A new species of Cavichiana from southeastern Brazil, with a key to the species of the genus and notes on the distribution of C. bromelicola (Insecta: Hemiptera: Cicadellidae)

ABSTRACT The sharpshooter genus Cavichiana Mejdalani et al., 2014 included so far only two species of about 6-8 mm in length: C. bromelicola Mejdalani et al., 2014 (type species) and C. alpina Quintas et al., 2020. Both are bromeliad specialists from southeastern and southern Brazil. Here a third species, C. caelivittata sp. nov., is described and illustrated based on males and females from the Serra dos Órgãos massif in state of Rio de Janeiro, southeastern Brazil. The new bromeliad specialist can be readily distinguished from the previously known species by the dorsum with a pair of broad whitish-blue stripes, each one extending from antennal ledge over lateroposterior area of crown, lateral portion of pronotum, basal portion of clavus, and forming an elongate arc on corium up to outer margin of first apical cell; the aedeagus bears a dorsoapical digitiform lobe directed posterad. A key to the three species, based on the coloration and morphology of the aedeagus, is provided. In addition, the distribution of C. bromelicola is discussed; it is suggested that the trade of bromeliads as ornamental plants is causing the spread of this species to various urban and suburban areas of southeastern Brazil.


INTRODUCTION
The sharpshooter genus Cavichiana Mejdalani et al., 2014 includes two beautifully colored species of about 6-8 mm in length: C. bromelicola Mejdalani et al., 2014 (type species) and C. alpina Quintas et al., 2020.Specimens of these species have so far been collected exclusively on bromeliads from southeastern and southern Brazil, where adults and nymphs usually hide deeply in the rosettes, getting protec tion, food (xylem sap), and moisture.This record is relevant because previously no other leafhopper species had been observed in its native habitat using bromeliads as a host plant (Mejdalani et al. 2014). However, Felix et al. (2020) recently reported bromeliads as the host plants of Portanus restingalis Felix & Mejdalani, 2016 (Aphrodinae, Portanini).
Cavichiana, which is morphologically similar to Aurigoniella Takiya et al., 2001, Juliaca Melichar, 1926, Microgoniella Melichar, 1926, and Platygonia Melichar, 1925(Mejdalani et al. 2014), can be recognized by the following features based on Quintas et al. (2020) and Quintas and Mejdalani (2021): (1) head deltoid and strongly produced anteriorly; (2) ocelli located distinctly anterad of anterior angles of compound eyes; (3) aedeagus tubular and elongate; and (4) paraphyses with both stalk and rami elongate; rami slender, each one with a basidorsal projection.The distinct red color of the compound eyes is perhaps another diagnostic feature of Cavi chiana.The use of bromeliads as host plants is appa rently an exclusive association of this genus within the Cicadellinae.In addition to morphological information on the two previously known species (Mejdalani et al. 2014, Quintas et al. 2020), a study has been recently conducted on the morphology (scanning electron microscopy) and biology of nymphs of C. bromelicola (Quintas and Mejdalani 2021).
In this paper, a new Cavichiana species is described and illustrated based on males and females from the Serra dos Órgãos massif, state of Rio de Janeiro, southeastern Bra zil.A key to species based on the coloration and aedeagus morphology is provided.We also discuss the distribution of C. bromelicola; our field studies and communications from colleagues indicate that the common use of bromeliads as ornamentals in public and private gardens is apparently related with the spread of this species to various localities in southeastern Brazil.
Male holotype.Body (Figs 1, 2), in lateral view, distinct ly flattened dorsoventrally in comparison with most sharp shooters.Head (Fig. 2), in dorsal view, strongly produced anteriorly; median length of crown approximately equal to interocular width and 6/10 of transocular width; anterior margin narrowly rounded; with distinct and elevated carina at transition from crown to face; ocelli located distinctly before line between anterior eye angles, each closer to median line of crown than to adjacent eye angle; surface of crown distinctly depressed; eyes, in dorsal view, oblong; frontogenal suture extending onto crown and attaining ocellus.Antennal ledge, in dorsal view, not protuberant, its outer margin slightly rounded and carinate; in lateral view, with anterior margin approximately rectilinear and oblique.Face without pubescence; frons flattened medially, muscle impressions distinct; epistomal suture obsolete medially; clypeus flattened, its contour continuing profile of frons, apex convex.
Thorax (Fig. 2) with pronotal width approximate ly equal to transocular width of head; lateral pronotal margins slightly convergent anteriorly; posterior margin concave, sinuous; central portion of disk of pronotum with transverse rugae, except anteriorly; dorsolateral carina complete, rectilinear, declivous anterad.Mesonotum with scutellum not distinctly striate.Forewing without distinct apical membrane; veins indistinct, not elevated; with four apical cells, base of fourth more proximal than base of third; three anteapical cells, their bases obscure.Hind wing with vein R2+3 incomplete.Hind leg with femoral setal formula 2.1.1;length of first tarsomere greater than combined length of second and third, with two parallel rows of distinct setae on plantar surface.
Male terminalia with pygofer (Fig. 3), in lateral view, well produced posteriorly; lacking processes; posterior margin broadly rounded; ventral margin with basal emar gination; macrosetae distributed mostly on posterior half of disk.Valve (Fig. 4), in ventral view, subrectangular; anterior margin distinctly emarginate.Subgenital plate (Fig. 4), in ventral view, subtriangular, distinctly narrowed on basal half, outer margin rounded at base; few uniseriate macrosetae distributed mostly on basal half; plates separated medially from each other throughout their length; in lateral view, not extending as far posteriorly as pygofer.Connective (Fig. 5) poorly sclerotized; in dorsal view, Yshaped; stalk slender.Style (Figs 5,6), in dorsal view, elongate, with distinct outer preapical projection; outer preapical area with few setae; apical portion digitiform, oblique, directed outwards.Aedeagus (Figs 7, 8) symmetrical; shaft, in lateral view, expanded apically, with dorsoapical digitiform lobe directed posterad, without apical crown of spines; gonod  2).Dorsum dark brown to black with pair of broad conspicuous stripes, each one extending from antennal ledge over lateroposterior area of crown, lateral portion of pronotum, basal portion of clavus, and forming elongate arc on corium up to outer margin of first apical cell; this pair of stripes mostly whitishblue except on crown and anterior portion of pronotum where dirty white hue pre dominates; inner margin of stripe with small emargination adjacent to ocellus.Crown with two pairs of small dirty white spots, one at apex and another at inner ocellar margins; eyes distinctly red.Forewing with transcommissural, elongate whitishblue marking on clavus extending from median portion to apex; corium with subtriangular whitishblue spot at apical portion, extending from costal margin to base of third apical cell.Face, lateral lobe of pronotum, and lateral and ventral areas of mesothorax mostly black; superior portion of frons with median dirty white spot; gena with bluishwhite spot adjacent to lorum.Labium, labrum, and legs mostly brownishyellow.Abdomen dark brown ventrally, posterior margin of sternites and laterotergites tinged with brownishyellow; pygofer and subgenital plates dark brown, macrosetae yellowishbrown; posterior and ventral margins of pygofer lobes brownishyellow.
Female paratypes.External morphology and color similar to the male holotype.Terminalia with sternite VII (Figs 10,11), in ventral view, broadly and deeply emargi nate posteriorly."Internal" sternite VIII, in dorsal view, without distinct sclerites.Pygofer (Figs 10,11), in lateral view, moderately produced posteriorly; posterior margin narrowly rounded; macrosetae distributed mostly on pos terior half and extending anteriorly along ventral margin (those located along ventral margin larger than other ones).Valvifer I, in lateral view, subrectangular.Valvula I (Fig. 12), in ventral view, with basal portion slightly expanded and directed outward; in lateral view, blade with apex (Fig. 13) acute, spiniform; dorsal sculptured area extending from basal portion to apex, formed apically mostly by scalelike processes arranged in oblique lines and basally by more linear processes; ventral sculptured area restricted to apical portion, formed mostly by scalelike processes; ventral in terlocking device elongate, restricted to basiventral area of blade.Valvula II (Fig. 14), in lateral view, expanded beyond basal curvature, dorsal margin convex; apex obtuse; ventral preapical prominence distinct; dorsal margin with about 18 continuous teeth, most of them subtriangular, elongate and flat (Fig. 15), basal teeth smaller and irregular; denticles dis tributed on teeth and on dorsal and ventral apical portions of blade, except on apex (dorsal dentate apical area longer than ventral area); valvula with ducts extending towards teeth and apex.Gonoplac (Fig. 16) extending approximately as far posteriorly as pygofer; in lateral view, with basal half narrow and apical half distinctly expanded; apex obtuse; surface with tiny tegumentary processes (denticuli) on apex and extending anteriorly along ventral margin; few setae also present.Pygofer lobes dark brown with posterior and ventral margins usually brownishyellow, macrosetae yellowishbrown; gonoplacs dark brown to black.
Type material.Atlantic Forest, southeastern Bra zil, state of Rio de Janeiro.Etymology.The specific name (declinable adjective, feminine), caelivittata, of Latin derivation, refers to the conspicuous dorsal pair of whitishblue longitudinal stripes.
Taxonomic notes.Cavichiana caelivittata sp.nov.can be readily distinguished from C. bromelicola and C. alpina by the coloration (Figs 1, 2) and aedeagal morphology (specific diagnostic character states are provided in the key below).The depressed crown surface and the distinct carina loca ted at the transition crownface (Fig. 2) are also diagnostic features of the new species.The large dorsal expansion at the bases of the paraphyses rami (Fig. 9) of the new species is located in the same area of the dentiform projection of the two previously known species (see Mejdalani et al. 2014, Quintas et al. 2020), suggesting that the former is homologous to the latter; the presence of this structure is a diagnostic feature of Cavichiana.Therefore, we conclude that the new taxon has the four diagnostic features of the genus proposed by Quintas et al. (2020) and listed above in the introduction.

Notes on the distribution of Cavichiana bromelicola
The type localities of the three known Cavichiana spe cies are in the Atlantic Forest biome in southeastern Brazil.Cavichiana bromelicola was originally described from an area of coastal vegetation (Restinga de Maricá), whereas C. alpina and C. caelivittata sp.nov.are from mountainous regions (Ita tiaia massif in Mantiqueira mountain range and Serra dos Órgãos massif in Serra do Mar mountain range, respectively).However, after the publication of the original description of C. bromelicola (Mejdalani et al. 2014), we and colleagues have commonly collected or observed specimens of this species from ornamental bromeliads in urban and semiurban areas in southeastern Brazil, including the Bromeliad Collection of the Jardim Botânico do Rio de Janeiro (JBRJ), a small square in Lumiar in the municipality of Nova Friburgo, and several public or private gardens (both lowland and mountainous localities).Quintas and Mejdalani (2021) provided a table that records ten bromeliad species as hosts of C. bromelicola, including plants from JBRJ and natural environments.Examination of the color pattern and male terminalia of specimens from anthropic environments indicates that they can be confidently identified as C. bromelicola, although small variations were observed, especially in the crown of apical spines of the aedeagus.Perhaps, such variations have arisen through genetic drift (Futuyma 1992, Ridley 2006), due to sexual reproduction limited to a very small number of closely related sharpshooters contained within a single or a few adjacent bromeliads (i.e., endogamy).In our field studies, we have observed C. bromelicola individuals almost exclusively on bromeliads, a fact suggesting that they do not use other kinds of plants as hosts.Apparently, these sharpshooters have very limited dispersal capacity, which means that they most likely do not fly for long distances from one bromeliad to another.The present hypothesis of a single species should be further investigated through molecular approaches and a phylogeographic analysis.As observed by Quintas and Mejdalani (2021), the trade of bro meliads can cause the artificial dispersal of C. bromelicola.Considering that sharpshooters are the main vectors of the gramnegative, xylemlimited bacterium Xylella fastidiosa Wells et al., 1987(Wells et al. 1987), which causes serious diseases in various cultivated plants in Brazil and elsewhere (Redak et al. 2004, Carvalho et al. 2022), we recommend that specimens of C. bromelicola be tested for the presence of this phytopathogen.

ACKNOWLEDGMENTS
GM is a research productivity fellow form Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, process 306197/20219).VQ and NHP are doctoral students from Programa de Pósgraduação em Ciências Biológicas -Zoologia (Museu Nacional, UFRJ); they receive stipends from CNPq and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior -CAPES, respectively.JAF is a postdoctoral fellow at Escola Superior de Agricultura "Luiz de Queiroz" (EsalqUSP); she is supported by Fundação de Estudos Agrários "Luiz de Queiroz" (FEALQ) and by Fundação de Apoio à Pesquisa Agrícola (FUNDAG; Hori zon Project 2020 1220IA).SRC and APS are undergraduate students and receive PIBIC stipends from UFRJ and CNPq, respectively.This work was partially supported by Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ, process E26/200.082/2019;Apoio Emer gencial ao Museu Nacional).Field works in Nova Friburgo benefited from grants provided by Sociedade Brasileira de Entomologia (SBE, chamada #20201) to VQ and NHP.We thank the curator of the Bromeliad Collection of the Jardim Botânico do Rio de Janeiro (JBRJ), Bruno Rezende, and the gardeners José Mourão and Rhanna H.N. Bastos for their support during the collecting work of C. bromelicola.We also thank the biologists André L.D. Ferreira, André A. Alves, and Gabriel A. Oliveira for the help during the field works in Nova Friburgo.The manuscript benefited from the useful comments of Ângelo P. Pinto (Universidade Federal do Paraná) and two anonymous reviewers.
The reader is referred to Mejdalani et al. (2014), Quintas et al. (2020), and Quintas and Mejdalani (2021) for details on the distribution and collecting data of specimens of C. bromelicola and C. alpina.