New macropterous leafhopper genera and species within the tribe Bonaspeiini from the Fynbos biome of South Africa (Insecta, Hemiptera, Auchenorrhyncha, Cicadellidae)

Two new genera of macropterous leafhoppers, tentatively included in the tribe Bonaspeiini Zahniser and Dietrich (2013) from the Fynbos biome of south-western parts of South Africa, are described. These are Retevolatus gen. nov. with type species R. flexiverpus sp. nov., R. semicurviverpus sp. nov. and R. subspiniverpus sp. nov. and Flavorubivolatus gen. nov. with type species F. glabriverpus sp. nov. and F. tensiverpus sp. nov. and F. curtiverpus sp. nov. Collection records and distribution modelling confirmed that species of both genera occur within a confined region of south-western parts of South Africa.


Introduction
This is another contribution to the study of leafhoppers (Insecta, Hemiptera, Auchenorrhyncha, Cicadellidae) in the Fynbos Biome of South Africa. Theron (1970Theron ( , 1972Theron ( , 1973Theron ( , 1974Theron ( , 1975Theron ( , 1978Theron ( , 1979Theron ( , 1980Theron ( , 1983Theron ( , 1984aTheron ( , 1984bTheron ( , 1986 and Davies (1987Davies ( , 1988aDavies and Geertsema 1998) described many species, but museum holdings and field work has revealed numerous new species. Many other new species in the South African Savanna biome with similar external appearance to these Fynbos leafhoppers requires ongoing examination of species and generic concepts. Zahniser and Dietrich (2013) placed 21 South African leafhopper genera, mostly described by Theron and Davies (op. sit.), into a new tribe Bonaspeiini. Recent additions were Bretega Stiller, 2016, Discolopeus Stiller, 2019and Geelus Stiller, 2020. The majority of Bonaspeiini do not fly, i.e. have short wings, at least with reduced hind wings, with a wide range of shrub associations. They are distributed in the Western Cape Province of South Africa, generally in the Fynbos Biome and adjacent biomes such as the Succulent Karoo, Nama-Karoo and Albany Thicket Biomes. Exceptional in the tribe is the macroptery of Renosteria Theron, 1984a that is associated with Dicerothamnus rhinocerotis (Asteraceae), a wide-spread, ruderal shrub in these biomes and threatened in agroecosystems (Mucina and Rutherford 2006). Additionally, Capeolix Linnavuori, 1961, Discolopeus, Geelus andKimbella Davies, 1988b have fully developed hind wings and tegmina as long as or longer than the abdomen. Male genitalia were and still are the best means of identification, with color and body shape generally not a reliable feature. The new genera described here are also macropterous, with Flavorubivolatus gen. nov. displaying a number of superficial similarities with Renosteria, but not associated with the same host plant. The other new genus, Retevolatus gen. nov., is distinct with overlapping distribution with the former new genus. Tribal placement of the new genera is not fully resolved due to the ambiguous definitions of Bonaspeiini and Athysanini in Zahniser and Dietrich (2013) which bear a number of similarities.

Material and methods
Numbers of examined specimens were 102 males and 74 females of Retevolatus and 51 males, 59 females and 14 nymphs of Flavorubivolatus. Dissections were made of KOH macerated specimens and examined under glycerine. Drawings were traced through camera lucida. Images were produced with Zeiss Axio Zoom V16 and Zeiss Axio Im-ager2 with Axiocam MRc camera. Image adjustments included crop, color brightness and contrast and unsharpen mask. Morphological terms mainly follow Blocker and Triplehorn (1985), Zahniser and Dietrich (2008) and Stiller 2016, Stiller 2019a, Stiller 2019b, chaetotaxy follows Rakitov (1997). Measurements included length from apex of crown to apex of tegmina, crown median length, crown length perpendicularly next to compound eye, pronotum length, head width across eyes, pronotum width, ocellus diameter and interocular distance between ocellus and compound eye. They were made with an eyepiece graticule, calibrated by a 2 mm scale on a microscope slide. Smallest measurements made at 70x magnification were converted to micrometers by multiplication with 14 or to millimeters by multiplication with 0.014. Measurements of the style were the sagittal length from the apex of the apophysis to the anterior margin of the medial arm and the greatest width across the base, between the medial and lateral arms, used as a length to width ratio. Connective measurements included width of the stem at its apex, greatest width across arms, length of stem from apex to base of arms (bottom of cavity between arms) and length of arms from apex of arms to bottom of cavity between arms. Measurements of the subgenital plate were the width at the base (confluence of subgenital plate and valve) and the length medially from the base to the apex. The crown angle, in part, at the apex of the head was calculated by trigonometry, specifically tangent, using crown median length and half head width across eyes, as adjacent and opposite sides of a right-angled triangle, and doubled for both sides. In the lateral view of the aedeagus the curvature was calculated as the angle subtended by its arc, using segment height and chord length of the arc in an online calculator (circular Male and female. Color. Base color ochraceous to stramineous. Crown with paired rhomboid or amorphous dark brown markings at apex (Figs 1A-I, 2A-C). Pronotum with light brown and dark brown markings, variable in size and shape (Figs 1A-I, 2A-C). Scutellum with light brown scutellar triangles, medially stramineous with dark brown paired markings and brown suture (Figs 1A-I, 2A-C). Tegmina uniformly reticulate and irrorate, costal cells and fourth apical cell sometimes with dark brown markings; vannal cells at margin variable, sometimes unmarked (e.g. Figs 1E,2A) or whitish (e.g. Figs 1A,H,2C). Legs ochraceous to stramineous; subapical brown, irregular band on pro-and meso-femur; metafemur with brown longitudinal line at apex; all tibia with areola of setae embrowned.
Wings. Tegmina with four apical cells; outer and central anteapical cells, inner anteapical cell and inner discal cells merged (Fig. 7L). Hind wing with vannal lobe large, greatest length 0.4 times greatest width of hind wing, four apical cells (Fig. 7M).
Metatibia. Setal formula at apex 2+2+1; apical pair subequal, shorter than long median seta, median setae with one very long and the other very narrow and about half as long. Basal seta subequal to long median seta.
Abdominal apodeme. Anterior abdominal apodeme with variable median lobes, rounded to angular, contiguous or separate. Lobes smaller in the female anterior apodeme.
Style. Elongate, with elongate apophysis; length to width ratio 3.1-3.7 (ratio length from apex of apophysis to apex of medial arm (sagittal plane) to greatest width across base). Width across preapical lobe slightly wider than width at base of apophysis (Figs 6I, 8D, 9E).
Etymology. Named for the net-like color pattern of the tegmina, in Latin, and verb in apposition. Net, rete, and flying, volatus, for functional wings. Gender masculine.
Discussion. This new genus is recognized by its color pattern, size and configuration of the male and female genitalia. The genus is tentatively placed in Bonaspeiini as it does not share all the features of the tribe. Similar color patterns to that of the new genus are found in the following genera of other tribes: Libengaia Linnavuori, 1969 (Opsiini), Dagama Distant (Fig. 2F, G), Allophleps Bergroth, 1910 (see Webb and Viraktamath 2017), Houtbayana decemnotata Linnavuori (Athysanini), an unknown and unplaced genus and species with elongate crown (Fig. 2D, E), undetermined Selenocephalini ( Fig. 2H-J) and even in some Discolopeus Stiller, 2019 species (Bonaspeiini) have a resemblance in marking at the apex of the crown. One male specimen of an unknown species of the new genus was examined, which suggests additional species of the genus are still to be found and described. This specimen has widely curvate style apophysis, oriented mediad, connective similar to that of R. semicurviverpus sp.nov. and R. subspiniverpus sp.nov., and the aedeagus very short, thickened, basal paired process and lateral denticulation. Externally the specimen resembles the three new species of Retevolatus described here. Key to Retevolatus gen. nov. males.
Style. Apophysis straight (Fig. 6I), length of apophysis 0.4 times greatest length through sagittal plane; apex of apophysis attaining apex of subgenital plate (Fig. 6H).  Discussion. The aedeagus of this species is strongly curved, with a right-angled anteriad apical process, that is variable in length, and broken in a number of examined specimens, specifically from Ceres and Darling. The length in most specimens depicted in Fig. 6A, with the longest found in one specimen from Garies, Fig. 6B. In this male from Garies the subgenital plate, pygofer process ( Fig. 7J, K), connective and style resemble those of the other specimens in this species. The pygofer process is apical, variable in shape and orientation, and damaged in some specimens. Plant distribution records for Salvia lanceolata in Fig. 18C.
Subgenital plate. Length to width 2.2-2.4 times longer than width at confluence of subgenital plate and valve. Macrosetae 5-10, variable between left and right plate (Fig. 8I). Apex of style near apex of subgenital plate (Fig. 8I).
Discussion. The aedeagus of this species is weakly curved, with an acutely angled dorsoanteriad, apical process that is constant in length, and whole in all examined specimens. The pygofer process is uniform in examined specimens, originating ventrally on the pygofer, acuminate and directed dorsoposteriad. The apophysis of the style of this species is curved mediad at its apex, and straight in R. subspiniverpus and R. flexiverpus. The deep, rounded notch of the sternite 7 of this species is distinct from the shallow or deep V-shaped notch of the other two species. The color and shape in dorsal and lateral views of this species resembles that of R. flexiverpus in Fig. 1A-I closely. Type specimen. Holotype male, pinned, with genitalia in a separate microvial. Original label: "South Africa, Nuwerus, 10-11-71, J.G. Theron" "SANC Pretoria Dbase CCDL27852" (blue paper).
Subgenital plate. Length to width 2.0-2.2 times longer than width at confluence of subgenital plate and valve. Macrosetae 9-15, variable between left and right plate. Apex of style extended half way into subgenital plate (Fig. 9H).
Distribution. Nuwerus, Fig. 18A, blue triangle. Discussion. The aedeagus of this species is strongly curved, with a subapical, rightangled dorsoanteriad process. Both or part of the apex and process of the shaft were damaged randomly in seven of 11 dissected of specimens, i.e. the process was broken at the base in four and broken at the apex in one specimen, and the shaft apex was partially or entirely damaged in two specimens. The pygofer process origin is ventrally and basally on the pygofer lobe, orientation posteriad, tubular and apex acuminate. The sternite 7 of this species (Fig. 9J) has a similar V-shaped, shallow notch to that of R. flexiverpus (Fig. 6L), but is narrowed distally, even more than in Fig. 6M. The valvifer 1 is more elongated than in the other species of Retevolatus.
Metatibia. Hind knee setal formula 2+2+1; apical pair sub equal, shorter than long median seta, median setae with one very long and the other very narrow and about half as long; basal seta longer than median and apical setae.
Metatarsi. Metatarsus 1 apex with five rounded platellae or four rounded and one acute platellum; metatarsus 2 apex with two lateral acute and two medial rounded platellae.
Abdominal apodemes. Anterior abdominal apodeme with desclerotized, rounded or angulate lobes, expanded anteriad and posteriad, not wider than width of apodeme, lobes contiguous in males, separate in females.
Connective. Stem and arms short, of similar proportions, greatest length 0.9-1.1 times greatest width; stem 0.5-0.7 times as wide as width across arms; arms widely separated, parallel to subparallel, U-shaped with stem; stem 0.8-1.1 times longer than length of arms (Figs 15G, 16F, G, 17E).
Discussion. The three species in this genus are recognized by their color and configuration of the aedeagus. Flavorubivolatus glabriverpus sp. nov. is yellowish (Fig. 10A, E) with a concavely curved aedeagal shaft without spines (Fig. 15A, C, E), and F. tensiverpus sp. nov. and F. curtiverpus sp. nov. are more reddish (Figs 10F-I, 11A-D, respectively) with the former with a straight shaft and spines about half as long as the shaft (Fig. 16A-D) and F. curtiverpus sp. nov. with the spines short, shorter than the diameter of the shaft (Fig. 17C, D). Low numbers of specimens of F. tensiverpus sp. nov. as well as lack of females of F. curtiverpus sp. nov. and their overlapping localities and collection dates with specimens of F. glabriverpus sp. nov. has detracted from the robust species hypothesis. However, no parasites or parasitoids were found in any of the examined specimens, although the sex-altering effect of Wolbachia (Negri et al. 2009;Saridaki and Bourtzis 2010) was considered, but the males of these species do not appear to be feminized. No parts were considered deformed or intermediate, or teneral. None of the specimens of the Flavorubivolatus gen. nov. collected by the author were on Dicerothamnus rhinocerotis, the only known associated plant of Renosteria.
This new genus resembles Renosteria in macroptery, similar triangular crown (i.e., species of Renosteria 71.9-78.7°, species of Flavorubivolatus gen. nov. 85.1-92.8°), and the corresponding longitudinal, rectangular sternite 7. It differs from Renosteria in color (species of Renosteria are yellow-green ( Fig. 13G-I) or brown (R. overbergia and R. cangica similar, as in Fig. 11F), compared to yellow or reddish in Flavorubivolatus) and in its slightly larger length from apex of vertex to apex of tegmina, i.e. 3.5-4.2 mm in Renosteria, compared to 2.6-3.8 mm in Flavorubivolatus. It also differs in the following aspects of the male genitalia.
The male pygofer lobe in Renosteria is obtusely or acutely triangular or rounded, with the process origin dorsomedial or subapical or apical, straight or sublinear, orientation ventrad or posteroventrad. In contrast the pygofer lobe in Flavorubivolatus gen. nov. is broadly rounded, with the process origin apical, curvate or straight, orientation ventrad or mediad.
The connective in Renosteria with ratio of greatest length to greatest width 1.09-1.31. In Flavorubivolatus gen. nov. the ratio of greatest length to greatest width similar, 0.95-1.22 in F. glabriverpus sp. nov., or slightly shorter, 0.82-1.00 in F. tensiverpus sp. nov., 0.93-1.04 in F. curtiverpus sp. nov. Other ratios such as width stem/width arm, length stem/length arm and length stem/width arm correspond.
The subgenital plate in Renosteria requires re-examination, but corresponds in arrangement of macrosetae, with length to width ratio 1.5-1.9 times longer than wide. In Flavorubivolatus gen. nov. it is 1.6-1.9 times longer than wide.
The style in Renosteria requires re-examination, with the anterior lateral lobe acutely angled to the sagittal plane through the apophysis and mid-section, and in Flavorubivolatus gen. nov. it is approximately in line with the sagittal line.
Valvifer 1. Dorsal and ventral margins rounded, asymmetric. Anterior and posterior margins narrowly rounded (Fig. 12J). Elongated or about as long as wide.
Etymology. Latin, adjective and noun in apposition, for hairless, smooth, bald, glabra, and the aedeagal shaft which is devoid of any spines or processes. Gender Masculine.
Discussion. Flavorubivolatus glabriverpus has the crown and pronotum yellowish, and the tegmina yellowish-red with this color pattern similar in males and females. Flavorubivolatus tensiverpus and F. curtiverpus are more reddish in color. Variation was found in male genitalia in apex of aedeagus, with the apex of the gonopore parallel or convergent (Fig. 15A, B). Measurement of subgenital plate variable, especially width due to imperfect horizontal orientation. The style was shorter than in other species of this genus. Connective with arms and stem of similar proportions. Dissections included 21 males and 13 females, thus 57% of males and 24% of females. No recently dissected males were parasitized, but one female abdomen contained a stylops and one female with a dryinid pupal case lateroventrally between the head and pronotum. Plant distribution records of some associated plants in Fig. 18E, F (Seriphium saxatilis, Myrovernix intricata respectively). The color pattern of Renosteria spadix corresponded to that of F. glabriverpus. Type specimen. Holotype male, pinned, with genitalia in a separate microvial. Original label: "South Africa, Sederberge, 21.xii.1976, J.G. Theron" "SANC Pretoria Dbase CCDL27824 (blue paper)". Paratypes
Aedeagus. Base or subbase of shaft ventrally with elongate paired process, about half as long as shaft, parallel and contiguous with shaft; shaft sublinear. Gonopore elongate, subapical dorsally to ventromedially. Dorsal apodeme and preatrium subequal in length (Fig. 16A-D).
Distribution. Cedarberg, Clanwilliam, Doltuin, Halfmanshof, Fig. 18B, red circle. Discussion. Flavorubivolatus tensiverpus is distinctly more red in the crown, pronotum and tegmina than yellowish color of F. glabriverpus. Measurements show that the crown length medially is slightly shorter, i.e. 0.4 mm and in the other species 0.4-0.5 mm. Genitalia in the male with aedeagal shaft linear to sublinear, position and orientation of basal processes (Fig. 16A-C), especially in Halfmanshof specimen, where the processes are more basal, slightly shorter and the apex of the shaft slightly wider than in the other species of this genus (Fig. 16B, compare Fig. 16A, C, Clanwilliam). Arms of the connective appear asymmetrical, but this is due to orientation during examination (Fig. 16F, G). The connective arms and medial arms of the styles are angled dorsad. Dissections included 8 males and 4 females and signs of parasites absent, i.e. no larva inside abdomen or pupal case. Plant records of Phylica oleifolia in Fig. 18D. A record of Renosteria waverena from Valve. Posterior margins broadly rounded (Fig. 17H). Aedeagus. Base of shaft ventrally with short paired acuminate process; length of process less than width of shaft. Shaft straight (Fig. 17B-D). Gonopore elongate, subapical dorsally to ventromedially (Fig. 17A, C). Shaft with striate microsculpture (Fig. 17A). Dorsal apodeme reduced, preatrium elongate (Fig. 17D).
Distribution. Driehoek farm, Dwarsrivier, Sederberge, Fig. 18B, yellow triangle. Discussion. The reddish color (Fig. 11B-D) is similar to the color of F. tensiverpus (Figs 11F-I, 12A) which distinguish these two species from F. glabriverpus, that is yellowish (Fig. 10A, E). Pygofer lobe processes variable in apical serration, i.e. length and number of teeth (Fig. 17K, L). The truncated, blunt apex of the subgenital plate and the elongated, transverse pygofer lobe process can clearly be distinguished on whole specimens (Fig. 11D), and that allow distinction between this species and F. glabriverpus and F. tensiverpus. The single female listed in material examined, (image in Fig. 10B) is tentatively included here. This species consists of a short series of males with overlapping records in one or other of the two species above. Signs of parasites or feminized specimens were absent, with all genitalia considered invariable, functional and distinct from F. glabriverpus and F. tensiverpus. All males were dissected. Records of an associated plant, Selago pustulosa, in Fig. 18G.

Distribution models
Distributions were modelled with the species from the most localities (i.e. R. flexiverpus) as well as most numerous species and all species merged, with the latter model for Retevolatus in Fig. 18A and for Flavorubivolatus in Fig. 18B, the former with a wider potential distribution. The first four Worldclim bioclimatic variables with the highest contribution to these models differed slightly. In Retevolatus they were bio19, 18,8 and 15,and in Flavorubivolatus bio8,15,18 and 19. Bio8 is mean temperature in wettest quarter (winter), bio15 is precipitation seasonality, bio18 is precipitation in warmest quarter (summer) and bio19 is precipitation in coldest quarter (winter). Despite the wide distribution of some of the associated plants ( Fig. 18C-G), the range of these leafhoppers appears restricted to a small area. Personal observation of species on plants resembling Seriphium saxatilis in the northern parts of South Africa have not produced specimens of Flavorubivolatus. Extensive work on the leafhoppers associated with Dicerothamnus rhinocerotis by Theron has not shown that Flavorubivolatus was associated with this plant.

Conclusions
Although the new genera described here are placed in Bonaspeiini this placement is not fully resolved due to the ambiguous definitions of Bonaspeiini and Athysanini in Zahniser and Dietrich (2013) which bear a number of similarities. The attempt here and previously (Stiller 2019a(Stiller , b, 2020 was to better quantify dimension of parts such as the anal tube, connective and style. At present three configurations of wings are recog-nized in genera included in Bonaspeiini, i.e. tegmina covering abdominal segments 2-3, hind wing reduced (Basutoia, Bretega, Tzitzikamaia); tegmina covering up to abdominal segments 7-9, hind wing reduced, narrower or shorter than tegmina (Bloemia, Bonaspeia, Caffrolix, Cerus, Colistra, Curvostylus, Gcaleka, Hadroca, Johanus, Kaapia, Megaulon, Proekes, Proekoides, Refrolix, Salsocolila, Salsolibia); tegmina extended beyond abdomen, hind wing as long as tegmina and with functional jugal lobe (Capeolix, Discolopeus, Geelus, Flavorubivolatus, Kimbella, Renosteria and Retevolatus); at present alary polymorphism has not been observed in these genera. An additional character which may be useful in grouping species is the presence or absence and position of the pygofer process; in some groups the process is marginal or medial with further subcategories for both such as dorsal, apical and ventral. Properties of the pygofer process such as denticulation, curvature and orientation appear unsuited for generic categories. The anal tube, connective, style and aedeagus require further investigation towards contribution of tribal placement.