The Higher Classification of the Ant Subfamily Ponerinae (Hymenoptera: Formicidae), with a Review of Ponerine Ecology and Behavior Author Schmidt, C. A. Author Shattuck, S. O. text Zootaxa 2014 2014-06-18 3817 1 1 242 journal article 5350 10.11646/zootaxa.3817.1.1 d66f1b27-5891-4fa5-96e0-f75cb3ec2445 1175-5326 10086256 A3C10B34-7698-4C4D-94E5-DCF70B475603 Bothroponera Mayr Fig. 8 Bothroponera Mayr, 1862: 713 , 717 (as genus in Ponerinae [Poneridae]). Type-species: Ponera pumicosa Roger, 1860: 290 ; by subsequent designation of Emery, 1901: 42 . Gen. rev. FIGURE 8. Worker caste of Bothroponera cambouei : lateral and dorsal view of body and full-face view of head (CASENT0034474, April Nobile and www.antweb.org); world distribution of Bothroponera . Bothroponera is a moderately large genus, with 53 described species and subspecies. The sister group of Bothroponera is unresolved, and the genus may be non-monophyletic as defined here. Bothroponera ( sensu stricto ) is restricted to Africa and Madagascar , while the B. sulcata species group ranges from Africa to the Philippines . Relatively little is known about Bothroponera ecology and behavior. Diagnosis. Bothroponera workers lack any obvious autapomorphies within the Ponerinae , and are therefore more difficult to diagnose than those of most ponerine genera. They can generally be distinguished by the following combination of characters: Body without a long dense golden pilosity, mesopleuron usually not divided by a transverse groove, metanotal groove obsolete, propodeal dorsum without spines or teeth, propodeal spiracles slit-shaped, petiole nodiform (and not semicircular in top view) and without posterodorsal spines or teeth, tergite of A3 without strong longitudinal striations, gaster with a strong constriction between A3 and A4, and metatibiae with two spurs. Species of Bothroponera ( s.s. ) also have strong sculpturing, large cordate frontal lobes, a broad propodeal dorsum, and a U-shaped cuticular lip posterior to the metapleural gland orifice, though members of the B. sulcata group lack these characters. Superficially, Bothroponera workers most closely resemble those of Phrynoponera and Pseudoneoponera , but they lack the bispinose propodeum, five-spined petiole, and unconstricted gaster of Phrynoponera and the shaggy pilosity, semicircular petiole, and longitudinally striate tergite A3 of Pseudoneoponera . Bothroponera can also be confused with Ectomomyrmex , though Bothroponera lacks the small eyes, angular sides of the head, divided mesopleuron (except in a few species), and weakly constricted gaster of Ectomomyrmex , and Asian Bothroponera species lack the strong sculpturing of Ectomomyrmex . The workers of Bothroponera ( s.s. ) also somewhat resemble the workers of Loboponera and Boloponera , given their stocky build, coarse sculpturing, large frontal lobes, obsolete mesopropodeal suture, broad propodeal dorsum, and nodiform petioles, but are readily separated from these genera by their metatibial spur formulae (two spurs versus one), propodeal spiracles (slit-shaped versus round), and metapleural gland orifice (opening posterolaterally versus laterally), among other characters. Synoptic description. Worker. Medium to large (TL 5.5–16 mm ) slender to robust ants with the standard characters of Ponerini . Mandibles triangular, usually with a faint basal groove. Frontal lobes either moderately large ( B. sulcata group) or very large and cordate ( Bothroponera s.s. ). Eyes of moderate size and placed anterior of head midline. Mesopleuron usually not divided by a transverse groove. Metanotal groove obsolete dorsally. Propodeum broad dorsally. Propodeal spiracles slit-shaped. Distinct pale glandular patch on posterior surface of metatibia, close to spur, present ( B. sulcata group) or absent ( Bothroponera s.s. ). Metatibial spur formula (1s, 1p). Petiole nodiform, widening posteriorly in dorsal view, with vertical anterior and posterior faces. Girdling constriction between pre- and postsclerites of A4 apparent. Stridulitrum either present ( Bothroponera s.s. ) or absent ( B. sulcata group) on pretergite of A4. Head and body either coarsely sculptured with abundant pilosity and moderate pubescence (most Bothroponera s.s. ; B. laevissima is shiny, with only weak sculpturing and sparse pilosity) or finely punctate with scattered pilosity and dense pubescence ( B. sulcata group). Color dark brown to black. Queen. Similar to the worker but slightly larger, winged and with ocelli ( Wheeler, 1922b ). Queens are absent in B. kruegeri , in which reproduction is performed by gamergate workers. Male. See description in Wheeler (1922b) . Larva. Described by Wheeler & Wheeler (1952 , 1971b , 1976 ). Geographic distribution. Bothroponera ( sensu stricto ) is restricted to Sub-Saharan Africa and Madagascar , while the B. sulcata group occurs from Sub-Saharan Africa through southern Asia to the Philippines . References in the literature to Australian “ Bothroponera ” species (and many Asian species as well) are actually references to Pseudoneoponera , which we are reviving as a distinct genus. Ecology and behavior. Relatively little is known about Bothroponera , and all studies of the ecology and behavior of these ants have focused on species in the B. sulcata group, so even less is known about the habits of Bothroponera ( s.s. ), with most information coming from anecdotal observations. B. mlanjiensis is reported to dwell in shady forest habitats and to nest in the ground ( Arnold, 1946 ), collection data for several other species indicates that they also nest in the ground and are often collected in leaf litter, and B. pachyderma has been found nesting inside abandoned termitaries (Déjean et al. , 1996), but otherwise ecological notes on Bothroponera ( s.s. ) are sorely lacking. At least some species feign death when disturbed ( e.g., B. pachyderma ; Wheeler, 1922b ). B. pachyderma is reported to be a generalist predator (Déjean et al. , 1999), but the prey preferences of other species are unknown. Some larger species, such as B. pachyderma , stridulate audibly when distressed (B. Bolton , pers. comm.). Perhaps the best studied species of Bothroponera is B. tesseronoda , a South Asian member of the B. sulcata group. This species forms subterranean nests, with from 50 to 170 workers per colony ( Jessen & Maschwitz, 1986 ). B. tesseronoda is apparently a generalist predator of arthropods, with termites making up a large portion of its diet ( Shivashankar et al. , 1995 ), though it is also known to visit extrafloral nectaries ( Agarwal & Rastogi, 2008 ). Workers recruit nestmates to food sources and new nest sites via tandem running, which is initiated by mechanical stimulation and a colony-specific chemical trail derived from the entire body surface ( Maschwitz & Mühlenberg, 1973 ; Maschwitz et al. , 1974 ; Jessen & Maschwitz, 1985 and 1986; Maschwitz & Steghaus-Kovac, 1991 ). Nest entrances are marked with a colony-specific chemical cue, and foraging workers also employ chemical signaling while scouting a new area for food. Remarkably, individual workers can identify their own trails and show a preference for them over that of other individuals ( Jessen & Maschwitz, 1986 ). Alarm pheromones and defensive secretions are produced in the mandibular glands, poison gland, and Dufour’s gland ( Maschwitz et al. , 1974 ). An African species in the B. sulcata group, B. crassa , also uses tandem running for nestmate recruitment ( Hölldobler & Wilson, 1990 ). Workers of another African member of the B. sulcata group, B. soror , use tandem running to recruit nestmates to large prey items or to groups of prey, but are also capable of recruiting nestmates from up to 150 mm away using a multicomponent mandibular gland secretion ( Longhurst et al. , 1980 ; Déjean, 1991). Arnold (1915) reports that workers of B. soror smell strongly like cockroaches; the mandibular gland secretions are possibly responsible for this. B. soror is a generalist predator of arthropods and a scavenger ( Longhurst et al. , 1980 ; Déjean et al. , 1999). Déjean (1991) describes some behavioral adaptations of B. soror to termite predation, and Orivel & Déjean (2001) measured the toxicity of B. soror venom. Nests are typically located under stones or in termitaries (Déjean et al. , 1996, 1997), and colonies are small with about two dozen workers ( Arnold, 1915 ). Haskins (1941) found that B. soror queens display semi-claustral nest founding during which they forage outside the nest but also metabolize their flight muscles, and that they prefer to build nests in soil around rotting logs and even in the logs. Ground nesting is probably the rule for Bothroponera , though B. silvestrii has also been reported nesting in rotting wood ( Taylor, 2008 ). Queenless but apparently self-sustaining populations of B. soror have been observed ( Longhurst, 1977 ; Villet & Wildman, 1991 ). An unusual member of the B. sulcata group is the South African species B. kruegeri , which has lost the queen caste and reproduces only via gamergate workers ( Peeters & Crewe, 1986b ). Colonies each have only a single gamergate, which is the only inseminated individual in the colony and the only individual with mature ovaries. Gamergates suppress the ovarian development of the other workers. This is apparently accomplished through chemical means, as no physical interactions between gamergates and non-gamergates have been observed ( Wildman & Crewe, 1988 ; Villet & Wildman, 1991 ). Villet & Wildman (1991) examined division of labor in this species. In the population studied by Wildman & Crewe (1988) , colony size ranged from 8 to 100 workers (mean = 43). This species is a generalized predator of arthropods ( Wildman & Crewe, 1988 ) and nests in the ground ( Villet & Wildman, 1991 ). Discussions in the literature of foamy defensive secretions from the sting apparatus of Bothroponera ( e.g., Wheeler, 1922b ) apparently all refer to species of Pseudoneoponera , which we consider a distinct genus. True Bothroponera presumably lack these unusual secretions, though data are lacking either way. Phylogenetic and taxonomic considerations. Bothroponera was erected by Mayr (1862) to house the single species Ponera pumicosa Roger. Emery (1895d) treated it as a subgenus of Ponera , and later ( Emery, 1901 ) as a subgenus of Pachycondyla . Bingham (1903) revived Bothroponera to genus status, and subsequent authors variously treated it as a distinct genus or as a subgenus of Pachycondyla . Most authors since Hölldobler & Wilson (1990) have taken the latter approach (but see Tiwari, 1999 ). Wilson (1958c) placed Pseudoneoponera as a junior synonym of Bothroponera . The type species of Phrynoponera , Ph. gabonensis , was originally placed in Bothroponera but was moved to a separate genus by Wheeler (1920). Recently Joma & Mackay (2013) removed Bothroponera from synonymy with Pachycondyla , treating it as a valid genus. However, little justification was provided and they included species we consider as belonging to Bothroponera , Ectomomyrmex , Pseudoneoponera and Ophthalmopone . Additionally, only about half of the species we place in Bothroponera were transferred, the remainder being placed in Pachycondyla . While we agree that Bothroponera should be given full genus status, our broader analysis of the subfamily Ponerinae suggests that the previous concept of the genus, that which included Pseudoneoponera as a junior synonym as well as various species placed at one time or another in the genera, should be modified. We are restricting Bothroponera to a subset of the species previously considered as belonging to the genus. Our concept of Bothroponera is based on both morphological and molecular evidence. Morphologically, Bothroponera ( s.s. ) most closely resembles Phrynoponera and Pseudoneoponera , as they all generally have robust builds, strong sculpturing, and obsolete metanotal grooves. These characters have evolved on multiple occasions in other ponerine genera, however, and are therefore unlikely to be good phylogenetic markers. In addition, the B. sulcata species group of Bothroponera lacks the robust build and strong sculpturing (but see below). Schmidt's (2013) molecular phylogeny of the Ponerinae places Bothroponera with strong support within the Odontomachus group. Members of Bothroponera ( s.s. ) have a U-shaped cuticular lip at the posterior edge of the metapleural gland orifice, which is a possible synapomorphy of the Pachycondyla group but was apparently independently derived in Bothroponera (and also maybe in Diacamma ). The sister group of Bothroponera is unresolved, but it is not closely related to Pachycondyla . Though a sister group relationship with either Phrynoponera or Pseudoneoponera cannot be statistically rejected, we consider Bothroponera , Phrynoponera and Pseudoneoponera to be separate genera, given their morphological distinctiveness and phylogenetic age consistent with that of other ponerine genera. Emery (1911) divided the species of Bothroponera into four groups: the B. pumicosa , B. rufipes-piliventris , B. sulcata , and B. perroti groups. The B. rufipes-piliventris group is actually the genus Pseudoneoponera . Schmidt's (2013) molecular phylogeny indicates that the B. perroti group is simply a Malagasy clade nestled within a nonmonophyletic African B. pumicosa group. This group is Bothroponera ( sensu stricto ), as it includes the type species of the genus, B. pumicosa . Emery’s B. sulcata group is problematic, as there is no obvious morphological synapomorphy linking it to Bothroponera ( s.s. ), and molecular data are lacking for this group. The B. sulcata group may form a monophyletic sister clade, or even paraphyletic basal grade, with respect to Bothroponera ( s.s. ), or it may represent an entirely independent lineage. For now we are choosing to be conservative by retaining the B. sulcata group species within Bothroponera , with the explicit caveat that it may actually be a distinct lineage requiring a new genus name.