Leptogenys

A widely ranging tropical genus represented by close to 300 species. Many form small colonies (<50 workers) and are predators of isopods. Nest are typically made in the leaf litter, soil, rotten wood, or occasionally in vegetation. Some Australasian and Oriental species have evolved an army ant lifestyle. Bolton (1975a) revised the Afrotropical species and Lattke (2011) treated the New World species. Both revisions include useful keys for their respective regions. Fast moving workers, with many species foraging at night, mean these ants are not as well collected for their diversity and likely abundance in the areas where they occur.

Identification
Lattke (2011) - The apparent monophyly of Leptogenys is nowadays not questioned, probably due to several outstanding characters that qualify as apomorphies when compared with other ponerines. Besides the traditional pectinate claws and carinate median clypeal lobe, additional characters useful for separating this genus from other Ponerini may be considered: the presence of a basal protarsal comb is common in the most Ponerini, but is lost in Leptogenys. In Pachycondyla impressa the anterior tentorial pit is located close to the antennal sclerite whilst in Leptogenys it is quite separated. The tentorial pit in Myopias has a similar location to Leptogenys, and the two also share the presence of a median clypeal lobe and basal mandibular sulcus. The clypeal lobe in Myopias is usually subquadrate to rectangular, thicker and opaque throughout when compared with Leptogenys. Additionally there are Myopias species lacking the lobe, all suggesting this lobe is not homologous for the two groups. Presence of the basal mandibular sulcus is considered a plesiomorphic character (Schmidt 2009, in thesis). The relative situation of the tentorial pit should studied in other ponerines.

Identification Keys

 * Leptogenys species groups (New World species)
 * Malagasy Leptogenys species groups
 * Key to Leptogenys of the New World
 * Key to Malagasy Leptogenys
 * Key to US Leptogenys species

Species richness
Species richness by country based on regional taxon lists (countries with darker colours are more species-rich). View Data



Modified from Lattke (2011) – Most New World species seem to have relatively modest, localized ranges, usually smaller than those observed in other Ponerini. None have a range that extensively includes both North and South America, or most of either continent such as the case for some species in Pachycondyla, Hypoponera, or Odontomachus. The species with the largest ranges almost all include the Orinoco-Amazon watershed as the core area of their distribution, with the most widely ranging species, Leptogenys famelica and Leptogenys unistimulosa, extending marginally into Colombia and Central America, and Leptogenys arcuata including some records from the Caribbean. L. arcuata may eventually prove to be more than a single species and the pubiceps complex, found throughout the Caribbean is most certainly a collection of species. Save for Leptogenys langi and Leptogenys nigricans, the species with the most extensive distributions are found outside of the pusilla clade. L. langi has winged queens and L. nigricans queens have a full set of sclerites associated with flight but it is unknown if their wings are functional.

African Leptogenys also suggest a similar pattern of reduced geographic ranges (Bolton 1975). The only exceptions to this are tramp species such as Leptogenys falcigera, Leptogenys maxillosa, and Leptogenys pavesii, which now enjoy a human mediated pantropical distribution (Bolton 1975), and Australasian Leptogenys with army ant behavior. Leptogenys diminuta, for example, ranges from Ceylon and India eastward into Melanesia and Queensland, Australia (Wilson 1958; Shattuck 1999). “True” army ants, despite possessing wingless queens, have far ranging geographic distributions (Kempf 1972; Watkins 1976). Presumably in these cases the nomadic life-style of the species has contributed to overcoming the apparent dispersal and colonization limitations of wingless queens (Peeters 2001) that conceivably make it more difficult to cross unsuitable habitat than in species with winged queens. Males in Leptogenys are not particularly robust and do not seem strong fliers, nor would be expected to fly very far since potential mates must be sought at ground level in forests. Male production per colony is probably not very high given the modest nest sizes in the genus, and the scant number of males seen in colonies collected by the author. With few exceptions males were not considered in this study, but a cursory look at the unidentified Leptogenys males in MIZA attracted to light traps show that in some places the number of flying males can be considerable. It remains to be seen if the males from a single light trap collection event are all from one nest or from several nests. If we add to this the frequent specializations in prey and habitat, and a highly diverse fauna with the certainty that this revision has missed a significant amount of species, it seems likely Leptogenys populations are more likely to be affected by potential vicariant processes than more mobile and less specialized species of other ponerine genera (Schmidt 2009, in thesis).

Habitat and Abundance
Modified from Lattke (2011) – In the New World Leptogenys may be locally abundant in some areas, which are usually of humid forest and with an abundance of isopod prey. In such an area direct collecting may uncover up to 5 or more nests in a single day, usually of the same species. Personal observations in at least 3 Venezuelan collecting localities with notably abundant Leptogenys colonies coincided with abundant isopod populations as evidenced by the scattering of numerous of these crustaceans upon treading the leaf litter late afternoon and early evening. Dejean (1997) confirms the pattern of abundant Leptogenys colonies in areas with an elevated presence of isopods in Mexico and Cameroon, even in anthropized areas. Collections during several years in single localities, such as La Selva in Costa Rica or Rancho Grande in Venezuela, indicate that as many as 5 – 6 species may be present in a given locality.

Nests may be found from ground level to higher than 2000 m above sea level, though most are lowland dwellers.

Even though these ants are more common in mesic habitats, they are no strangers to arid, desert conditions, with a number of species endemic to xeric areas such as Baja California (Leptogenys peninsularis), the Sonora desert (Leptogenys sonora), or the Galapagos Islands (Leptogenys santacruzi, Leptogenys cf. gorgona). The dry, interior part of Australia is also habitat for a number of Leptogenys species (Shattuck 1999), and several species of Leptogenys are known from semiarid parts of Africa (Bolton 1975). One species has been found deep in two caves in Laos, both caves being rich in guano and isopods (Roncin & Deharveng 2003).

Two species (Leptogenys khammouanensis and Leptogenys elongata) have been found in caves.

Biology
The following synopsis of the biology of Leptogenys is based on Lattke (2011), who in turn based much of his account on Schmidt’s unpublished thesis work (Schmidt 2009).

Colony Attributes
Nests of New World Leptogenys may vary from 20 to 30 workers, rarely surpassing 50 (pers. obs). Small nest sizes (30 <) are also reported for over 15 species of this genus in the Oriental Tropics by Ito (2000) with the outstanding exception of some SE Asian Leptogenys with army ant habits, which have colonies numbering in the thousands (Witte & Maschwitz 2000). During the course of this study no evidence was found hinting at army ant habits in any of the New World Leptogenys.

Nesting Biology
The nests may be found in rotten wood on the ground, usually within cavities in logs or large branches, and also beneath bark. The wood-soil and rock-soil interfaces are also used for nesting, as well as rock crevices, and a few may nest directly in the soil such as Leptogenys famelica. Some species may be adapted to disturbed areas, such as the pantropical tramp Leptogenys maxillosa, which has been found nesting in cracks and fissures of buildings in urban areas in Brazil (Freitas 1995). Nest entrances of the larger ground nesting species may be recognized by the scattered exoskeletons of isopod prey discarded from the nest, At least one species, Leptogenys elegans Bolton, 1975 from West Africa, will nest in dead wood above the ground and forage on tree-trunks (Bolton 1975). Dejean & Olmsted (1995) report 4 species of Leptogenys (Leptogenys donisthorpei, Leptogenys maya, Leptogenys sianka, Leptogenys wheeleri) nesting in epiphytic bromeliads and orchids in inundated forests in northern Yucatan. This record is of interest due to the arboreal nest sites, a situation unknown for any other New World species of the genus, and rarely encountered in other faunas (Bolton 1975). The Reserva Sian Ka’an site is lowland forest flooded with 20 – 60 cm of water, a situation discouraging the establishment of nests on the ground, but the rich epiphytic flora has created a suspended soil and is home to abundant isopods. At least three of the aformentioned Leptogenys species have also been found nesting on the ground (see species accounts) so there is no evidence yet for obligate arboreal Leptogenys, at least for the New World fauna. In this reserve Dejean (pers. comm.) frequently found Leptogenys in mangroves on the brackish water lagoon side of the stands, but not on the oceanic side.



Foraging/Diet
Specialized predation seems to dominate in this genus, with observations throughout its range corroborating oniscoid isopods as the prey of choice for most species (Dejean & Evraerts 1997; Dejean 1997). That a group of ants should specialise on oniscideans should come as no surprise as these crustaceans frequently constitute an important part of the soil fauna with population densities reaching in the hundreds per square meter (Quadros & Araujo 2008). The earliest known fossils of oniscideans are from Baltic Amber, with fossils also known from the lower Miocene of Kenya and Dominican Amber as well. The earliest known remains of other crustaceans within the Scutocoxifera, in which terrestrial isopods are included, range from the Middle to Upper Jurassic (Schmidt 2008), implying the existence of a diverse terrestrial isopod fauna before the major radiation in Leptogenys (Schmidt 2009, in thesis). Other predatory specializations have been detected such as earwigs (Steghaus-Kovac & Maschwitz 1993) for an Oriental species and termite predation for the African nitida group (Bolton 1975), and also for Leptogenys unistimulosa Roger, 1861 in Brazil (Mill 1982). The Oriental species that follow an army ant lifestyle are generalists, taking diverse prey (Witte & Maschwitz 2000). Individuals in the larger species carry isopod prey slung beneath their bodies, between the legs, and smaller species will carry large prey items amongst several individuals. In Costa Rica a group of workers of Leptogenys volcanica were observed and photographed carrying an isopod considerably larger than the size of each individual ant, suggesting either group raiding (E. Rodríguez, D. Kronauer, pers. comm.), or perhaps recruitment to a victim disabled by one original worker (B. Bolton, pers. comm.). Duncan & Crewe (1993) report both individual foraging and group raiding amongst South African species of Leptogenys. Dejean & Evraerts (1997) also report group raiding in the genus.

Reproduction
The two most common modes of reproduction in the genus are either through ergatoid queens or egg-laying workers (Ito 1997; Ito & Ohkawara 2000). The queen in Leptogenys is generally an ergatoid, and only in three New World species are “normal” queens found in addition: in Leptogenys langi, in Leptogenys nigricans, and in an undescribed species of Leptogenys collected in Texas (Cokendolpher et al. 2009). The latter apparently is a member of the pusilla clade judging from the images illustrating the publication. Wingless queens, but with well-developed wing-base sclerites and ocelli are also known for Leptogenys ergatogyna in Africa (Bolton 1975). Ergatoid queens generally lack any trace of ocelli, though when present it is a single, somewhat reduced median ocellus. The gaster is usually larger in comparison to the workers, the petiolar node, when seen dorsally, is generally wider than long, and the propodeal margin is more convex when observed laterally than in the worker. Additional characters that may characterize queens are larger overall body dimensions, broader head with more convex sides when seen in full-face view, and the compound eyes may be slightly larger.

At least three species belonging to the pusilla group: Leptogenys pusilla, Leptogenys ritae, and Leptogenys josephi have ergatoid queens with greatly swollen mandibles of a pale yellow color. This unusual character was first reported for L. josephi by MacKay & MacKay (2004), and was thought to be diagnostic for the species. The only queens so far known for L. josephi and L. pusilla are of the enlarged mandibular type, whilst in L. ritae normal mandibulate queens as well as swollen mandibulate queens have been found. The first two species are sympatric in Central America, with L. ritae perhaps having some sympatry with the other two in Panama, though it is mostly distributed throughout northern South America, including Trinidad. The possibility exists the two queens may correspond to different species. Not only is the purpose of the unusual mandibles a mystery, but also the nature of the relation between the two forms. Perhaps the enlarged mandibles habour hypertrophied glands. It should also be noted that the gaster of the queen of these species is comparatively larger than in most other Leptogenys queens. The morphology of these queens is not unlike that of dichthadiiform queens of nomadic species such as Simopelta or ecitonines on account of the enlarged gaster and modified mandibles, but no New World species are known to have an army ant lifestyle.

Reproduction by workers is suspected in all species of at least 3 species groups (arcuata, ingens, and unistimulosa) and in one species each of two groups, Leptogenys famelica of the famelica group and Leptogenys gaigei of the luederwaldti group. This possibility is based on the total lack of recognizable queens in these species despite the repeated collection of many nests. Gamergate reproduction could be more widespread in Leptogenys as indicated by Bolton’s (1975) observation that out of the 56 species known for the Ethiopian Region, females are known for only 6 or 7. Freitas (1995) reported males copulating with “workers” just outside the nests of Leptogenys maxillosa, and never observed winged queens during nuptial flights or nest relocation events. In a survey of 11 species from Malaysia and Indonesia, Ito (1997) found gamergates in 4 species, with the rest having ergatoid queens.

The predominance of ergatoid queens leaves open the question of colony reproduction. While most nests apparently are monogynous, nests have been found with two or more queens in Leptogenys ritae and Leptogenys orchidioides (see individual species accounts). A small, apparently incipient, nest of approximately 3 queens and 4 workers was found in L. ritae (R. Johnson, pers. comm.), suggesting polygyny. Perhaps some of the nest series represent virgin queens that have yet to leave the nest, indicating that several virgin queens may accumulate in the nest before dispersing. Do they leave the nest by themselves or does each take an escort of workers, or perhaps they leave as a group or groups? Nest fission, as in ecitonine army ants, is known for the army ant lifestyle diminuta group in SE Asia (Ito & Ohkawara 2000).

Associations with other Organisms
Beyond the specialized diet of many species (see Foraging/Diet section above), there are numerous other insects that have been found associated with Leptogenys. A phorid fly of the genus Megaselia Rondani, 1856 has been reported attacking Leptogenys mutabilis in Borneo (Disney & Fayle 2008). Pheidole megacephala, a common invasive ant in the tropics, has been scored in laboratory experiments (Dejean et al. 2008) as an effective enemy of Leptogenys, killing off colonies of a native species of Leptogenys in Cameroon, sometimes in less than a hour, as well as Leptogenys elongata from Mexico. A diverse number of symbionts have been recorded for army ant Leptogenys, including the only known molluscan myrmecophile in Leptogenys processionalis distinguenda (Witte et al. 2002).

Nomenclature

 *  LEPTOGENYS [Ponerinae: Ponerini]
 * Leptogenys Roger, 1861a: 41. Type-species: Leptogenys falcigera, by subsequent designation of Bingham, 1903: 52.
 * Leptogenys senior synonym of Microbolbos: Wilson, 1955b: 136.
 * Leptogenys senior synonym of Dorylozelus: Taylor, 1969: 132.
 * Leptogenys senior synonym of Machaerogenys, Lobopelta: Bolton, 1975a: 240.
 * Leptogenys senior synonym of Odontopelta: Snelling, R.R. 1981: 390; Taylor & Brown, D.R. 1985: 32.
 * Leptogenys senior synonym of Prionogenys: Taylor, 1988: 33.
 * DORYLOZELUS [junior synonym of Leptogenys]
 * Dorylozelus Forel, 1915b: 24. Type-species: Dorylozelus mjobergi (junior secondary homonym in Leptogenys, replaced by Leptogenys tricosa), by monotypy.
 * Dorylozelus junior synonym of Leptogenys: Taylor, 1969: 132.
 * LOBOPELTA [junior synonym of Leptogenys]
 * Lobopelta Mayr, 1862: 733. Type-species: Ponera diminuta, by subsequent designation of Bingham, 1903: 54.
 * Lobopelta subgenus of Leptogenys: Forel, 1892k: 520.
 * Lobopelta revived status as genus: Dalla Torre, 1893: 43; maintained as genus: Bingham, 1903: 54.
 * Lobopelta junior synonym of Leptogenys: Emery, 1896e: 177 (footnote)
 * Lobopelta revived from synonymy as subgenus of Leptogenys: Forel, 1899c: 18; Forel, 1900d: 304; maintained as subgenus: Wheeler, W.M. 1910g: 135; Emery, 1911d: 101; Forel, 1917: 238; Wheeler, W.M. 1922a: 653; Creighton, 1950a: 50; Kempf, 1972a: 131.
 * Lobopelta junior synonym of Leptogenys: Bolton, 1975a: 240.
 * MACHAEROGENYS [junior synonym of Leptogenys]
 * Machaerogenys Emery, 1911d: 100 [as subgenus of Leptogenys]. Type-species: Leptogenys truncatirostris, by original designation.
 * Machaerogenys junior synonym of Leptogenys: Bolton, 1975a: 240.
 * MICROBOLBOS [junior synonym of Leptogenys]
 * Microbolbos Donisthorpe, 1948f: 170. Type-species: Microbolbos testaceus, by original designation.
 * Microbolbos junior synonym of Leptogenys: Wilson, 1955b: 136.
 * ODONTOPELTA [junior synonym of Leptogenys]
 * Odontopelta Emery, 1911d: 101 [as subgenus of Leptogenys]. Type-species: Leptogenys turneri, by monotypy.
 * Odontopelta junior synonym of Leptogenys: Taylor & Brown, D.R. 1985: 32.
 * PRIONOGENYS [junior synonym of Leptogenys]
 * Prionogenys Emery, 1895g: 348. Type-species: Prionogenys podenzanai, by monotypy.
 * Prionogenys junior synonym of Leptogenys: Taylor, 1988: 33.

Lattke (2011) - The genus Leptogenys as presently defined stems most recently from Bolton’s (1975) revision of the fauna from the Ethiopian Region and Madagascar. The generic worker diagnosis given below was taken using only New World species into account, and comparing with Bolton (1975). In a revisionary study of the ponerine genera based on molecular markers, Schmidt (2009, in thesis) found unequivocal molecular evidence for grouping Leptogenys along with 17 other genera into a monophyletic entity called the Odontomachus group, mostly of Old World distribution but with many species of Leptogenys, Anochetus and Odontomachus present in the New World. Within this group, Myopias is considered sister to Leptogenys, with evidence for both genera constituting one part of the basalmost divergence in the group, and the remaining genera the other part. Schmidt (2009, in thesis) argues for an Old World diversification of the Odontomachus group with age estimates ranging from 22 – 40 my according to diverse constraints, with a preferred estimate of 30 (38 – 24) my. The major radiations in Leptogenys have occured with an estimated crown age of 32 my (Schmidt 2009, in thesis). Examination of a few random Paleotropical species show character states not seen in New World Leptogenys such as a heavily dentate mandible, scale-like petiole, frontal lobe covering almost all of the antennal condyle or a moderately convex anterior clypeal margin. At least some traits are arguably plesiomorphic, thus lending a bit of morphological support to Schmidt’s conclusions based on molecular data

Worker
Lattke (2011), New World.

Head shape ranges from elongate to wider than long in full-face view; vertexal carina present, usually visible from cephalic full-face view; compound eye usually situated anterad of mid-cephalic length, occasionally at mid-length, usually dorsolaterally situated on head; eye diameter may be very reduced, just a few ommatidia, to very large and prominent, its length covering more than one-third the lateral cephalic margin with the head in full-face view. Mandible variously shaped, either triangular or subtriangular, frequently elongate with subparallel internal and external margins, also falcate and incapable of closing against clypeus; masticatory margin usually shorter than basal margin, generally edentate or nearly so; rarely masticatory margin crenulate or with series of blunt denticles; apical tooth usually present, mandible usually with laterobasal sulcus. Clypeus relatively long with median portion of clypeus projecting anterad as distinct, variously shaped, usually triangular lobe, with a median longitudinal crest extending from apex to level of frontal lobes, clypeus extends briefly posterad between frontal carinae as narrow wedge; clypeus with narrow carinae or lobe bordering anterolateral cephalic margin; frontoclypeal suture straight and transverse, tentorial pit closer to compound eye than to antennal sclerite. Labrum frequently with tubercles on external face; palp formula 4,4 or 4,3. Frontal lobe covers less than half antennal insertion, frontal carinae very brief; antennae 12-segmented, usually weakly incrassate; scape usually surpasses posterior cephalic margin, rarely shorter; funicular segments subcylindrical or moniliform.

Propleural lateral and ventral face separated by curvature, not carinate; pronotum never with carinate sides; mesopleuron with anepisternum and katepisternum usually indistinctly separated; propodeal spiracle separated from declivitous margin in lateral view by at least 3 diameters; mesopleural carinae usually distinct, sometimes weakly developed, especially ventrad; pronotal suture mobile; mesonotum always distinct; metanotal groove distinctly impressed, from shallow and fine to deep and wide, smooth or scrobiculate; mesometapleural suture distinct, scrobiculate, posterodorsal edge of mesopleuron forming a distinct ledge; metapleural-propodeal suture varies from indistinct to well impressed. Propodeum with or without lateral teeth or lobes, declivitous face usually with broad transverse sulcus next to insertion of postpetiole.

Posterior face of anteroventral petiolar process medianly concave; prora well developed, usually shaped as transverse crest or lobe; gaster usually smooth and shining, sometimes abdominal segments III – IV punctate; pretergite of abdominal segment IV with stridulitrum; constriction between abdominal segments III and IV ranging from weakly developed to well developed; pygidium with or without longitudial crest; hypopygidium usually with a row of small setae. Tibiae without setae on lateral or dorsal surface; protibial apex usually without setae apicad of insertion of strigil (except L. panops); first protarsal segment without comb of stout setae opposite strigil; strigil without velum at base (except L. gaigei and L. cuneata); mesotibial apex usually with one small seta on external face, sometimes absent or several setae present; metatibial apex usually lacking apical setae; claws usually pectinate (this state is reduced in minute species such as L. pusilla or L. gorgona with few or no preapical teeth), arolium absent; meso- and metatibial apex each with 2 apical spurs. General body color ranges from black to ferruginous; mandibles, clypeus, antennae, legs, and gastral apex usually lighter colored than head, mesosoma, and gaster; blue or purple opalescence present in some species; sculpture varies from smooth and shining to striate, or punctate.