens taken with a Canon EOS 7D mounted on a P-51 Cam-Lift (Dun Inc., VA, USA) and with the help of Adobe Lightroom (ver.5.6).Body length was measured from the anterior oral margin to the posterior end of the abdomen, in lateral view.Wing length was measured from the wing tip to the basicosta.

DNA-barcoding
Specimens with a DNA barcode are indicated in the text by a GenBank accession number and/or a BOLD Process ID.GenBank accession numbers starting with KX05 and BOLD IDs starting with SYC were generated using the protocols explained in Ramage et al. (2017) as part of the SymbioCode initiative (https://doi.org/10.5883/DS-SYMC).New DNA barcodes generated for this study as well as other marker sequences (GenBank accession numbers starting with MF44 or MH28) were produced using the DNA primers and PCR amplification protocols described in Mengual et al. (2008Mengual et al. ( , 2012) ) at the ZFMK.Entire specimens or remnants of specimens were preserved and labelled as DNA voucher specimens for the purpose of morphological studies, except for one male of Melanostoma polynesiotes sp.nov., which was totally destroyed in the DNA extraction process and no body parts remain.

Results
A total of 228 specimens were studied for the present survey.Details are given in the Material examined section under each species.One species was recorded for the first time for French Polynesia, Syritta aenigmatopatria Hardy, 1964.In addition, two species new to science were found at high elevations on the island of

Differential diagnosis
Species with yellow face with a medial black vitta, scutum black with a continuous lateral yellow vitta from postpronotum to scutellum, scutellum yellow with a median black macula, terga 2 and 5 with two lateral yellow maculae and terga 3 and 4 with a broad yellow fascia.It differs from A. nigripilosa only in the wing microtrichia, as stated in the key.

Status in French Polynesia
Present.

References
Aubertin & Cheesman 1929: 172 (records);Hull 1937: 83 (catalogue);Vockeroth 1969: 129 (list); Thompson & Vockeroth 1989: 441 (catalogue); Mengual et al. 2009: 15 (list).Aubertin & Cheesman (1929) recorded this species for the first time from French Polynesia (Tahiti, Raiatea and Bora-Bora) and mentioned that it was abundant on fern-covered slopes.Hull (1937) listed this species from Fiji, the Marquesas Islands and Tahiti, while Vockeroth (1969) mentioned it from Samoa.We were not able to collect specimens during any field expeditions, even though it might be abundant.The only studied specimen from French Polynesia that has the same wing microtrichia pattern as the holotype of A. amphotera is a male from Rurutu in the BMNH.This species and A. nigripilosa are extremely similar, and after the study of the type material by XM, it is still not clear whether they are the same taxon or two different species, one located in the western and southern parts of the south Pacific Ocean (A. amphotera) and the other (A.nigripilosa) restricted to the central part.In the BMNH, there is a male (Cook Islands: Rarotonga, Avatiu Valley, 28 Mar.1999, C. Wilkinson leg.) with the cell bm bare on the basal ⅓.This male does not match the type of A. amphotera, indicating a potential intraspecific variability or some damage during preservation of this specimen.This specimen might also be a male of A. nigripilosa that reached the Cook Islands, broadening the distribution range of this species.At this point, we should consider the records from French Polynesia by Aubertin & Cheesman (1929) as doubtful, since they were reported prior to the description of A. nigripilosa by Hull (1944).More specimens are needed to understand the variability of these Oceanic species of Allograpta, but the analysis of the available DNA barcodes, including the specimens of Allograpta from Fiji mentioned above, resolved Fijian and Polynesian specimens in the same cluster, with a bootstrap support value of 100 in the Neighbour-Joining analysis.Moreover, Fig. 2. A. Allograpta amphotera (Bezzi, 1928), holotype, ♂, dorsal view.B. Allograpta jacqui Mengual & Ramage sp.nov., holotype, ♂ (ZFMK-DIP-00019707), dorsal view.C. Allograpta nigripilosa (Hull, 1944), ♂, dorsal view (photograph by F. Jacq).D. Ischiodon scutellaris (Fabricius, 1805), ♂ (ZFMK-DIP-00019713), dorsal view.Scale bars = 1 mm.(Hull, 1944), ♂ (photographs by F. Jacq).E. Lateral view.F. Frontal view.G-H.Ischiodon scutellaris (Fabricius, 1805), ♂ (ZFMK-DIP-00019713).G. Lateral view.H. Frontal view.Scale bars = 1 mm.BOLD groups all of these specimens in the same Barcode Index Number (BIN), BOLD:AAZ6685 (https://doi.org/10.5883/BOLD:AAZ6685).Although resolved in the same clade, the barcodes of Fijian and Polynesian specimens form two different clusters.In further studies we will try to collect more individuals of Allograpta to test the molecular variability of these two species; at the current stage, the synonymy of A. amphotera and A. nigripilosa seems plausible.

Differential diagnosis
This species differs from the two other species of Allograpta in this paper by having the scutellum entirely yellow and terga 4 and 5 with two longitudinal yellow vittae connected (or not) basomedially with the lateral oblique maculae.

Remarks
This is a widespread species on the American continents, introduced in Hawaii, and not present in French Polynesia.We have included this taxon in our species list because Fluke (1942) mentioned records from the Crocker Expedition in Rikitea (French Polynesia, Gambier Islands, Mangareva Island). Charles Templeton Crocker (1884-1948) made several expeditions from 1931 to 1938 (SNAC 2016), but visited French Polynesia only once.Between September 15 th 1934 and April 16 th 1935, the American Museum of Natural History funded a scientific expedition to the South Pacific, including Eastern Polynesia, the so-called Templeton Crocker Pacific Expedition, with the following ports: Marquesas, Tuamotus, Australs, Mangareva, Pitcairn, Easter, Valparaiso, Chinchas and the Galapagos Islands.Van Duzee (1937) gave details on this expedition.Unlike other expeditions (Curran 1934(Curran , 1936)), we could not find any report of Diptera collected during the 1934-1935 journey.The interesting part for the present dilemma is that the expedition visited only one island of the Gambier Islands, Mangareva (Rikitea is a small town on Mangareva), to get some fuel before they left towards Pitcairn Island (Chapin 1935).On the return journey, the expedition visited Easter, Juan Fernandez and the Galapagos Islands, and A. exotica has been reported from Easter Island (Thompson 2015) and from Juan Fernandez (Fluke 1955).We strongly believe that A. exotica does not occur in French Polynesia and if a specimen with such a label exists, it is very likely due to mislabelling.Another option is a misidentification of A. nigripilosa by Fluke, although this seems unlikely based on his taxonomic expertise.Consequently, A. exotica is not included in the identification key.Mengual & Ramage sp. nov. urn:lsid:zoobank.org

Differential diagnosis
Species with face medially black, yellow laterally, scutum and scutellum black, and abdomen black with lateral small yellow maculae on terga 1-5.Easy to distinguish from other species of Allograpta by the general dark coloration.

Etymology
The specific epithet refers to the family name of the first collector of the species, Fred Jacq, a great naturalist and photographer.The species epithet is to be treated as a noun in the genitive case.

Male
Length (n = 1).Body 10.5 mm; wing 9.0 mm.3d, 6C, e).Face with a distinct tubercle, shiny, yellow with a medial broad black vitta, which does not reach oral margin, and two small black maculae on eye margin, yellow pilose with some black pile laterally and dorsally; gena black; lunule black; frons shiny black, with two small yellow maculae on eye margin at level of antennal insertion, black pilose; vertical triangle black, black pilose; antenna dark, black pilose; arista bare; eye bare, holoptic; occiput dark, covered with thick silver pollinosity on ventral ⅔, silver-white pilose on ventral ⅔ and black pilose on dorsal ⅓.

head (Figs
thorax (Figs 2B,3C,.Scutum shiny, black with small yellow markings on posterior notopleuron and posterior postpronotum, with relatively long yellow pile with some abundant black pile; scutellum shiny black with long yellow and black pile, subscutellar fringe with long dark pile.Pleuron black, except posterior anepisternum yellow on posterior ½ and katatergum with a yellow macula, mostly shiny with some pale pollinosity anteriorly, yellow pilose; metaepisternum bare; metasternum with long yellow pile; calypter pale basally, darker apically, with dark pile; plumula long, yellow; halter yellow; posterior spiracular fringes yellow.
Legs.Coxae and metatrochanters dark; pro-and mesofemora yellow with a dorsal dark area; pro-and mesotibiae yellow with a medial dark annulus; pro-and mesotarsi black; metaleg black; black pilose with some yellow pile on metacoxa.

Female
Unknown.

Geographical distribution
Species only known from Tahiti (French Polynesia).

Status in French Polynesia
Endemic.

Flowers visited
No records.

Remarks
This is a singular species due to its body coloration, without the common yellow fasciae of other species of the genus Allograpta and the with scutellum entirely shiny black.The Australian species of this genus were reviewed by Mengual & Thompson (2015), but a broader systematic revision is needed for the Australasian and Oceanian Regions (see Mengual et al. 2009).

Differential diagnosis
Extremely similar to A. amphotera, but it has a different microtrichia pattern on the wings, as stated in the key.

Geographical distribution
French Polynesia (Society Islands).

Status in French Polynesia
Endemic.

Flowers visited
No records.

Differential diagnosis
Species with basoflagellomere elongate and subacute apically, face yellow, metasternum bare and abdominal terga 2-5 distinctly marginated (Figs 2D, 3G-H).It differs from S. grandicornis by having a dorsal yellow macula on the katepisternum and the male metatrochanter with a ventral spine-like process or calcar.

Geographical distribution
Widespread species in the Oriental and Australasian Regions (Japan west to India and south to Australia, Papua New Guinea, New Caledonia, Micronesia, Samoa and other Pacific islands, including French Polynesia).

Status in French Polynesia
Present; recorded from Marquesas Islands, Gambier Islands, Tuamotu Islands, Austral Islands and Society Islands.

Flowers visited
Ischiodon scutellaris has been seen visiting two plant species, Hippobroma longiflora (L.) G.Don and Apetahia raiateensis Baill.(Campanulaceae).Apetahlia raiateensis is a shrub endemic to the three trachytic plateaus of Raiatea and it is listed as Critically Endangered by the IUCN (UICN France, MNHN & DIREN Polynésie française 2015).In terms of conservation, the identification of the pollinators of this species may be useful (F.Jacq, pers.comm.).A couple of males at the MNHN were collected on flowers of Messerschmidia argentea (L.f.) I.M.Johnst.(= Tournefortia argentea L. f. = Heliotropium foertherianum Diane & Hilger, 2003).Due to the overall similarity between females of I. scutellaris and A. nigripilosa in the field, some records of visited flowers cannot be confidently assigned to one of these species and they are not included.Cochereau (1966Cochereau ( , 1974) ) observed the food web of I. scutellaris (as Xanthogramma sp.) in Mangareva, Gambier Islands.Larvae of I. scutellaris attacked several aphids (Hemiptera: Aphididae) on different plants: Rhopalosiphum maidis (Fitch, 1856) and Sitobion avenae (Fabricius, 1775) on several species of Poaceae; Brevicoryne brassicae (Linnaeus, 1758) on Brassica oleracea L., (Brassicaceae); and Aphis gossypii Glover, 1877 on several plants belonging to Araceae, Asteraceae, Malvaceae and Myrtaceae.He also observed that the pupae of I. scutellaris were parasitized by a braconid wasp (Hymenoptera: Braconidae).

Remarks
Old records of this species are not easy to verify, as some authors used S. grandicornis and I. scutellaris as synonyms for a long time.

Differential diagnosis
Species with face entirely black (Fig. 4C), scutum and scutellum black (Fig. 4A), and metasternum greatly reduced, with deep posterior incision on each side (typical metasternum for this genus).Easy to distinguish from other species in French Polynesia by its overall black coloration.

Etymology
The specific epithet polynesiotes refers to the country where this species occurs, French Polynesia, and it means 'belonging to, pertaining to' in Greek.The specific epithet is to be treated as a noun in apposition.

thorax (Figs
Wing.Hyaline, entirely microtrichose, except cell c basally bare. Legs.Coxae dark; metatrochanters pale; femora black with basal and apical apices yellow; pro-and mesotibia yellow with a medial dark annulus; pro-and mesotarsomeres yellow; metatibia dark with basal 1/ 5 yellow; metatarsomeres dark.aBdomen.Parallel-sided, unmargined, entirely black.Terga 2-5 black pilose medially and yellow pilose laterally; sterna dark.maLe genitaLia.Enlarged (Fig. 4G); surstylus tapers to slender acute apex, curved towards dorsal part; superior lobes (postgonites) elongated, with rounded apex, with a spur-like process medially on the ventral margin and another spur-like process dorsally pointed anteriorly; hypandrium with two arms ending with two spur-like processes (one small and one larger); phallus one-segmented, distiphallus with two spur-like processes on each side, one pointed dorsally, the other pointed ventrally (Fig. 9D-F).

Female
Similar to male except for normal sexual dimorphism and as follows: frons shiny black, with pale pollinosity along eye margin on ventral ½ between antennae and anterior ocellus; postalar callus black; sternum 2 dark on anterior ½ and pale on posterior ½.

Geographical distribution
Species only known from Tahiti (French Polynesia).

Status in French Polynesia
Endemic.

Flowers visited
No records.

Geographical distribution
Pantropical: from southern USA south to Argentina, including the West Indies, introduced in the Afrotropical (including Madagascar, Mauritius, Reunion and Seychelles), Oriental and Australasian Regions, and Oceania.

Status in French Polynesia
Introduced; recorded from Marquesas Islands, Gambier Islands, Austral Islands and Society Islands.

Flowers visited
Ornidia obesa has been observed visiting two plant species, belonging to two families: Citharexylum spinosum L. (Verbenaceae) and Rauvolfia sachetiae Fosberg (Apocynaceae), the latter being endemic to French Polynesia.

Geographical distribution
From USA to Argentina, the West Indies and French Polynesia.

Status in French Polynesia
Introduced; recorded from Marquesas Islands, Gambier Islands and Society Islands.

Differential diagnosis
Simosyrphus grandicornis has often been confused with species of Ischiodon, especially I. scutellaris, as they are sympatric and the males of both species have large genitalia.The morphological characteristics to distinguish S. grandicornis are the broadly rounded basoflagellomere (subacute apically in I. scutellaris), black katepisternum (Fig. 5F) (with a dorsal yellow marking in I. scutellaris; Fig. 3G), simple male metatrochanter (Fig. 6B) (with a spine-like process in I. scutellaris, Fig. 6A) and black metafemur, yellow on the apical fourth (mostly yellow, black only subapically in I. scutellaris).

Material examined
Not collected or studied from French Polynesia, but Nishida (2008) reported it from the Society Islands.We believe that Nishida's records might be a misidentification of I. scutellaris.

Geographical distribution
Very common Australasian species, found throughout Oceania (from New Caledonia and Fiji to Hawaii), New Zealand and Australia (all states).Simosyrphus grandicornis is absent from the island of New Guinea.

Status in French Polynesia
Unknown.

Remarks
Among the material examined there were no individuals of this species from French Polynesia; thus, we believe that S. grandicornis is not present in the archipelago.Mengual (2015) stated that the dispersal of this species into many of the occupied Oceanic islands in its range was due to human activities, probably introduced by the early Polynesians who might have brought them in as larvae on fruits and plants they were transporting.Because it is plausible that S. grandicornis will reach the islands of French Polynesia in the near future, we have included this species in the identification key (indicated with an asterisk *).Hardy, 1964 Figs 4F, 5C-D, G-H Syritta aenigmatopatria Hardy, 1964: 409 (holotype: ♂, BPBM; type locality: USA, Hawaii, Oahu).

Differential diagnosis
Species with vein R4+5 straight, metafemur without basoventral patch of black setulae, arista bare, face carinate (Fig. 4F) and metafemur greatly enlarged with a ctenidium on the posteroventral half.It is similar to S. oceanica, but differs by having the spurious vein well sclerotized (Fig. 5C), as distinct as the neighboring R and M veins (spurious vein not sclerotized in S. oceanica, only formed by microtrichia; Fig. 5B, E), the ventral surface of the metatibia with an anteroventral carina forming a prominent lamina in males, less evident in females (metatibia without lamina in S. oceanica) and a distinct abdominal coloration, as stated in the key.

Geographical distribution
Widely distributed in the Oriental and Oceanic Regions: Vietnam, Sumatra, Java, Philippines, Palau Islands, Samoa, Mariana Islands, Hawaii and French Polynesia.

Status in French Polynesia
New species record; known from the Society Islands.

Flowers visited
No records.

Remarks
This is a new species record, since this taxon has not previously been recorded from French Polynesia.Macquart, 1855: 112 (misspelling).

Differential diagnosis
See above under S. aenigmatopatria.

Geographical distribution
South-central and central parts of the Pacific: Austral Islands over the Cook Islands, Marquesas Islands and Society Islands to the Hawaiian Islands (Lyneborg & Barkemeyer 2005).

Status in French Polynesia
Present; recorded from the Marquesas Islands, Society Islands, Austral Islands and Tuamotu Islands.

Remarks
We report here the first record of this species from the Tuamotu Islands.Lyneborg & Barkemeyer (2005) clarified the status of this species and the allopatric Syritta luteinervis de Meijere, 1908, and they explained the error of the New Zealand record for S. oceanica.(Hull, 1944) Key to Australasian and Oceanian species of Melanostoma  discovery of these new species in the mountainous areas of Tahiti prompts us to hypothesize that new endemics from French Polynesia are possible.With the newly obtained DNA barcodes, together with the barcodes obtained within the SymbioCode initiative, and with the taxonomy of Allograpta amphotera and A. nigripilosa to be corroborated, French Polynesia has become the first country in the world whose flower fly fauna is fully barcoded, a small but meritorious achievement.We hope that new material and new records will become available after this survey and that they help to resolve some taxonomic questions that remain open, such as the systematics of the genus Melanostoma in the Oceanian and Australasian Regions.