Pseudoneoponera Donisthorpe

Schmidt, C. A. & Shattuck, S. O., 2014, The Higher Classification of the Ant Subfamily Ponerinae (Hymenoptera: Formicidae), with a Review of Ponerine Ecology and Behavior, Zootaxa 3817 (1), pp. 1-242 : 131-134

Pseudoneoponera Donisthorpe

Fig. 25

Pseudoneoponera Donisthorpe, 1943a: 439 (as genus). Type-species: Pseudoneoponera verecundae Donisthorpe, 1943a: 439 ; by original designation. Gen. rev.

Pseudoneoponera is a moderately large genus, with 18 described species, and ranges from India to Australia. Its sister group is unresolved. Pseudoneoponera is notable for the unusual foamy defensive secretions its workers exude from the sting, for its small colonies, and for its unusual social systems, including the frequent occurrence of gamergates.

Diagnosis. Workers of Pseudoneoponera are distinguished from other ponerines by their combination of robust build, coarse sculpturing, shaggy pilosity, obsolete metanotal groove, a nodiform petiole which is semicircular in dorsal view and often has a denticulate posterodorsal margin, longitudinally striate tergite A3 (rarely otherwise sculptured), and strong gastral constriction between A3 and A4. Pseudoneoponera most closely resembles Bothroponera and Phrynoponera , and to a lesser extent Ectomomyrmex , but these genera lack the shaggy pilosity, semicircular denticulate petiole, and longitudinally striate tergite A3 of Pseudoneoponera , and Pseudoneoponera lacks the spinose propodeum (except Ps. bispinosa ), five-spined petiolar node, and unconstricted gaster of Phrynoponera , and the small eyes, angular sides of the head, divided mesopleuron (except in a few species), and weakly constricted gaster of Ectomomyrmex .

Synoptic description. Worker. Medium to large (TL 9–18 mm) robust ants with the standard characters of Ponerini . Mandibles triangular and usually with a distinct basal groove. Eyes moderate to large in size, placed anterior of head midline. Mesopleuron usually not divided by a transverse groove. Metanotal groove absent. Propodeum broad dorsally, the posterodorsal margin usually unarmed (bispinose in P. bispinosa ). Propodeal spiracle slit-shaped. Metatibial spur formula (1s, 1p). Petiole surmounted by a wide node, which is roughly semicircular in dorsal view (the anterior face convex and the posterior face flat or concave), the posterodorsal margin often denticulate (rarely medially incised or trispinose). Gaster with a strong girdling constriction between pre- and postsclerites of A4. Head and body coarsely sculptured, the tergite of A3 usually deeply longitudinally striate. Head and body shaggy, clothed in dense pilosity, which is often reddish in color. Color ferrugineous to black.

Queen. Queens have been formally described only for P. sandakana ( Wheeler, 1919) , for which the worker caste is still undescribed. Queens have also been noted to occur in P. tridentata ( Sommer & Hölldobler, 1992) and an undescribed species from Java ( Ito, 1993). Gamergates are common in the genus, and the queen caste has apparently been completely lost in many species. From the description of P. sandakana , it appears that queens, when they exist, are similar to workers but are alate and have the typical modifications of the head and thorax found in other alate ponerine queens.

Male. See descriptions in Forel, 1900a, Wheeler (1919), and Donisthorpe (1943a).

Larva. Described by Wheeler & Wheeler (1971b, 1976).

Geographic distribution. Pseudoneoponera occurs from India through Southeast Asia to Australia, where it reaches its greatest species diversity. At least one species, P. rufipes , ranges all the way from India to Australia.

Ecology and behavior. Pseudoneoponera has received a fair bit of attention from researchers due to its unusual reproductive and social strategies. While most species have not been carefully surveyed, the queen caste has apparently been found in only a few species, while gamergates have been found in several species ( Monnin & Peeters, 2008). In at least two species, both queens and gamergates may cohabit a single colony (see below). Just how widespread reproduction by gamergates is in Pseudoneoponera is unknown, but it may be characteristic of much or most of the genus. Interestingly, queens are unknown from all Australian Pseudoneoponera species , having only been found in some Indonesian species. It is tempting to think that the complete loss of the queen caste may be synapomorphic for the Australian species, but too little is known about relationships within the genus to test this hypothesis.

Pseudoneoponera species studied to date all have very small colonies (roughly 10 to 20 workers; Ito, 1993; Peeters et al., 1991; Ito & Higashi, 1991; Higashi et al., 1994; Sommer et al., 1994). Peeters et al (1991) report that P. sublaevis workers forage individually and prey on insects, and Shivashankar et al. (1995) report that P. rufipes feeds on a wide diversity of arthropods and other invertebrates, but otherwise few direct observations of Pseudoneoponera feeding have been published. Presumably these ants are generalist predators and scavengers, like most ponerines.

An unusual characteristic of this genus is that workers produce a foamy thread-like defensive excretion from their venom glands. This has been observed in at least P. bispinosa , P. insularis , P. rufipes , P. sublaevis , and P. tridentata ( Bingham, 1903; Maschwitz et al., 1981; Peeters et al., 1991; Sommer et al., 1994). The foaming is made possible by the atrophication of the Dufour’s gland and the resulting mixing of venom gland proteins with air ( Maschwitz et al., 1981; Buschinger & Maschwitz, 1984). The report by Bingham (1903) that P. bispinosa and P. rufipes release the foam from their mouths is almost certainly a misinterpretation of the phenomenon, as suggested by Maschwitz et al. (1981); indeed, one of us (CS) personally witnessed a P. rufipes worker emit foam from the tip of its abdomen. Maschwitz et al. (1981) report that P. insularis and P. tridentata retain a normal sting response, in addition to the foaming mechanism, and that their sting is painful. They hypothesize (and provide supporting experimental evidence) that the foaming mechanism is more effective than the sting against small fast moving enemies such as other ants. The foam apparently acts as a physical obstacle requiring active cleaning rather than as a neurotoxin ( Buschinger & Maschwitz, 1984). A similar foamy secretion has been independently evolved by at least one species of Pachycondyla ( P. harpax ; Overal, 1987).

P. tridentata has an unusual social system in which colonies can have both multiple dealate queens and multiple gamergates, which compete with each other for reproductive dominance ( Sommer & Hölldobler, 1992). Some colonies lack queens, in which case reproduction is performed solely by gamergates. A large proportion of both queens and workers in a colony are mated (47–100% and 81–100%, respectively; Sommer et al., 1994), and multiple individuals can be reproductively active. Since so many individuals in a colony are mated, reproduction seems to be correlated with age and behavioral dominance rather than with mating status. Workers apparently mate with males produced by their own colony, while queens do not ( Sommer et al., 1994). Colony founding in P. tridentata may occur via several different mechanisms, including pleiometrosis and fission (Hölldobler et al., 1992; Sommer et al., 1994). Tandem recruitment is used during nest emigration ( Maschwitz et al., 1981). Workers of P. tridentata forage individually at night, and nesting generally occurs in the ground, though nests have also been found in trees ( Maschwitz et al., 1981; Sommer et al., 1994). Jessen & Maschwitz (1983) found that P. tridentata is endowed with a large number of abdominal glands whose functions are unknown.

Ito (1993, 1999) studied the reproductive strategy of an undescribed Pseudoneoponera species in Java. This species is unusual in that multiple mated workers may be present in a colony, but only the top-ranked worker is a gamergate ( Ito 1993). Queens do occur, but most colonies are apparently queenless and have around 10 workers, of which one or more are mated ( Ito 1993). The workers in a colony are ranked in a dominance hierarchy structured via frequent antagonistic interactions; low-ranking workers are the principal foragers ( Ito, 1993). Males attempt to mate with their nestmates but are usually rejected; mating only occurs with foreign males, and only when a gamergate is absent ( Ito, 1999).

P. sublaevis also reproduces via gamergates, but in yet another variation on the theme, only a single worker in the colony is mated and performs all reproduction for the colony ( Ito & Higashi, 1991; Peeters et al., 1991). The members of a colony are organized in a strict linear hierarchy determined by age and by ritualized dominance displays, and queens are apparently absent ( Higashi et al., 1994). This species occurs in Australian Eucalyptus forests, and nests in the ground (Peeters et al., 1991). Gamergates are also known to occur in P. porcata (Peeters, 1993) , but the details of its mating system are unknown.

Phylogenetic and taxonomic considerations. Pseudoneoponera was erected by Donisthorpe (1943a) to house his new species P. verecundae , known from a single male specimen. He believed it to be closely related to Neoponera , based on “various characters,” though he didn’t explain how or why he came to this conclusion in any additional detail, and he noted numerous differences between the genera. Donisthorpe also erected a new section of Ponerinae , Exeuponerinae, to house Pseudoneoponera , as its males have retractile genitalia and do not correspond to any of the sections erected by Emery (1911) based on male and larval characters. Wilson (1958c) found that the P. verecundae type was extremely similar to males tentatively associated with P. tridentata (then Bothroponera tridentata ), and synonymized Pseudoneoponera with Bothroponera . Pseudoneoponera later became a junior synonym of Pachycondyla along with Bothroponera ( Brown, 1973) .

Based on Wilson’s comparison of the P. verecundae type to males of “ Bothroponera ” tridentata , we believe that P. verecundae is a member of a cluster of species formerly considered to be in Bothroponera . These taxa, which form a geographically compact group, are characterized by their robust builds, coarse sculpturing, shaggy pilosity, obsolete metanotal groove, semicircular petiolar node, and longitudinally striate tergite A3. Based on both molecular and morphological evidence, we are removing this group of species to its own genus. As the type species of Bothroponera ( B. pumicosa ) does not belong to this cluster of species, Pseudoneoponera becomes the only available name.

Schmidt's (2013) molecular phylogeny places P. rufipes with strong support within the Odontomachus group but does not resolve its sister group. It is certainly not closely related to Pachycondyla or Neoponera (to which Donisthorpe thought it was related), but a sister relationship with Bothroponera or Phrynoponera cannot be rejected. Morphologically, Pseudoneoponera most closely resembles Bothroponera and Phrynoponera , and to a lesser extent Streblognathus , though there are no obvious synapomorphies linking these genera (see the discussions under Bothroponera and Phrynoponera for more). Even if Pseudoneoponera is found to be the sister to one or more of these genera, they are morphologically and behaviorally distinct enough and phylogenetically old enough to warrant separate generic status.

Figures

FIGURE 25. Worker caste of Pseudoneoponera oculata: lateral and dorsal view of body and full-face view of head (CASENT0172433, April Nobile and www.antweb.org); world distribution of Pseudoneoponera.