Odontomachus

Seen up close, a large Odontomachus worker moving slowly across the leaf litter in a dipterocarp forest in the Philippines (or elsewhere) is spectacular. The bizarre head of species of this genus bears strong jaw-muscles, and large, elongate trap-jaws which are carried sideways when foraging, ready to snap shut as potential prey comes in touch with the long frontal trigger hairs. A powerful stinger at the apex of the abdomen completes the image. These ants resemble nothing so much as small alligators lost among the dimensions of the litter. (Sorger and Zettel 2011)

Identification
Schmidt and Shattuck (2014) - Workers of Odontomachus are so distinctive that they are difficult to confuse with those of any other genus except Anochetus, the sister genus of Odontomachus. The unusual trap mandibles and head shape of Odontomachus are synapomorphic with Anochetus, but the genera are readily differentiated by examination of the posterior face of the head. In Odontomachus the nuchal carina is V-shaped medially, and the posterior surface of the head has a pair of dark converging apophyseal lines. In Anochetus the nuchal carina is continuously curved and the posterior surface of the head lacks visible apophyseal lines. These genera also tend to differ in size (Anochetus are generally smaller, though there is some overlap), propodeal teeth (absent in Odontomachus but usually present in Anochetus), and petiole shape (always coniform in Odontomachus, but variable in Anochetus).

Odontomachus species groups

Distribution
Schmidt and Shattuck (2014) - Odontomachus is abundant in the tropical and subtropical regions of the world, though it is most diverse in the Asian tropics and the Neotropics. Australia boasts a handful of species, while Africa has two species (Odontomachus assiniensis and Odontomachus troglodytes) and the Malagasy region has three species (Odontomachus coquereli, Odontomachus troglodytes, and Odontomachus simillimus, the latter apparently introduced to the Seychelles; Fisher & Smith, 2008). A few species extend into temperate regions, notably in the southwestern United States, northeastern China, central Argentina, and southwestern Australia (reviewed in Brown, 1976).

Biology
Hita Garcia, Wiesel and Fischer (2013) - Mainly tropical and subtropical in its distribution but some species can also be found in temperate zones or even in semi-desert areas (Brown, 1976). The more than 70 species are roughly equally split between the New and the Old World. Species of Odontomachus are large, often conspicuous ants that are mainly predaceous (Brown, 1976; Fisher & Smith, 2008). Like Anochetus, they are trap-jaw ants and can also use these specialised mandibles to jump away if they feel disturbed (Brown, 1976). They hunt other arthropods, often termites, but a few also tend hemipterans. They generally nest in the soil, the leaf litter, or in rotten logs while few species live arboreally (Brown, 1976)

Schmidt and Shattuck (2014) - In most respects Odontomachus are fairly typical ponerines. The nesting habits of many species have been observed, and most of these nest in soil or rotting wood (e.g., Odontomachus affinis: Brandão, 1983; Odontomachus bauri: Ehmer & Hölldobler, 1995; Odontomachus brunneus, Odontomachus clarus, Odontomachus relictus, and Odontomachus ruginodis: Deyrup & Cover, 2004; Odontomachus cephalotes: Wilson, 1959b; Odontomachus chelifer: Fowler, 1980; Passos & Oliveira, 2004; Odontomachus coquereli: Molet et al., 2007; Odontomachus erythrocephalus: Longino, 2013; Odontomachus opaciventris: de la Mora et al., 2007; Odontomachus rixosus: Ito et al., 1996; Odontomachus simillimus: Wilson, 1959b; van Walsum et al., 1998; Odontomachus tyrannicus: Wilson, 1959b), though some species will nest in more unusual locations such as in abandoned termite nests (Déjean et al., 1996, 1997) or arboreally (e.g., Odontomachus troglodytes: Colombel, 1972; O. brunneus, Odontomachus hastatus, and Odontomachus mayi; Brown, 1976; O. bauri and O. hastatus: Longino, 2013). The nests of O. bauri are apparently polydomous (Ehmer & Hölldobler, 1995). Odontomachus workers are monomorphic and are epigeic foragers, and some species are at least partially arboreal in their habits (Brown, 1976; Longino, 2013). Most species are generalist predators of arthopods, though many species partially specialize on certain types of prey, especially termites (e.g., Fowler, 1980; Lévieux, 1982; Ehmer & Hölldobler, 1995). At least some species will also tend honeydew-secreting insects or visit extrafloral nectaries (e.g., O. affinis: Borgmeier, 1920; O. bauri, O. hastatus, and Odontomachus panamensis: Schemske, 1982; Longino, 2013; O. troglodytes: Evans & Leston, 1971; Lachaud & Déjean, 1991a), and the Neotropical species O. chelifer is known to eat fruit and the arils of certain seeds, which the ants ultimately disperse (Pizo & Oliveira, 1998; Passos & Oliveira, 2002, 2004). Odontomachus laticeps and Odontomachus minutus (=Odontomachus meinerti) also collect seeds with nutritious arils (Horvitz & Beattie, 1980; Horvitz, 1981). [[Odontomachus malignus is notable for its habit of foraging among corals at low tide (Wilson, 1959b). Foraging workers of O. bauri navigate using visual cues from the forest canopy overhead as well as chemical cues (Oliveira & Hölldobler, 1989). Recruitment of nestmates via tandem running was observed in O. troglodytes'' (Lachaud & Déjean, 1991a).

Colony size is highly variable across the genus, ranging from an average of only 18 workers in O. coquereli (Molet et al., 2007) to as many as 10,000 workers in O. opaciventris (de la Mora et al., 2007). Most species seem to have colony sizes of several hundred workers: O. chelifer colonies average between 100 to 650 workers (Fowler, 1980; Passos & Oliveira, 2004), colonies of O. rixosus had an average of 142 workers (Ito et al., 1996), and O. bauri is reported to have up to 300 workers per colony (Jaffe & Marcuse, 1983), though O. troglodytes colonies can have over 1,000 workers (Colombel, 1970a).

Most species of Odontomachus have typical winged queens and semi-claustral nest founding (Brown, 1976), though O. coquereli has wingless ergatoid queens and colonies apparently reproduce by division (Molet et al., 2007). An undescribed species from Malaysia is also reported to have ergatoid queens (Gobin et al., 2006), and colony reproduction by fission is suspected to occur in some other species (Brown, 1976). While some Odontomachus species are likely to be monogynous, many species are polygynous (e.g., Odontomachus assiniensis: Ledoux, 1952; O. cephalotes: Peeters, 1987; O. chelifer: Medeiros et al., 1992; O. rixosus: Ito et al., 1996; Odontomachus troglodytes: Ledoux, 1952). Queens of O. rixosus perform many of the tasks more typical of the worker caste, including foraging outside the nest (Ito et al., 1996). In the most detailed series of studies on a single Odontomachus species, Colombel examined various aspects of the behavior of O. troglodytes, including caste determination (Colombel, 1978), egg development (Colombel, 1974) reproduction by workers (Colombel, 1972), ecology, nest structure, demographics and population dynamics (Colombel, 1970a), egg-laying by queens (Colombel, 1970b), and alarm pheromones (Colombel, 1968). The laying of haploid eggs by workers has also been observed in O. chelifer (Medeiros et al., 1992), O. rixosus (Ito et al., 1996), and O. simillimus (van Walsum et al., 1998). Wheeler et al. (1999) examined the egg proteins of O. chelifer and O. clarus.

Only a handful of papers have been published on the social behavior of Odontomachus. Polyethism in Odontomachus troglodytes was studied by Déjean & Lachaud (1991), while division of labor in Odontomachus affinis was examined by Brandão (1983). Powell & Tschinkel (1999) discovered that the workers of Odontomachus brunneus organize themselves into a social hierarchy via ritualized dominance interactions, with repercussions for task specialization within the nest. Whether similar heirarchies exist among workers in other Odontomachus species is unknown, though dominance heirarchies exist among queens in colonies of the polygynous species Odontomachus chelifer (Medeiros et al., 1992). Jaffe & Marcuse (1983) observed both nestmate recognition and territorial aggression in Odontomachus bauri. Aspects of the mating behavior of Odontomachus assiniensis, the other African Odontomachus species, were studied by Ledoux (1952).

Wheeler & Blum (1973) identified the mandibular glands as the source of alarm pheromones in O. brunneus, O. clarus and O. hastatus. Morgan et al. (1999) examined the mandibular gland secretions of O. bauri, while Longhurst et al. (1978) studied the mandibular gland secretions of O. troglodytes and the response of males to these secretions. Oliveira & Hölldobler (1989) identified the roles of pygidial, mandibular and poison gland secretions in O. bauri for recruitment, alarm and attack behaviors. Alarmed Odontomachus workers can also stridulate (e.g. Carlin & Gladstein, 1989).

The trap mandibles and associated behaviors of Odontomachus (and Anochetus) rank among the most specialized of any ponerine. When hunting, Odontomachus workers hold their highly modified mandibles open at 180° and shut them with extreme force and speed on their prey. In fact, this is the fastest movement ever measured in any animal (Patek et al., 2006; Spagna et al., 2008). The contact of trigger setae (located beneath the mandibles) with the prey triggers the mandibular closure. The morphological, physiological and neurological characteristics of trap mandibles (and associated structures and behaviors) have been extensively studied (e.g., Gronenberg et al., 1993; Gronenberg & Tautz, 1994; Gronenberg, 1995a, 1995b; Ehmer & Gronenberg, 1997; Just & Gronenberg, 1999; Paul & Gronenberg, 1999; Spagna et al., 2008). Kinematic data indicate that the force of jaw closure in Odontomachus scales positively with body size, while acceleration scales inversely with body size (Spagna et al., 2008). The significance of these scaling relationships for the optimal foraging strategy in a given species is unknown.

The sequence of actions taken during prey capture by a hunting Odontomachus worker was summarized by de la Mora et al. (2007). Upon detection of a suitable prey item, the worker antennates it, then withdraws the antennae and snaps its mandibles shut on the prey. Generally the prey are held in the mandibles, lifted off the substrate, stung, and then transported back to the nest, though sometimes stinging is not necessary (Brown, 1976). The exact behavioral sequence used during prey capture varies somewhat depending on the Odontomachus species and the identity of the prey. For example, multiple mandibular strikes may be used to stun or dismember the prey. Odontomachus workers are often cautious during prey capture, especially with potentially dangerous prey such as termites. De la Mora et al. (2007) describe the predatory behavior of Odontomachus opaciventris in detail; the foraging behaviors of several other Odontomachus species have been described by other authors (e.g., Odontomachus assiniensis: Ledoux, 1952; Odontomachus bauri: Jaffe & Marcuse, 1983; O. chelifer: Fowler, 1980; Odontomachus troglodytes: Déjean, 1982, 1987; Déjean & Bashingwa, 1985). Déjean (1987) found that workers of O. troglodytes learn to avoid noxious prey.

Rapid mandibular strikes are used by Odontomachus to perform a variety of specialized tasks in addition to prey capture. Patek et al. (2006) found that workers of O. bauri utilize the force of their mandible strikes to bounce to safety (or to bounce onto intruders), and also to eject intruders away. This latter behavior (the bouncer defense) was studied in detail in O. ruginodis by Carlin & Gladstein (1989). In addition to these highly specialized tasks, the mandibles of Odontomachus remain functional for more typical activities such as nest construction and brood care (Just & Gronenberg, 1999).

Nomenclature

 *  ODONTOMACHUS [Ponerinae: Ponerini]
 * Odontomachus Latreille, 1804: 179. Type-species: Formica haematoda, by monotypy.
 * Odontomachus senior synonym of Pedetes: Dalla Torre, 1893: 51.
 * Odontomachus senior synonym of Champsomyrmex, Myrtoteras: Brown, 1976a: 67.
 * CHAMPSOMYRMEX [junior synonym of Odontomachus]
 * Champsomyrmex Emery, 1892d: 558 (footnote). Type-species: Odontomachus coquereli, by monotypy.
 * Champsomyrmex subgenus of Odontomachus: Forel, 1893a: 163; Forel, 1917: 238.
 * Champsomyrmex as genus: Emery, 1892d: 558 (footnote); Emery, 1911d: 111; Wheeler, W.M. 1922a: 653.
 * Champsomyrmex junior synonym of Odontomachus; Brown, 1976a: 96.
 * [ Thempsomyrmex Forel, 1893a: 163, incorrect subsequent spelling.]
 * MYRTOTERAS [junior synonym of Odontomachus]
 * Myrtoteras Matsumura, 1912: 191. Type-species: Myrtoteras kuroiwae, by monotypy.
 * Myrtoteras junior synonym of Odontomachus: Brown, 1976a: 96.
 * PEDETES [junior synonym of Odontomachus]
 * Pedetes Bernstein, 1861: 7. Type-species: Pedetes macrorhynchus, by monotypy.
 * Pedetes junior synonym of Odontomachus: Dalla Torre, 1893: 51.

Worker
Schmidt and Shattuck (2014) - Medium to large (TL 6–20 mm; Brown, 1976) slender ants with the standard characters of Ponerini. Mandibles straight and narrow, articulating with the head medially, capable of being held open at 180°, and with a trio of large apical teeth and often a row of smaller teeth along the masticatory margin. Head with a pair of long trigger setae below the mandibles. Clypeus truncate laterally and anteriorly. Frontal lobes small and relatively widely spaced. Head strangely shaped: much longer than wide, with a distinct constriction behind the eyes and then often a gradual broadening posteriorly, the posterior margin of the head straight or mildly concave, the nuchal carina V-shaped medially, the posterior surface of the head with a pair of dark converging apophyseal lines. Eyes fairly large, located anterior of head midline on temporal prominences. Metanotal groove shallowly to deeply impressed. Propodeum broadly rounded dorsally, as broad as mesonotum but narrower than pronotum. Propodeal spiracles small, circular to ovoid. Metatibial spur formula (1s, 1p). Petiole surmounted by a conical node, topped by a posteriorly-directed spine of variable length. Gaster without a girdling constriction between pre- and postsclerites of A4. Stridulitrum almost always present on pretergite of A4. Head and body shiny to lightly striate, with very sparse pilosity and pubescence. Color variable, orange to black.

Queen
Brown (1976) - Like worker, but differing in the characters usual for ponerines; in most species normally alate when virgin. Ocelli present, small, placed near the middle of the dorsum of the head. Head usually relatively broader, mandibles and scapes shorter, petiolar node broader and more compressed axially, than in workers of the same species, and trunk with the usual development of flight sclerites; gaster modestly more voluminous. The overall size difference between queen and corresponding workers is usually not very great, making the queen often hard to spot among the· circulating workers when a nest is newly opened.

Wing venation in both wings of the complete ponerine pattern, without any trace of 1r in forewing, but Mf2 most often present and uncontracted. Anal lobe of hind wing usually present; may be lost in some of the smaller Anochetus species.

Male
Brown (1976) - Typically euponerine in habitus as well as the definitive characters: mandibles reduced, non-opposable; genital capsule not fully retractile; cerci well developed and conspicuous. In fact, I can find no character or combination of characters in this sex that will unequivocally separate Odontomachiti from Poneriti, or even that will separate males of Odontomachus or Anochetus from those of a pone rite genus such as Pachycondyla s. lat. Even the downcurved pygidial spine is shared by males of the more primitive members of both tribes, as well as some species of Platythyrea (Platythyreini).

Odontomachite males are usually told, in my experience, by this generally ponerite habitus and character set, plus the particular shape of the petiolar node, which usually reflects in a more or less muted fashion the shape of the node in the corresponding workers and queens. Thus, the node in Odontomachus males is more or less conical, with a narrowly rounded or even pointed apex; in Anochetus, the male node varies greatly with the species, so that in A. mayri, for example, it is axially compressed and squamiform, with its sharp crest forming two points separated by a deep emargination.

Size varying from 3 or 4 to more than 15 mm total body length, with correspondingly well developed wings. Head broader than long, with half to just about all of the sides of the head occupied by large compound eyes, which usually have weakly emarginate mesal margins in front (dorsal) view, and also weakly emarginate lateral margins as seen from the side. Clypeus small, convex, with convex or nearly straight anterior free margin. Ocelli large, close together, usually situated on an eminence. Mandibles small, narrow, with acute to bluntly rounded apices; at base with a dorsal whitish membranous basin, possibly at least partly the same as Ettershank's (1966) mandallls, which may possibly serve to dispense pheromones during mating flights. Distad, a curved dorsal ridge separates the concave outer face from the mainly convex dorsomedial face of the mandible. One or more stout, tapered, porrect setae issue from near the apex of each mandible.

Labrum small, bilobed, with a median cleft, fitting into the space between the mandibles. Maxillary palpi 6-, 5-, or 4-merous, labial palpi 4- or 3-merous, so far as I have been able to count. Some small Anochetus may possibly have lower counts, such as 3, 2, but I have not confirmed any counts this low.

Antennae always 13-merous, with short, cylindrical scape and still shorter pedicel; flagellar segments beyond pedicel longer and slender cylindrical, arranged so that the antennae usually curve circularly up and then downward in dead specimens, as in most ponerine males.

Trunk rather robust, scutum with notauli forming a distinct, impressed Y in some Anochetus, an indistinct V in others, and a generally very indistinct or obsolete V in Odontomachus.

The unconstricted gaster of most odontomachite workers and queens is usually a feature also of the conspecific males, and we may expect that the gaster of males in such species as A. inca has the constriction between the first and second gastric segments wide and deep, as in the worker of that species.

Pygidium produced as a sharp, slender downcurved spine in all known Odontomachus males and in some Old World species of Anochetus; in other Anochetus, the pygidium has the free margin with a median crease and point, or is entire and rounded. Hypopygium linguiform, with parallel sides and rounded or subtruncate apex in Odontomachus, but variously formed in Anochetus: as a rounded, convex plate; produced as a single setose rod-like piece; produced as paired slender rods; as an emarginate plate, etc.

Genitalia proper much like those of Poneriti, relatively complete and conservative in Odontomachus, but modified in various striking ways in the few Anochetus species available.

Legs slender; tarsal claws slender, with or without a small submedian tooth. Tibial spurs 1, 2, 2 or 1, 1, 1, paralleling their development in workers and queens of the same species.

Pilosity simple and fine, usually rather sparse; pubescence sparse to dense, usually suberect to decumbent.

Head, petiole and gaster usually smooth or nearly so, though often finely and rather densely punctulate. Trunk usually largely rugulose, rugulose-striate, or densely punctulate in Odontomachus, finely and more or less densely punctulate in Anochetus; in both genera the pleural areas are smoother and often more shining than dorsal sclerites.

Color ranging from testaceous to black; in Anochetus, head sometimes much darker than remainder of body; scutum sometimes with a fuscous pattern consisting of a pair of broad longitudinal bands connected in part by a median element.

Larva
Brown (1976) - (characterization based on that of G. C. and J. Wheeler, 1971: 1213): Body profile like that of Pachycondyla, i. e., neck long and slender ; head large ; abdomen subovoidal, but with the ventral profile nearly straight; anus ventral. Beset with numerous (80-116) tubercles, but none on midventral surface. A typical tubercle consists of a frustum bearing a circle of 4-6 relatively long, slender hairs that are constricted at the point of attachment; seated on this frustum is a spire that bears at its apex a heavy, straight, spine-like hair; integument of spire with short transverse rows of spinules. On the mid-dorsal surface of abdominal segment IV there is usually one or a pair of modified holdfast tubercles in form ranging from almost flush, glabrous areas to low pulley-like, bulbous, or even nearly doorknob-shaped projections, with a sparse fringe of short hairs. Similar structures may be found also on segment V, or may be lacking there; in O. tyrannicus, the Wheelers (1952: 652) found that the dorsal holdfast tubercles of segments IV and V were replaced by ordinary seta-tipped tubercles like those on the rest of the body.

Head hairs few and short; mouthparts rather large. Mandibles like those of Ectatomma or Odontoponera: moderately broad at base, tapering apicad, and weakly to strongly curved, with an acute apical tooth and a medial blade carrying 2 more teeth; on a part of the anterior surface are a few minute spinules in short, transverse rows.

The available information on the morphology and bionomics of odontomachite larvae is summarized by G. C. and J. Wheeler (1952: 642-652, pl. 5, fig. 1-17, pl. 6, fig. 22-29) and added to by them in 1964: 455-456 and 1971: 1212-1213, fig. 25-26.

In view of the present revision some of the old identifications listed by the Wheelers require correction or amplification:

Anochetus sp., [Mt.] Tobang, Borneo, 1300 m, E. Mjoberg, is A. princeps.

Odontomachus haematoda specimens listed from Costa Rica and Jamaica and figured in Plate 6, fig. 22-29, are very likely O. bauri; the samples from Ceylon that Clausen found parasitized by chalcidoids belonged to O. simillimus, and the specimens mentioned by Eidmann and by Emery could have been haematodus, bauri, or some other species. «Odontomachus haematoda clarus» is O. clarus; the species investigated by Haskins and Haskins is O. brunneus; the Cuban var. pallens is O. insularis, and the var. bruneipes discussed by Eidmann probably was O. bauri. The O. biolleyi larvae described by the Wheelers (1952: 651) should be questioned, because the only adults of this species I know to exist in collections are the types. O. ruficeps coriaria and O. cephalotes of the Wheelers are O. ruticeps, and their O. haematoda insularis is O. brunneus.

The larval characters of the Odontomachiti are in my opinion worth only generic rank within the Ponerini. The tubercle form and the hold fast structures of abdominal segments IV and V are the best characters, but the holdfasts are not universal in the subtribe (O. tyrannicus lacks them), and the form of the ordinary tubercles is not really strikingly different from that of some Pachycondyla (s. lat.) species, or species currently placed in Myopias.

Pupae
Brown (1976) - Normally enclosed in cocoons; I know of no exceptions. The silk is light to dark brown in color in the finished cocoon. As usual for ponerines, the spinning larvae require some sand or loose dirt as a substrate in order to complete the cocoon normally.