Warimiri Tavares, de Mello & Mendes g en. nov.

urn:lsid:zoobank.org:act: DE1766C9-E04C-452C-9EA3-116D6DC9215C

Type species. Warimiri madiba sp. nov. (described below).

Etymology. The name is an arbitrary combination of letters derived from two words of the Nheengatu: wariní = warrior + mirĩ = small. Nheengatu is a language created by the Jesuit priests to homogenize several native idioms of the Tupi branch, making it possible for them to communicate between the tribes. Nheengatu is considered the New Tupi language. The gender of the name is being established as neuter.

Diagnosis. Fastigium of vertex blunt, moderately protruding, wider than scapus, not dentate below, and contiguous with the fastigium of frons (Figs. 2A–E, G–H; 6C–D; 7B–D; 11B–D; 14C–D, G–H; 19C–D). Dorsal surface of head, frons, and genae conspicuously rugose and punctuate (Figs. 2C–E; 7B–C; 11B–C; 14C–E); antennal scape with an inward blunt tooth (Figs. 2D; 11B; 14D), posterior portion of each antennomere darker than the remaining. Eyes subglobose, slightly broader than the fastigium of vertex (Figs. 2A–E, G–H; 6C–D; 7B–D; 11B–D; 14C–D, G–H; 19C–D). Pronotum slightly convex in lateral view; pronotal disk distinctly lighter than the lateral lobes, with rounded lateral keels and an almost inconspicuous medial keel, more notable on the posterior half, resembling a roof (Figs. 2B; H; 6B, D; 7C, E; 11C, E; 14B, H; 19B, D); anterior and posterior margins truncated, this last produced behind. Lateral lobes produced laterally, with cephalic margins straight and oblique, anteroventral angles obtuse, anteroventral margins sinuous and oblique, posteroventral angles obtuse, posterior margins sinuous, and humeral sinus inconspicuous (Figs. 2A, G; 6A, C; 7A, D; 11A, D; 14A, G; 16A, C). Prosternum smooth, meso- and metatasternum wider than long, trapezoid (Figs. 2F; 7F; 11F; 14F). Wings significantly reduced (micropterous), almost entirely covered by the pronotal disk posterior margin (Figs. 2G–H; 4A–B; 6A–D; 7A, C–E; 11A, C–E; 14G–H; 16A–B; 19A–D). In males, both tegmina slightly surpassing the first tergite anterior margin (Figs. 2G–H; 7A, C–E; 14G–H), and hind wings trifling (Figs. 2A–B; 14A–B). Left tegmen anal margin conspicuously prominent, very similar to the right tegmen protruding plectrum (Figs. 4A; 8A; 16A). Mirror of both tegmina membranous, hyaline, and subtriangular (Figs. 4A–B; 8A–B; 16A–B). Right tegmen plectrum flexible, with a hyaline and more membranous cell formed between two anal veins (Figs. 4B; 8B; 16B). In females, both pairs of wings even more reduced, trifling (Figs. 6A–D; 7A, E; 19A–D). Legs short and stout; anterior tibiae’s dorsal surface elevated into two low lateral keels (Figs. 3A–C; 7G; 11G; 15A–C); all femora dorsally unarmed (Figs. 3A–B, D–E, G–H; 7A, G–H; 11A, G–H; 15A–B, D–E, G–H); anterior femora with minute ventral spines internally (Figs. 3B; 7G; 11G; 15B) and externally generally smooth (Figs. 3A; 15A); mid femora usually smooth internally (Figs. 3E; 15E) and externally with minute ventral spines (Figs. 3D; 7H; 11H; 15D). Male subgenital plate widely expanded laterally and dorsally, emarginated posteriorly, with two cylindrical styli (Figs. 3J–L; 7I–K; 15J–L); cerci short, bearing inward spines distally or medio-proximally (Figs. 3M–N; 7I; 15M–N). Titillatory process (ti) comprising a continuous sclerotized area that can be limited to the dorsal fold (df) (Figs. 5A–B, E–F) or extends from the fold df to the dorsal cavity (dc) (Figs.17A–B, D–F). Titillator’s sclerites (TS) paired and long, rod-shaped (Figs. 1A–F; 5A–B, D–E; 9A, D–E; 17A–D); anterophallic membranous processes of the dorsal lobe (mp.dl) pared, attached to the apical most portion of sclerite VS (Figs. 5B–C, F; 9B; 17A–C, F); sclerite of the ventral fold of the dorsal lobe (VS) large, inverted Yshaped. Female subgenital plate narrow (basally narrower than the proximal portion of ovipositor ventral valves), emarginated posteriorly, not produced dorsolaterally, and basally flanked by two membranous invaginations (better seen when the subgenital plate is lowered) (Figs. 1H–I; 6G; 11K; 19G), located just before the anteriormost margins of the ventral valves (Figs. 1H–I, white arrows); ovipositor strongly upcurved (Figs. 6E; 11I; 19E).

Warimiri gen. nov. is morphologically similar to Hyperomerus and Dectinomima and probably is closely related to them. However, Hyperomerus has two distinct sclerotized areas comprising the processes ti; processes mp.dl attached to the process es ti anterophallic surface; cerci usually elongated, bearing a compressed projection and a basal appendage or spine (only Hyperomerus almeirina Tavares, Sovano & Gutjahr has short cerci, with reduced compressed projection). On the other hand, Dectinomima has no processes ti; the sclerites TS are neither elongated nor rod-shaped, and the cerci bear an elongate mesointernal tapered projection anteriorly or posteriorly upcurved.

Comments. Montealegre-Z et al. (2011) recorded two undescribed species tentatively determined as “near Uchuca ” (i.e., Hyperomerus) to the pacific coast of Colombia and Ecuador. Till the moment, it is not sure if these species belong to Hyperomerus or another undescribed genus. The available knowledge is that all species already described to Hyperomerus are limited to the East side of the Andes, extending to the Amazonian region (Tavares & Nunes-Gutjahr 2021), and the two species within Dectinomima are recorded to Panama and the Pacific coast of Colombia, on the trans-Andean region (Montealegre-Z & Morris 2003). Warimiri gen. nov. is the first genus of Agraeciini with blunt and large fastigium of vertex (larger than scapus) recorded in the Atlantic forest. Nowadays, this biome is isolated from the Amazon by two biomes of the ‘open formation diagonal’ (also known as ‘dry diagonal’): Cerrado and Caatinga (Batalha-Filho et al. 2013; Werneck 2011).

The new species described here were collected in the States of Ceará, Alagoas (one species with disjoint distribution), and Bahia (two species). We believe that Warimiri gen. nov. is closer to Hyperomerus than Dectinomima due to the similarity of the phallic complex, but a phylogenetic approach is needed to confirm this hypothesis.

Comments on the phallic components. Males genital structures tend to diverge faster than other morphological structures, driven by different select pressures (Simmons 2014). In katydids, the sclerites TS play an important role in female acceptance for copula by stimulating the internal membranes of the copulatory chamber (acting as a courtship device) (Wulff et al. 2015, 2018), applying pressure that forces the subgenital plate to stay open (Wulff et al. 2017) or mechanically supporting the mating position and the spermatophore transference (Wulff & Lehmann 2020). The copula duration varies among Tettigoniidae subfamilies, but it is significantly longer in species with sclerites TS than in species without these structures, and more complex sclerites TS transfer spermatophorous quicker (Vahed et al. 2011). In addition, many species of Tettigoniidae have processes ti, sclerotized areas that bear sclerotized microstructures, which probably also act as stimulation devices (Chamorro-Rengifo & Lopes-Andrade 2014), increasing the stimulatory capability of the phallus. However, morphofunctional studies have not considered these last structures (Vahed et al. 2011; Wulff et al. 2015, 2017, 2018; Wulff & Lehmann 2014, 2020). In fact, any other component of the phallic complex has not been considered, especially the internal ones.

In Warimiri gen. nov. and Hyperomerus, the posterior portion of the phallic complex bears very long and paired sclerites TS and paired or single processes ti. Internally (or anteriorly), a very large sclerite VS and a pair of processes mp.dl are also present in both genera. Despite being on different faces of the phallus, in these two genera, the sclerite VS supports the sclerite TS eversion, acting as an umbrella’s runner (Figs. 1A–E). So when the phallus is everted (by the hydrostatic pressure of hemolymph), it is physically easier to push a unique structure (the runner) that, in turn, forces the two sclerites TS outward (acting as the umbrella’s stretchers) than directly forcing two independent mobile bars (Fig. 1G, black arrow). For us, this sizeable central sclerite also helps to maintain the phallus wholly everted.

On the other hand, it is also easier to retract the phallus by pulling a unique central structure. However, we believe that a second component facilitates the phallus retraction, the processes mp.dl. These processes are located on the dorsalmost portion of the anterophallic face in both genera and are connected directly to the sclerite VS apex (in Warimiri gen. nov.) or the processes ti (in Hyperomerus). We believe these processes act like tendons, pulling back inside the sclerite VS (Fig. 1G, white arrow). Ventrally, a membranous projection of the phallus is connected to the subgenital plate and acts like a frenulum, helping to retract the phallus (Fig. 1F, white arrow).

Dectinomima has shorter sclerites TS and apparently no processes ti, but we do not have any information about the anterophallic components of this genus, so we will not discuss it.