Protalaridris

Lattke et al. 2018 recognize seven species, 6 of which are described as new. Sporadic observations of one long-mandibulate species suggest they are sit-and-wait ambush predators that open their jaws to approximately 180° when stalking. A nest has never been found.

Identification
Lattke et al. (2018) - The inclusion of Protalaridris leponcei, with its strikingly different mandibles, obliged a redefinition of the genus (i.e., changes to the genus as described in Brown 1980), but most changes are limited to mandibular morphology, as the majority of other characters are shared, including the number of antennal segments. The labral margin hairs are very short in Protalaridris leponcei with none longer than 1/4 the labral width while in the other species some hairs will always exceed half the labral width in length. A monotypic genus could have been proposed but there is a long history of ant genera established because of variances in mandibular morphology, with further studies forcing a taxonomic weeding out of excess names. Differences in mandibular morphology and antennal segmentation have traditionally been important in separating genera within the Basiceros and Strumigenys groups, but recent work has gathered convincing evidence to the contrary, especially in the latter genus. It was only after the extensive work of several myrmecologists (Baroni Urbani & De Andrade, 1994; Bolton, 1999; Ward et al., 2015) that many genus-group names were sunk as junior synonyms of Strumigenys. A tipping point was finally reached with Protalaridris leponcei when J. Longino and M. Branstetter graciously shared the preliminary results of their UCE molecular analysis of the Basiceros group of genera and these clearly indicate the inclusion of Protalaridris leponcei as an ingroup within Protalaridris.

The two strikingly different mandibular morphologies permit easy recognition of two informal species groups within the genus: (1) The armata group, with elongate mandibles, and (2) the leponcei group, with short, triangular mandibles. Most species of the armata group have the frontovertexal ridge arching laterally until it either meets the eye or misses it by an ocular diameter, whereas in Protalaridris leponcei the ridge is separated from the eye by several times its diameter. This latter condition is also found in the ascrobicula group of Octostruma (Longino, 2013a). Within the armata group Protalaridris aculeata and Protalaridris bordoni share the prominent frontovertexal ridge and a mostly concave vertex except for a narrow flat strip that borders the occipital carina, in sharp contrast with the blunt frontovertexal ridge with a mostly convex vertex of the other known species.

Brown (1980) - General characters as in Rhopalothrix and Talaridris, but with the following differences: Antennae 9-merous; segments III- VII short and transverse. The following details describing mandibular features do not apply to Protalaridris leponcei, as noted in the previous two paragraphs. Mandibles long, slender, their insertions remote, but the shafts converging to cross at apices when closure is complete, each tapering toward an acute, incurved, straight apical spine. As seen from the side, shafts curved markedly dorsad from base toward apex away from main axis of cranium, much as in Talaridris. Inner margins of mandibular shafts each armed with 2 long, spaced, slender teeth and 3 smaller teeth or denticles in addition to apical spine.

Rhopalothrix and Talaridris are similar but have antennae 7-merous. In Talaridris, mandibles are also upcurved, but are short compared to Protalaridris armata, and the shorter and longer teeth are concentrated in a small area near to the apex, though they still appear to be homologous with the more widely-spaced teeth of Protalaridris.

Biology
Lattke et al. (2018) - Known specimens of this genus had all been taken from leaf litter samples. One litter sample from Venezuela contained several workers and a queen of Protalaridris punctata, implying a ground nesting site within leaf litter or decomposing branches, twigs or other plant matter. Protalaridris is found in mesic forests ranging from 200 to 2500 m a.s.l., although by far most records are from montane cloud forests between altitudes of 800 and 2000 m. Most have been taken from isolated collecting events though some may be locally abundant. Specimens were collected in 10 percent of 2 transects using the ALL protocol in cloud forests of the Venezuelan Coastal Range on the southern facing slopes between 1225 to 1450 m (Lattke & Riera, 2012), but none were recorded from one cloud forest transect, 16 km to the west, also using the ALL protocol, at 1100 m on north facing slopes (Rodriguez & Lattke, 2012). In some localities they may be quite frequently encountered such as in the cloud forest Otongachi Reserve at 850 m in the western Andean slopes of Central Ecuador, where Protalaridris armata has been found in 20% of leaf litter samples (Donoso & Ramon, 2009) or in Reserva Biológica San Francisco, in the eastern range of the south Ecuadorian Andes, where two species of Protalaridris, Protalaridris leponcei and Protalaridris loxanensis, have been found in a total of 6 (24%) out of 25 Winkler samples (but with only one specimen in each sample). Nevertheless sampling montane cloud forests in northern South America does not guarantee finding these ants as surveys in apparently suitable sites in SW Colombia failed to recover any specimens (Estrada & Fernandez, 1999; Bustos & Ulloa-Chacon, 1997). An unpublished survey (Farfán, 2014) using the ALL protocol in three cloud forest sites situated between 1300–1700 m in the Sierra de Aroa mountains of northern Venezuela, 122 km W from the Rancho Grande locality, also failed to detect any Protalaridris. Given the large areas of potentially suitable habitat in Colombia and Peru it is expected that more species of Protalaridris should be uncovered in these countries. Its presence in Bolivia and Brazil is also possible considering the few kilometers separating the type locality for Protalaridris aculeata from the borders of these countries. During the course of this study we had access to images of a disarticulated specimen of long-mandibulate Protalaridris from Risaralda, Colombia. Its cephalic and mandibular morphology indicate it represents an undescribed species.

No field observations of habits are yet recorded for this genus though the mandibular morphology suggests specialized predation. The biology of most Basiceros group species is poorly known as they are generally described from leaf litter samples. Casual observations of a single live worker of Protalaridris loxanensis taken by JEL from a leaf litter sample offered some clues to their predatory habits. It was placed in a petri dish with a humid paper towel. The ant would walk slowly with its head directed anterad, elevated from the ground. Whilst walking the head would slowly sweep right and left with the antennae describing vertical elongate ovals. During movement the mandibles would be either closed or open up to 90 degrees. The labrum and its surrounding hairs project forward. Several times it was observed with its mandibles opened almost 180°, but only when it was immobile. During this stance it would keep its body and head closer to the ground than when walking, its scapes extended posteroventrally and tucked beneath the head with the funiculus extending anterolaterally. A fine-hair paint brush as well as a human hair glued to a toothpick were both used in attempts to stimulate the ant into closing its mandibles, but in both cases it would close its mandibles slowly and back away. Thus, the mandibles may not be of the rapidly closing type or the attempts to stimulate a reaction simply failed.

With mandibles closed and in lateral view, it is possible to discern the major anteroventral tooth and the protruding labrum with its suite of long hairs. The labral hairs are subparallel to the ventral tooth, and their apices frequently surpass in length the tooth’s apex. The labrum protrudes at a roughly 45° angle with the ventral mandibular margin in all studied specimens. The mandibles can be opened with comparative ease in ethanol-preserved specimens (Protalaridris armata and Protalaridris loxanensis) to a gape of approximately 180°, such as in trap-jaw ants of the genera Odontomachus, Anochetus, Acanthognathus, or Strumigenys. The finely pointed and polished apices of both the mandible and its ventral tooth suggest either penetration of the body of its prey or a more secure grip to facilitate use of its sting, or both. The protruding labrum with the row of long apical hairs suggest a latch mechanism controlled by trigger hairs that make contact with the prey (Cerdá & Dejean, 2011). But Bolton’s (1999) studies of mandibular morphology indicate that a T-shaped labrum is associated with a sudden release, kinetic mode of mandibular closure. The labral morphology in Protalaridris is more like that of a static mandibular closing mechanism, but the wide gape is typical of ants with a kinetic mechanism. Another, closely related genus, Rhopalothrix, also presents acutely pointed mandibles and a similarly protruding labrum with specialized hairs, and it is also possible to open their mandibles to a gape as in Protalaridris (Longino & Boudinot, 2013) but they lack the well-developed basal mandibular process. There is no conclusive evidence regarding the nature of the jaw closure mechanism in Protalaridris. The Otongachi Reserve in Ecuador offers excellent opportunities for natural history studies of this genus or for collecting nests for lab study as Protalaridris armata is relatively common there.

Mandibular morphology
Lattke et al. (2018) - The mandibular morphology of the Protalaridris armata group contrasts with that of most other ant genera due to its upturned dorsal margin when seen laterally. In Hymenoptera the mandibular shape is typically crescent-shaped in lateral view and triangular in dorsal view. Ant mandibles, as in most insects, are dicondylate and this constrains mandibular movement along a transverse (laterally) oriented plane (Staniczek, 2000; Blanke et al., 2014). The abduction and adduction are produced by abductor and adductor muscles originating on the dorsal and lateral internal surface of the head capsule (Snodgrass, 1935; Grimaldi & Engels, 2005). Given the overwhelming dominance of the triangular – subtriangular mandible variants, it can be assumed this is a plesiomorphic and conserved general shape and an inversion of the mandibular apex from downturned to upturned should be rare. Deviations from this are the specialized linear mandibles of genera such as Acanthognathus; Anochetus; Daceton; Mystrium; Myrmoteras; Odontomachus, and Strumigenys, which are straight when seen laterally (Larabee & Suarez, 2014). Even rarer are the instances of “up-turned” (and always more or less elongated) mandibles.

Ants in the ponerine genus Harpegnathos have elongate upturned mandibles convergently shaped with those of Protalaridris. Both genera also share the presence of a large crescent-shaped basal ventral tooth with a convex anterior margin in lateral view. The monotypic genus Talaridris has an elongate scooped-shaped mandible but with a distinct ventral denticle that has a concave posterior margin when seen laterally. Besides these extant groups the extinct sphecomyrmicine genera Ceratomyrmex and Haidomyrmex also bear elongate mandibles with an upturned apical part and ventral tooth at the base (Perrichot et al., 2016). Long-mandibulate Strumigenys of the gundlachi group have relatively slender, tapering mandibles that in profile may be linear to weakly concave (Bolton, 2000: 176). Their mandibles differ from the aforementioned genera in lacking a basal ventral tooth and the dorsal concavity, when present, is feeble at best. The masticatory margin of triangular mandibles generally bear variously shaped teeth that lie in the same plane as the dorsal mandibular surface. This configuration is different in the ascrobicula group of Octostruma (see Longino, 2013a) and Protalaridris leponcei. Their mandibles are triangular but the masticatory margin bears a series of relatively large, irregularly sized teeth that protrude dorsomedially, clearly visible in lateral view. One of the preapical teeth of Protalaridris leponcei is particularly large compared with the others (Fig. 6a), and further enlargement may lead to the situation found in the remaining species of the genus. Such enlargement may be exemplified by an undescribed species Rhopalothrix UFV sp. 1 imaged in Antweb (2017; specimen UFV-LABECOL-000326). In the latter, the mandible bears a very large preapical tooth that rivals the mandibular shaft in size. If the mandibles in Protalaridris leponcei represent the plesiomorphic state within the genus, further enlargement of a dorsomedially projecting tooth might have proved advantageous for certain tasks, such as prey securement (Ohkawara et al., 2016). A tooth or denticle projecting from the basal mandibular angle is known in some species of Neivamyrmex (Borgmeier, 1955). Thus, the “upturned” part of the mandibles in the armata group could be derived from a preapical mandibular projection. That possibility is supported by the presence of a large incurved ventral tooth close to the mandibular base of in all these ants; this tooth may actually represent the original apical part of the mandible. Additionally, in the long mandible, only the part distal to the ventral tooth is distinctly upturned in Protalaridris, Talaridris, and Harpegnathos. The trap-jaw mandible in the genus Acanthognathus bears a slender and curved basal tooth or process (Gronenberg et al., 1998) that is anteriorly convex and with apical denticles, suggesting the possibility it may be the original mandible, but the morphology is much different than of the previously mentioned genera.

Our interpretation may also be useful for addressing enigmatic mandibular morphology in non-extant ants. What Barden & Grimaldi (2012: 8) interpret as the apical tooth of Haidomyrmex scimitarus may actually represent a hypertrophied preapical tooth whilst the apex of the so-called ventrobasal mandibular tooth (the posteriormost member of the basal teeth) is most probably the original apical mandibular tooth. Likewise, what is labelled as the basal tooth in Ceratomyrmex ellenbergeri (Fig. 1C in Perrichot et al., 2016) would be the principal mandibular shaft and the apical portion of the mandible would be a hypertrophied preapical tooth or projection. As in extant taxa, the upturned part of the mandibles in these early ants begins close to the “basal mandibular tooth”. Fig. 11 depicts our idea of what is the main mandibular shaft and what is the extension of the preapical dentition or margin.

The mandibles of Haidomyrmex and Ceratomyrmex were interpreted as being unlike anything present in modern ants (Barden & Grimaldi, 2016; Perrichot et al., 2016; Barden 2017) and even implying a vertical movement (Perrichot et al., 2016). In our interpretation, the mandibles of Haidomyrmecini possess analogues in extant genera and it is not necessary to postulate vertical mandibular movement which would imply a 90 degree twist of the mandibular condyles and associated radical structural modifications of the head capsule, mandibular apodemes, and muscles. Haidomyrmecine and Talaridris mandibles correspond to the planar type (sensu Keller, 2011), with the external surface facing laterally and the external margin facing ventrally, whilst mandibles in Protalaridris and Harpegnathos are torqued, with their external surface facing dorsally and the external margin facing laterally (Keller, 2011). What still makes haidomyrmecine mandibles remarkable is the length and the angle at which the dorsal tooth projects, almost perpendicular to the longitudinal axis of the body, something unseen in extant taxa. Hopefully, detailed phylogenies and careful comparative studies of closely related taxa with different mandibular morphology will help elucidate the developmental pathways of such transformations.

Wing Morphology
Lattke et al. (2018) - Only a few winged individuals are known for the genus. The following discusses differences and similarities between the wing venation of Protalaridris and some of the genera of the Basiceros group. The fore wing of Protalaridris bordoni and Protalaridris loxanensis is very much like that of Basiceros scambognathus (Brown, 1949) as illustrated in Brown & Kempf (1960: 173), but the Cubital vein separates from M-Cu at the junction with cu-a in Basiceros scambognathus whereas in Protalaridris the separation of Cu occurs at a distance from cu-a. In Protalaridris 1A continues briefly apicad of cu-a but it stops at cu-a in Basiceros. The fore wing of Protalaridris loxanensis has the costal cell well-defined between tubular veins C and Sc+R as in Rhopalothrix Mayr, 1870; basal cell well-defi ned between tubular veins Sc+R and M+Cu (similar to Rhopalothrix but M+Cu darker, better defined in Protalaridris); submarginal cell mostly well-defined though Rs is partly spectral between M and 2rs-m; subbasal cell closed as M+Cu, 1A, and cu-a markedly tubular; Cu tubular for a distance. Fore wing with 4 closed cells in Protalaridris and Basiceros (costal, basal, submarginal, and subbasal); in Rhopalothrix with 2 (costal and basal). The hind wing in Protalaridris loxanensis has tubular C+Sc+R, Sc+R, and Rs+M; 1A is briefly tubular and M+Cu is also partially tubular. In contrast the fore wing venation of Rhopalothrix subspatulata is relatively reduced, with Rs+M, M, and Rs apicad of M all reduced to spectral veins, cu-a is absent and A has a vestigial tubular section (Longino & Boudinot, 2013). Additionally the hind wing of Rhopalothrix subspatulata has only C+Sc+R developed as tubular veins whilst Sc+R and Rs+M are spectral. As in Protalaridris, 1A is briefly tubular but M+Cu is totally spectral.

Castes
Males have not been collected for any species in this genus.

Nomenclature

 *  PROTALARIDRIS [Myrmicinae: Basicerotini]
 * Protalaridris Brown, 1980a: 36. Type-species: Protalaridris armata, by original designation.
 * Protalaridris junior synonym of Basiceros: Baroni Urbani & De Andrade, 2007: 88.

Description
Lattke et al. (2018) - Worker. Head in dorsal view about as long as wide, widest posterior to eye; cephalic dorsum with transverse blunt frontovertexal ridge, posteriorly broad convex; occasionally ridge sharp and narrow (Protalaridris aculeata & Protalaridris bordoni). Vertex broad and convex when ridge is blunt, mostly concave when ridge is sharp with narrow flat strip bordering the occipital carina. Brief median longitudinal carina present on frontal area, extending posterad from posteromedian clypeal margin. Epistomal suture well defi ned, shaped as open inverted V; clypeus posterolaterally forms anterior part of antennal fossa, anterior margin of frontal lobe meets posterior clypeal margin through narrowly arched space. Vertex both posterior and anterior to frontovertexal ridge feebly but broadly impressed on each side; anterior clypeal margin generally with broad median concavity. Head in lateral view with broadly convex posterodorsal margin that meets dorsal margin at an angle; cephalic dorsal margin straight to broadly concave.

Compound eye relatively small, directed anterolaterally, with 1–12 usually indistinct ommatidia; eye separated from dorsal cephalic surface by at least one diameter, sometimes less but never directly bordering the dorsal cephalic surface. No erect pilosity on cephalic dorsum but Protalaridris leponcei bears one erect spatulate (see the hair shape webpage for definitions and illustrations) hair on each side of head, posterior to compound eye and lateral to frontal carina; medially pointing subspatulate to lanceolate appressed hairs generally present. Antenna issues forth on each side through a deep semicircular notch in the dorsolateral cephalic margin; antennal fossa shaped as sinuous emargination along anterolateral cephalic margin in fullface view, interior of fossa mostly finely areolate; antennal scrobe very shallow and broad, extending posterad of compound eye. Antenna 9-segmented with 2-segmented club, pedicel suboval, segments 3–7 short and transverse, apical segment longer than preceding seven segments. Scape flattened, robust with flat to convex dorsal surface, ventral surface finely areolate, anterior basal lobe slightly expanded, except in Protalaridris aculeata where the lobe is lamellate, anterior margin with 5–10 spatulate to reniform hairs.

Mandible: 1. armata species-group: mandibles long and slender, in full length view straight to slightly arched, insertions remote, shafts crossing at apices when completely closed, each tapering apicad towards medially directed apical tooth, tooth dark brown, shining, and acutely pointed. Mandibles form complex cradle or cage mainly bound by the main axis of each mandible as well as one, or two, prominent ventral teeth. Base of mandible at cephalic insertion expands into flange with smooth rim and areolate dorsal surface; mandibular dorsum strongly sculpted, ventral surface weakly sculpted; apical tooth and apex of massive ventral tooth/teeth smooth. In lateral view mandibular dorsal margin concave or forms obtuse angle with main longitudinal axis of cephalic capsule. Mandible generally with two internal rows of preapical teeth, one ventral another dorsal; dorsal teeth sometimes absent or reduced in number (Protalaridris armata). Dorsal preapical teeth numbering 1–7 pointing mesad, relatively small, never crossing each other; ventral preapical teeth numbering 4–5, of varying size with at least some teeth that cross, including one or two massive ventromedially directed teeth. Mandibular dorsum with abundant appressed and elongate lanceolate to simple hairs that arch anterad. 2. leponcei species-group: mandible short, stout, and triangular; dorsal margin in lateral view strongly bowed with basal portion in same plane as clypeus and apical portion strongly bent downward. Masticatory margin with single row of 9 preapical teeth of irregular size, protruding dorsomedially. Apical and longest preapical teeth crossed with mandibles closed. In frontal view first basal tooth of mandible continuous with basal rim.

Labrum extended, shape varying from rounded to rectangular with an anteromedian notch ranging in depth from shallow to over half the length of the labrum, six to 32 usually flattened hairs present along margins, the longest anterolaterally placed and the shortest posterolaterally placed and within the anterior notch. With head in dorsal view the labrum mostly visible with mandibles closed in armata group but entirely hidden from view in the leponcei group.

Mesosoma subpyriform to pyriform in dorsal view, broadest across anterior pronotum, cervix marked off by a blunt arcuate margin. Mesosomal dorsal margin continuously convex in lateral view, curvature ranging from weakly convex to strongly convex; no sutures evident across dorsum, except for shallow narrow furrow usually obscured by particulate matter, immediately anterad of transverse carina marking top of concave dorsal part of the propodeal concavity, and presumably marking the anterior margin of propodeum. Propodeal spiracle surrounded by elevated ring of cuticle, opening faces posterolaterally, located approximately at half length of declivity less than one diameter from tooth; dorsal propodeal margin very brief; declivity finely areolate ventrad of upper margin of teeth, rugulose dorsad, separated from dorsum by distinct transverse carina, not covered by debris or encrustations. Petiole relatively short with poorly developed node, node transverse in dorsal view, obliquely subtruncate in lateral view, postpetiole twice as wide as petiole, rounded above, with trace of median longitudinal sulcus posterad, and weak median posterior emargination. Underside of petiole biconvex, anteroventral process shaped as discrete angle or lacking. No erect hairs on dorsum of mesosoma, petiole, and postpetiole; mesosomal lateral surface without standing pilosity. Encrustations present on dorsal surface of petiolar node and postpetiole.

Abdominal tergite IV with broadly convex main dorsal surface, dorsolaterally bordered by longitudinal blunt ridge that defines elongate lateral vertical region of tergite, transverse ridge present along anterior margin; gastral tergum with 20–50, sometimes more, erect to suberect spatulate hairs arranged in at least four longitudinal rows; anteroventral gastral process lacking. Legs short and compact, femora gradually thickened apicad, and tibiae even thicker, tarsi more slender; 1–2 spatulate hairs present on apex of tibiae, meso- and metatarsal segments with pairs of erect spatulate hairs in V; tarsal claws small.

Queen
Lattke et al. (2018) - Wing Venation (for Protalaridris bordoni and Protalaridris loxanensis gynes). Fore wing with 4 closed cells (costal, basal, submarginal, and subbasal); veins C, Sc+R, M+Cu, Cu, cu-a, and 1A tubular; Cu tubular for a distance; Rs partly spectral between M and 2rs-m; separation of Cu from M-Cu occurs at a distance from cu-a; 1A continues briefly apicad of cu-a. Hind wing with tubular C+Sc+R, Sc+R, and Rs+M; 1A briefly tubular and M+Cu partially tubular.