Leptogenys pubiceps

Lattke (2011) has suggested this species is part of a complex. The resolution of the status of the various forms requires a detailed taxonomic treatment.

Identification
Lattke (2011) - A member of the unistimulosa species group but this nominal species is a complex that requires additional collections and study. Some members of this species complex could be confused with Leptogenys arcuata on account of similar Gestalt, both sharing the same general shape and with overlapping ranges in the Guianas and some of the Lesser Antilles. L. arcuata is smaller and lacks the presence of the crest or tooth on the petiolar apex and has the posterior petiolar margin usually straight to weakly sinuate in lateral view, a margin that is always strongly sinuate in the pubiceps complex. The third antennal segment is more than twice longer than wide in the pubiceps complex, whilst it is less than twice its apical width in ''L. arcuata. L. arcuata also has the eyes apparently more laterally placed on the head, a trait that resembles some L. pubiceps'' populations. The propodeal declivity in L. arcuata tends to be flat and separated from the lateral propodeal face by a relatively sharp margin, this in contrast with the blunt, curved margin which prevails in L. pubiceps.

Distribution
Lattke (2011) - This species complex has a distribution range restricted to the Caribbean Basin. It is found on the mainland from Heredia and Limon Provinces of Costa Rica to northern Colombia, the northern slopes of the Venezuelan Andes, the Venezuelan Coastal Range and Trinidad. It is found on both the Greater and Lesser Antilles with no records from the Bahamas nor southern Florida.

Distribution based on Regional Taxon Lists
Neotropical Region: Barbados, Colombia, Cuba, Haiti, Puerto Rico, Venezuela.

Biology
Lattke (2011) - These ants have been collected in numerous habitats that range from dry forests to humid and cloud forests, lowland and montane, besides coffee plantations. They have been found beneath rotten logs, and in leaf litter. The usual reaction of the ants upon disturbance to the nest is to rapidly flee and hide amongst the leaf litter.

Haiti
"A nest of L. puncticeps at Diquini was in the ground, beneath debris, in an unused tobacco shed." (Wheeler and Mann 1914). Despite the misspelling, this material was being referred to the species Leptogenys punctaticeps. Based on the ranges of L. punctaticeps and L. pubiceps, and without having examined these specimens, this ant is most likely to be part of the pubiceps species complex.

Castes
The queen and male are unknown.

Nomenclature

 *  pubiceps. Leptogenys pubiceps Emery, 1890a: 62 (w.) VENEZUELA. Forel, 1893g: 362 (q.m.). Unresolved pubiceps-complex, containing the nominal taxa columbica, cubaensis, mucronata, pubiceps, vincentensis: Lattke, 2011: 217.

Lattke (2011) - Slender, slightly arched mandibles of uniform width; eye large, convex, occupying more than one-fourth lateral cephalic margin. Hypostomal tooth varies from clearly visible to just showing apex in cephalic full-face view; second antennal segment two-thirds length of third. Apex of petiolar node with blunt projection ranging from crest to tooth, generally overhanging posterior margin in lateral view, posterior margin sinuate.

Perhaps reflecting the complicated geological history of the Caribbean, this complex has a bewildering array of morphological diversity that challenges analysis. Variations in overall size, scape length, head shape, mandibular width, color, and development of the apical petiolar process will distinguish individual populations, yet when all series are taken into account there is enough crossing of characters from one population to the next to throw any sense of order into disarray. The shape of the node apex permits rough separation of two trends within this complex: one has the petiolar node topped by a very modest posterior crest, whilst the other has a distinct blunt tooth on the node apex.

The crest is a posterior flattened projection that varies from thinly laminate in northeastern Venezuelan specimens, to more thickened further west, but not swollen medially to form a bulge that could be defined as a tooth or tubercle. Leptogenys pubiceps, Leptogenys vincentensis (= Leptogenys pubiceps vincentensis), and Leptogenys cubaensis (= Leptogenys pubiceps cubaensis) are forms with a crested node. Forel (1901b: 328) described L. vincentensis by contrasting it with Leptogenys specimens he calls L. pubiceps collected by himself in the Sierra Nevada foothills of Magdalena, Colombia. He cites the following differences of the insular species from the Colombian specimens: smaller eyes, with a diameter that does not reach the anterior cephalic margin, a very short and obtuse median clypeal lobe, ferruginous legs and scape, convex posterior cephalic border, and a poorly develop petiolar tooth and different node shape. From these contrasts and the fact that all L. pubiceps complex samples known from the Sierra Nevada area correspond to the toothed node forms, it seem obvious what he is calling L. pubiceps is most certainly Leptogenys columbica (=Leptogenys mucronata columbica). Santschi (1930: 76) describes L. cubaensis without providing any information on how it differs from L. pubiceps. Samples from northeast Venezuela, Hispaniola, Cuba, Virgin Islands, St. Vincent, and Puerto Rico correspond to this form. Most of these have the head longer than broad except for the series from the Serranía de San Luis (Northcentral Venezuela) which tend to be broad. The specimens from Cuba and Hispaniola are slightly smaller in than the rest of the studied material. The angle separating the posterior petiolar face from the lateral face is blunt in specimens from Puerto Rico and St. Vincent compared with samples from other sites.

The color and petiolar shape of the L. pubiceps holotype resembles more that of specimens from St. Augustine, Trinidad than from the Venezuelan Cordillera de la Costa. The St. Augustine specimens have broader heads, subquadrate mesonotum, scape surpassing the posterior cephalic margin by one-fourth its length, and ferruginous appendages. The petiolar node lacks the sulcus and has a very modest crest. Specimens collected nearest to the type locality, such as Los Caracas or Naiguatá, differ in the darker coloration, subrectangular mesonotum, scape surpassing the head by one-third its length, and the petiolar shape which has a shallow pre-apical sulcus and posterodorsally directed thin crest.

Leptogenys mucronata and Leptogenys columbica are forms with a toothed node. The syntypes consist of 2 pins, each with three point-mounted workers. Forel (1901b: 328) separates L. columbica from L. mucronata on account of the convex posterior cephalic margin with a clearly visible vertexal carina when seen in full-face view. He also discerned the more abrupt petiolar tooth in dorsal view that contrasts with the gradually tapering tooth of L. mucronata, and the transversely striate pronotum of L. columbica that differs from the more smooth pronotum of L. mucronata. Both, the shape of the posterior cephalic margin and visibility of the vertexal carina hinge upon the angle of observation, so these two characters are not reliable. Other useful differences are the visibility of the hypostomal teeth in L. mucronata, its mandibles are not as arched and the head tends to be broader anterad when compared with L. columbica. Samples from Grenada, Jamaica, St. Vincent, Trinidad, northern Colombia, northwestern Venezuela, Panama, and Costa Rica have a definite blunt point on the node apex. Specimens from Magdalena, Colombia and a single specimen from Tovar, Mérida, Venezuela have the head distinctly wider anterad than posterad, with the series from the Sierra de San Luis, Venezuela only slightly wider anterad than posterad. All other material has parallel-sided lateral cephalic margins, with the Arima Valley (Trinidad) specimens having the head broader than long. The mandibles are usually quite slender but in specimens from Trinidad they are noticeably thicker and have fine longitudinal irregular striae. The Costa Rica specimens are rather homogenous and quite distinct on account of their broad head and particularly massive mandibles. Lattke & Longino (2009) cite the Costa Rican population-species as JTL-002 (cf. punctaticeps) in the Ants of Costa Rica website. The series from Arima, Trinidad are also quite distinct due to their broad head, thick mandibles (not a as thick as the Costa Rican samples) and large size, the largest of the complex. Two specimens from Bolivar, Colombia are smaller in size, have a flattened mesosomal dorsal margin in lateral view, relatively short scapes, and longer second funicular segment compared with most other specimens.

The crested node specimens have the head with subparallel lateral margins in dorsal view, usually longer than broad and relatively narrow mandibles, whilst the toothed specimens will vary in mandible thickness, and some will have the head wider anterad than posterad. The toothed node specimens tend to have the eyes more dorsolaterally situated than the crested node specimens, this being especially obvious in the Arima and Costa Rican series. The two forms are sympatric on Trinidad, and St. Vincent. The differences in size and sculpture support assuming they are different species. Some specimens from Cuba have an intermediate shape that can not be sorted into these two categories, and have been labeled as L. pubiceps complex.

It is tempting to synonymize L. columbica with L. mucronata, and consider L. cubaensis and L. vincentensis synonyms of L. pubiceps, as it is also tempting to describe the Costa Rican and the Arima, Trinidad populations as distinct species. Nevertheless the degree of subjectivity involved in such decisions would be great, since arguments could be made for recognizing the species status of several other populations. While such action could give the illusion of order, it would also hide the obviously complex situations of many of these populations, a reflection of their reproductive history, paleogeography, sea level changes, vicariance and dispersal. The only obvious conclusion is that this species complex should eventually be object of a thorough revision, but more material from many localities is needed before there is any hope of obtaining better resolution. The fact that queens are wingless make it particularly interesting for studies of Caribbean biogeography. Care should be exercised since some patterns could be confounded by human mediated population dispersions due to the movement of agricultural products such as cocoa and coffee seedlings, banana rootstock, and even barrels of soil.

Worker
Lattke (2011) - Metrics (n = 9) : HL 1.28 – 1.55; HW 1.21 – 1.42; ML 0.88 – 1.15; EL 0.30 – 0.40; SL 1.35 – 1.72; PW 0.84 – 1.01; WL 2.19 – 2.63; PH 1.01 – 1.15; PL 0.67 – 0.78; DPW 0.54–0.78 mm. CI 0.84–0.95; MI 0.70–0.84; OI 0.24 – 0.29; SI 1.08 – 1.32; LPI 1.45 – 1.63; DPI 0.73 – 1.05.

Head as long as wide or wider than long in full-face view; lateral margins semi-parallel to wider anterad than posterad, lateral margin slightly arched to straight, posterior margin convex to broadly convex. Eye large, convex, occupying more than one-fourth of lateral cephalic margin. Clypeus with transverse striae laterad, becoming longitudinal medially; anterior clypeal margins converge to brief anteromedian lobe, lobe shorter than basal scape width, ending in 3–4 setae. Hypostomal tooth varies from clearly visible to just showing apex in cephalic full-face view. Mandible elongate and slender, slightly arched, of uniform width, with single apical tooth, single preapical denticle usually present; mandibular dorsum usually smooth and shining, sometimes with fine longitudinal striae. Scape extends one-third its length beyond cephalic posterior border; second antennal segment two-thirds length of third. Cephalic dorsum punctulate, ranging from dense to sparse; dorsal surface of labrum mostly smooth and shining with scattered low tubercles; PF: 4,4.

Mesosoma with convex dorsal promesonotal margin separated by metanotal groove from convex propodeal margin in lateral view; promesonotal convexity ranging from very convex to broadly convex; propodeal dorsal margin broadly convex, separated by blunt angle from declivitous margin, declivitous margin one-third the length of dorsal margin; mesonotum sometimes forming third convexity. Mesosomal sculpturing mostly transversely rugulose with smooth areas along lateroventral pronotum and mesosomal dorsum; propodeal spiracle ellipsoid; metapleural-propodeal suture broad, with transverse crests, widest posterad; mesometapleural suture well impressed; propodeum without teeth or prominent lobes or crests.

Petiolar node in lateral view subquadrate, with more or less continuously curved anterodorsal margin; anterior margin usually shorter than dorsal margin, dorsal margin inclined at varying angles relative to longitudinal axis of petiole; apex of node with blunt projection ranging from crest to tooth, generally overhanging posterior margin in lateral view, posterior margin sinuate; sides of node with oblique rugosity; posterior face separated from sides by sharp margins, tending to flat with fine transverse rugulae; node slightly longer than wide in dorsal view. Subpetiolar process forms convex lobe in lateral view; as wide as sternite in ventral view, posterior face deeply concave, laterally bound by thin cuticle. Anterior margin of abdominal segment III mostly vertical in lateral view. Gaster smooth and shining with scattered piligerous punctulae. Body has abundant erect and semi-erect hairs; scape and cephalic dorsum with appressed pilosity. Body black; legs, antennae and mandibles brown to dark brown; clypeus black to dark brown.

Type Material
Lattke (2011) - Leptogenys pubiceps Emery, 1890a: 62. Holotype worker: Venezuela, La Guaira [examined].

Leptogenys pubiceps var. vincentensis Forel, 1901b: 328. Syntype workers: Antilles, St. Vincent [examined].

Leptogenys mucronata var. columbica Forel, 1901b: 328. Syntype workers: Colombia, Sierra Nevada de Santa Marta, Narancho (A. Forel) [examined]. Leptogenys pubiceps st. cubaensis Santschi, 1930: 76. Syntype workers: Cuba, Habana, Almandares (M.L. Jaume) [examined].

Leptogenys mucronata Forel, 1893: 360. Syntype workers: Antilles, St. Vincent, Richmond Valley, 800 ft, 13.xi.[no year given], shady damp place near the river, under log (H.H. Smith) [examined].