Revision of the asychis species group of Aphelinus ( Hymenoptera : Aphelinidae )

Aphelinus (Hymenoptera: Aphelinidae) is a genus of parasitoid wasps that has a long history of use in biological control programs against aphids. Past research shows that species delimitation within Aphelinus is greatly complicated by lack of comprehensive literature and the existence of cryptic species complexes. One of these complexes is the Aphelinus asychis species group. Through the development of a morphological character set, a revision of the Aphelinus asychis species group was conducted. Two new species, Aphelinus sinensis sp. n., and Aphelinus kazakhstanensis sp. n., are described, and the two existing valid species within the asychis group, Aphelinus asychis and Aphelinus semiflavus are redescribed and lectotypes are designated for Aphelinus semiflavus and Aphelinus brevipennis (a junior synonym of A. semiflavus). We also provide a key for identifying species in the asychis group.


Introduction
Aphelinus Dalman, 1820 is a genus of chalcidoid wasps in the family Aphelinidae, subfamily Aphelininae (see Hayat 1972).All species of Aphelinus are endoparasitoids of aphids, and the genus has a long association with biological control research against aphid pests (Arce Gomez and Rumiatto 1989, Cate et al. 1973, Clausen 1978, Elliott

Diagnosis of asychis species group
Following Hayat (1998), we consider a submarginal vein with only two setae to be diagnostic for the A. asychis group.In females, F1 and F2 are subquadrate, and F3 is 1.2-2.0×longer than wide.In males, F1 and F2 are wider than long, and F3 is >3.0× longer than wide.Brachyptery is also common in this group, particularly in males.There is marked sexual dimorphism, particularly in antennal proportions and coloration, which is not found in other Aphelinus species.

Material and methods
Specimens used in this study were killed in 95% ethanol and stored in freezers.Most were then critical-point-dried using a Samdri 790 CPD unit.Critical-point-dried specimens were card mounted with Franklin International's water soluble Titebond Liquid Hide Glue.Selected specimens were slide-mounted following Noyes ' (1982) protocol.All card-mounted and slide-mounted specimens were assigned individual accession numbers (e.g., TAMU-ENTO X0852885, USNM ENT 4532898, etc.).Label data for type specimens are reported verbatim, where | signifies a new line on a label and || separates different labels.The following acronyms for museum collections are used, followed in some cases by the actual acronym used with specimen accession numbers: ANIC, Australian National Insect Collection (Canberra); BMNH [BMNH(E)], Natural History Museum (London); CNC (CNC HYMEN), Canadian National Collection of Insects and Nematodes (Ottawa); EMEC, Essig Museum Entomology, University of California (Berkeley); NHW, Naturhistorisches Museum (Vienna); TAMU (TAMU-ENTO), Texas A&M University Insect Collection (College Station, TX); UCR (UCRC ENT), University of California (Riverside, CA); USNM (USNM ENT) National Museum of Natural History (Washington, D.C.).Complete label data for any or all material examined are available upon request to the authors.
Images for figures were acquired using digital imaging and image-stacking.Specimens photographed for coloration were removed from alcohol storage, placed on a layer of water-based, water-soluble jelly in a small watch glass, submerged in alcohol, and photographed using a Leica M205 FA stereomicroscope and Leica Applications Suite software (ver.4.5) as were card-mounted or point-mounted specimens.Slide-mounted specimens were photographed using an Olympus BH2 microscope with DIC illumination and Image-Pro Plus software (ver.7.0).Zerene Stacker (http://zerenesystems.com)was used for all image stacking.Adobe Photoshop CS6, Adobe Lightroom 5, and Adobe InDesign CS6 were used for final modifications to images and layout of plates.All images were deposited in mx, a web-based content management database system.Morphological codings were conducted in mx.The mx system is open source, with further documentation available at http://mx.phenomix.org.

Morphology
See Appendix Table 1 for a list of the morphological terms used in this study, followed by a definition and a URI (uniform resource identifier) that links to more information on that character in the Hymenoptera Anatomy Ontology (Yoder et al. 2010) database.

Morphology measurements
Measurements from slide mounts were taken using an eyepiece reticle in a Zeiss standard 16 microscope.Measurements from card mounts were taken using an eyepiece reticle in a Leica MZ16 microscope.Raw measurements are only reported for body length, which is followed by the range and the number of specimens measured.The remaining measurements are reported as ratios.
Head.The length of the head was measured from the anterior to the posterior margin in dorsal view (Fig. 1a, hl).The frontovertex length was measured in dorsal view from the dorsal margin of the scrobal impression to the occiput (Fig. 1a, fvl).Head and frontovertex widths were both measured at their widest points.The posterior ocellar diameter (Fig. 1a, ow), distance from posterior ocelli to eye margin, (Fig. 1a, ool), distance between ocelli (Fig. 1a, pol) and distance from posterior ocelli to occipital margin (Fig. 1a, ocp) were measured as illustrated.The widths of each antennal segment (scape, pedicel, F1, F2, F3, and club) were measured at their widest points.The lengths of each antennal segment were measured from proximal to distal end.The scape of males in the asychis group bears a longitudinal row of pores on a convex ridge on the anterior (ventral) surface (Fig. 2), that appear to be openings for internal glands.
Meso/Metasoma length.Meso/metasoma length of specimens was measured from the anterior margin of the pronotum to the apex of the epiproct using slidemounted specimens.The lengths of the mesosoma, the mid lobe of mesoscutum, and the scutellum were measured from their anterior to posterior margins along the mid line and widths were measured at their widest points.
Wings.Fore wing measurements are shown in Figure 1b; hind wing measurements follow those of the fore wing.
Male Genitalia and Ovipositor.The length of the phallobase was measured from the anterior margin of the genital capsule to the posterior end of the digiti (Fig. 1c, phl).The width of the phallobase was measured at its widest point (Fig 1c,phl).The length of the digitus was measured between its most anterior to most posterior points, and its width was measured at its widest point (Fig. 1c, dig).Ovipositor length was measured as illustrated in Figure 1d.

Results
Key to species in the asychis group of Aphelinus, male or female specimens.
Mesosoma (Figs. 3d,3f and 4g).Mid lobe of mesoscutum length 0.7-0.8×mid lobe width with two pairs of long setae (one pair lateral and one pair posterior) and 29 short setae; side lobes of mesoscutum each with one pair of long setae and one pair of short setae; scutellum with two pairs of long setae (one pair anterior and one pair posterior); mesotibial spur length 0.6× mesobasitarsus length, metatibial spur length 0.6-0.8×metabasitarsus length.
Fore wing (Fig. 4d).Length 2.8× fore wing width, longest marginal seta 0.2× fore wing width; costal cell length 0.7× length of marginal vein, with one line of 5 setae on ventral surface and 1 dorsal setae in apical quarter; marginal vein with two rows of 10 large dorsal setae, one row of 5 small dorsal setae, and one row of 7 ventral setae; interspace between basal cell and linea calva with 25 setae arranged in three complete line and one incomplete line; linea calva closed with three setae at its posterior end, setae bordering linea calva proximally are arranged uniformly and evenly to posterior margin of wing.Hind wing (Fig. 4e).Length 4.4× hind wing width, longest marginal seta 0.5× hind wing width.
Distribution.Old World and some New World populations, discussed below.Discussion.The collections from France, Italy, Morocco, and Spain correspond to those discussed in Kazmer et al. (1996).The leg coloration of of A. asychis specimens as described in the diagnosis is unique among species in the asychis group.We are treating collections from Spain, Morocco, and Italy as conspecific, noting that the metafemora are dark brown [not yellow].We are also treating the collection from England, Sussex, as conspecific, noting that the profemora and mesofemora are yellow [not dark brown with apex yellow or pale].We are treating the population from Hoda, Hawaii, and Bangalore, India, as conspecific, noting that the protibia are brown [not yellow].Although the male antennal club in asychis is usually longer than in other species, we note that it is of comparable length (4.0-4.5 L/W) in the series from Montpellier and Antibes, France.
Numerous populations of A. asychis are present in North America.The population from Randall Co., Texas, is presumed to have originated from biological control program releases of A. asychis against the Russian wheat aphid in that area.The population from Wayne Co., Missouri, was collected from a Malaise trap, and one specimen from Florida was collected by a flight interception trap, both in the late 1980's.The populations from Stillwater, Oklahoma, originated from a lab culture of A. asychis for Schizaphis graminum biological control.The two specimens collected from Manhattan, Kansas, from spotted alfalfa aphid in a greenhouse probably result from agricultural research at Kansas State University.The series from Alameda Co., California, 1962, is from the former UC Insectary, Oxford Tract, and was possibly being researched as a candidate for biological control of yellow clover aphid or other pest aphids.
However, in a few cases, A. asychis was found in North America before any documented biological control releases.An A. asychis specimen was collected in Florida in 1952, which does not have any associated host data.Two specimens from Maryland, one from Myzus persicae and one from strawberry aphids were collected in 1962 and 1950, respectively.One specimen from Maine in 1958 was collected from a Capitophorus aphid mummy.Two specimens from Colorado were collected from Myzus persicae mummies, one in 1940 and the other in 1988.These records are especially interesting because they are geographically close to the type locality and host of A. semiflavus.The specimens from Minnesota, Ohio, and California (UCRC ENT 75429 and 75431) did not have collecting date or host information.
The lectotype and paralectotype females of affinis Förster were examined (NHW, Vienna).They are point-mounted specimens in good condition.We concur with Graham (1976) that affinis Förster is a junior synonym of asychis Walker.The lectotype of brachyptera Kurdjumov (NHW, Vienna) was examined.It is a brachypterous female mounted on a minuten pin through the mesosoma.We concur with Graham (1976) that brachyptera Kurdjumov is a junior synonym of asychis Walker.The lectotype female of brevicalcar Thomson (LUZN, Sweden) was examined.It is a card-mounted female in reasonably good condition.We concur with Graham (1976) that brevicalcar Thomson is a junior synonym of asychis Walker.
The type material of dubia Kurdjumov (NHW, Vienna) which consists of a female lectotype and five female paralectotypes, mounted together on a small wooden block on minuten pins, and a second pin with a wooden block bearing five additional female paralectotypes, was examined.A red dot next to one pin identifies the lectotype.The specimens are dirty and in poor condition and the minuten pins are rusting.Although we note some color variation in the metasoma of these specimens, we concur with Graham (1976) that dubia Kurdjumov is a junior synonym of asychis Walker.The lectotype female and paralectotype females of euthria Walker (BMNH, London) were examined.The lectotype and one paralectotype are mounted on the same card and are in poor condition (lectotype is missing all of the metasoma).Three additional paralectotypes individually card mounted are in reasonably good condition.We concur with Graham (1976) that euthria Walker is a junior synonym of asychis Walker.Diagnosis.Females and males.Legs with procoxa brown (Fig. 5d) [not yellow (Fig. 9d)], profemur and mesofemur light brown with apex yellow or pale (Fig. 5d) [not entirely yellow (Figs.7d, 9d) or dark brown with apex yellow or pale (Fig 3d)], mesotibia yellow (Fig. 5d) [not dark brown with apex yellow or pale (Fig. 3d)], metatibia light brown with apex yellow (Fig. 5d) [not entirely yellow (Fig. 9d), or dark brown with apex yellow or pale (Fig. 3d)].Description.Female (Figs. 5b,5d,5f and 6b,.Color (Fig. 5b, 5d, 5f ).Head and mesosoma dark brown; radicle and basal portion of scape yellow/white, apical portion of scape and pedicel brown, and F1, F2, F3, and club yellow, tip of club dusky; legs with mesocoxa and metacoxa brown, metafemur yellow, protibia yellow; metasoma yellow from base to apex, lateral margins of metasoma darker than mesal area except in basal quarter.Body length.0.7-0.9mm (n=3; slide mounts) (Holotype 0.7 mm).Head (Figs. 5b and 6b).Width 1.1-1.2×head length in anterior view; frontovertex width 0.5-0.6×head width and 1.9-2.6×frontovertex length; posterior ocelli diameter 0.6-0.8×posterior ocelli to eye margin distance and 1.5× the posterior ocelli to occipital margin distance; antenna as in Figure 6b with scape length 5.8-7.3×scape width, pedicel length 1.8-2.2×pedicel width, F1 and F2 subquadrate, length of both 1.0× width, F3 length 1.4-1.6×F3 width, and club length 2.9-3.7×club width and 2.8-3.1×F3 width, club with 8 longitudinal sensilla.

Aphelinus kazakhstanensis
Mesosoma (Figs. 5d, 5f and 6g).Mid lobe of mesoscutum length 0.8× mid lobe width, and with two pairs of long setae (one pair lateral and one pair posterior) and 34-35 short setae; side lobes of mesoscutum each with one pair of long setae and one pair of short setae; scutellum with two pairs of long setae (one pair anterior and one pair posterior); mesotibial spur length 0.7-0.8×mesobasitarsus length, metatibial spur length 0.4-0.6×metabasitarsus length.
Fore wing (Fig. 6d).Length 2.3-2.4×fore wing width, longest marginal seta 0.1-0.2×fore wing width; costal cell 0.6-0.7×length of marginal vein, with one line of 6-7 setae on ventral surface and 1-2 dorsal setae in apical quarter; marginal vein with two rows of 14-15 large dorsal setae, one row of 7-8 small dorsal setae, and one row of 8-10 ventral setae; interspace between basal cell and linea calva with 30-33 setae arranged in three complete lines and two incomplete lines; linea calva closed with 2-3 setae at its posterior end, setae bordering linea calva proximally arranged uniformly and evenly to posterior margin of wing.
Hosts.The original material was collected from Diuraphis noxia in the field in Dmitrievka, Kazakhstan.In lab culture, Diuraphis noxia on wheat was used as the host.
Distribution.The species is known only from type material from Dmitrievka, Kazakhstan.
Discussion.The collections from Kazakhstan, Dmitrievka correspond exactly to those discussed in Kazmer et al. (1996).The most notable distinction between A. kazakhstanensis and all other A. asychis group species is the yellow mesotibia, combined with the profemur, mesofemur, and metatibia light brown at base and yellow at apex.Aphelinus asychis specimens have mesotibia, metatibia, profemur, and mesofemur dark brown at base and yellow at apex.Aphelinus semiflavus has mesotibia, profemur, and mesofemur yellow, and metatibia brown at base and yellow at apex.Aphelinus sinensis specimens have mesotibia, metatibia, profemur, and mesofemur all yellow.
Mesosoma (Figs. 7d, 7f and 8g).Mid lobe of mesoscutum length 0.7-0.8×mid lobe width with two pairs of long setae (one pair lateral and one pair posterior) and 30-31 short setae; side lobes of mesoscutum each with one pair of long setae and one pair of short setae; scutellum with two pairs of long setae (one pair anterior and one pair posterior); mesotibial spur length 0.6-0.7×mesobasitarsus length, metatibial spur length 0.4-0.5×metabasitarsus length.
Fore wing (Fig. 8d).Length 2.3-2.8×fore wing width, longest marginal seta 0.1× fore wing width; costal cell length 0.6-0.7×length of marginal vein, with one line of 10-12 setae on ventral surface and 1-2 dorsal setae in apical quarter; submarginal vein with two setae; marginal vein with two rows of 14-17 large dorsal setae, one row of 7-13 small dorsal setae, and one row of 7-10 ventral setae; interspace between basal cell and linea calva with 31-40 setae arranged in three complete line and one to two incomplete lines; linea calva closed with 3 setae at its posterior end, setae bordering linea calva proximally are arranged uniformly and evenly to posterior margin of wing.
Discussion.There has been confusion in the past about whether or not A. asychis and A. semiflavus are separate species.Based on the material examined, leg coloration patterns clearly differ between them.Regarding the New World populations examined, we are treating most North American populations as semiflavus, noting that in the Mexico and Winnipeg, Canada, populations, the metatibia are yellow/brown [not dark brown at base with apex pale].
We consider the Bangalore, India, and Dezful, Iran populations as sp.nr.semiflavus, noting that the India population has mid tibia and mid femora with brown [not yellow] and the Dezful, Iran, population has legs like semiflavus except one female with brown [not yellow] on mid tibia and mid femora.There is one specimen from Virginia that we are treating as possibly semiflavus, noting that the fore femora and fore tibia are brown [not yellow].
We have examined the lectotype female and paralectotype male of brevipennis Girault and agree with the conclusion of Gahan (1924) that it is a junior synonym of semiflavus Howard.Although Ferrière (1965), Nikol'skaya and Yasnosh (1966), and Yasnosh (1978) treated semiflavus as a junior synonym of asychis Walker, for reasons discussed above we consider it to be a distinct and valid species.
Future work should use reciprocal crosses and perhaps molecular data to determine whether populations of A. semiflavus in Old World vs. New World are in fact one species, are races of one species with different host ranges, or are two distinct cryptic species.Diagnosis.Females and males.Legs with procoxa yellow (Fig. 9d) [not brown (Figs. 3d,5d,7d)], profemur, mesofemur, mesotibia, and metatibia yellow (Fig. 9d) [not brown at base with apex yellow or pale (Figs. 3d,5d,7d)].
Mesosoma (Figs. 9d, 9f and 10g).Mid lobe of mesoscutum length 0.7× mid lobe width, with two pairs of long setae (one pair lateral and one pair posterior) and 31-33 short setae; side lobes of mesoscutum each with one pair of long setae and one pair of short setae; scutellum with two pairs of long setae (one pair anterior and one pair posterior); mesotibial spur length 0.7-0.80×mesobasitarsus length; metatibial spur length 0.5× metabasitarsus length.
Fore wing (Fig. 10d).Length 2.5-2.8×fore wing width, longest marginal seta 0.1-0.2×fore wing width; costal cell 0.8× length of marginal vein, with one line of 6-7 setae on ventral surface and 1-2 dorsal setae in apical quarter; submarginal vein with two setae; marginal vein with two rows of 12-18 large dorsal setae, one row of 7-11 small dorsal setae, and one row of 7-10 ventral setae; interspace between basal cell and linea calva with 18-33 setae arranged in two complete lines and two incomplete lines; linea calva closed with 2-3 setae at its posterior end, setae bordering linea calva proximally are arranged uniformly and evenly to posterior margin of wing.
Hosts.The original material was collected from Diuraphis tritici (Gillette 1911) (the junior synonym Diuraphis agropyronophaga is given on the holotype label) in the field in China.In lab culture, Diuraphis noxia on wheat was used as the host.
Distribution.Northern China and Japan.Discussion.The collections from China, Pingluo, Ningxia correspond exactly to those discussed in Kazmer et al. (1996).The most notable traits of the type specimens of A. sinensis from Pingluo, China, that distinguish them from other A. asychis group specimens examined, are the presence of yellow procoxae, yellow femora, and yellow tibiae.In all other asychis group species, all coxae are brown, and femora and tibiae are patterned with brown.The series from Harbin, China, exhibits the same leg-pattern coloration as the type series.We are treating the Harbin series as conspecific, noting that the head and metasoma are much darker, almost black, and the antennal club is darker at the apex than in material from Pingluo.There is one specimen from the Honshu, Japan, series that has the same leg coloration patterns as the type series from Pingluo, however, the other specimens in this series resemble A. asychis.

Other Potential New Species
There are two additional potential new species, but there is not enough material to describe them.One potential new species is from Harrow, Canada, and has leg colora-tion like asychis, but includes a brown metafemur [not yellow].There are two other Canadian series with one specimen in each series that also has this leg coloration.The other potential new species is from Dorking, England, and is represented by a single specimen that has leg coloration with all segments very dark brown.