Metapone

The genus Metapone Forel remains one of the most unusual and enigmatic ant genera in the Old World. Most of the 17 described taxa are known from very few specimens, often single queens or males collected during dispersal flights and occasionally workers collected from dead wood. Forel (1911) established Metapone based upon a series of workers, larvae and reproductive pupae of the type species Metapone greeni Forel, collected from Peradenyia, Sri Lanka. In the same paper, Forel designated Metapone as the type genus of a new tribe Metaponini and provisionally placed this tribe into a new special section among the Ponerinae, which he called the Promyrmicinae.

Metapone was properly transferred to the subfamily Myrmicinae a year later by Emery (1912) when he realized that the larvae were characteristic of the Myrmicinae and not the Ponerinae. Unfortunately, Emery misinterpreted certain ponerine-like morphological traits of Metapone as primitive rather than as derived adaptations to a highly specialized habit of foraging in galleries of wood. As a result, Emery retained Forel’s section Promyrmicinae, transferred it to the Myrmicinae, and expanded it to include the tribes Metaponini and Pseudomyrmicini. Wheeler (1919) was able to clear up much of the earlier confusion after additional species of Metapone became available for study. Wheeler rejected the term Promyrmicinae even though he could not determine the position of the Metaponini among the other tribes within the Myrmicinae. The genera Metapone, Liomyrmex, Vollenhovia and Xenomyrmex were included until recently in the tribe Metaponini (Bolton, 1994, 1995). Bolton (2003) later considered Metapone to be the only genus in tribe Metaponini. The placement of Metapone within the Myrmicinae (tribe Crematogastrini) has been confirmed by molecular data (Ward et al. 2015).

Metapone is widely distributed throughout the Indo-Australian, Oriental and Malagasy Regions (Taylor, 1991; Bolton, 1995; Eguchi, 1998; Alpert, 2007). Even though specimens are rare, new species continue to be discovered. Eguchi (1998) described a new species from Borneo, an undescribed Metapone has been found in Gabon (B. Bolton, B. Fisher, pers. comm.) and another undescribed Metapone was found recently in New Caledonia (C. Burwell, pers. comm.) and in the Philippines (Gary Alpert, pers. comm.). There are several series of undescribed Metapone from Papua New Guinea and Australia in the. A species of Metapone was recently collected in South Africa (Brigitte Braschler ).

Identification
Metapone has the following derived characters (Bolton 2003):
 * Procoxae smaller than meso- and metacoxae
 * Metafemur highly anteroposteriorly compressed, extremely deep in anterior view
 * Apex of mesotibia, metatibia and basitarsi of all legs with traction spines
 * Presclerites of abdominal segment IV very large
 * Articulation of abdominal segments III and IV very broad

Distribution
Taylor and Alpert (2016) - Metapone is now known to range at least from Gabon and Madagascar, eastwards to Sri Lanka and the Nicobar Islands, Sumatra, Borneo, Java, Bali and Lombok, and from Taiwan, south through the Philippines and Micronesia to New Ireland, New Guinea and New Britain, thence eastwards to the Solomon Islands and Fiji, and southwards through eastern Australia, at least to the vicinity of Adelaide, South Australia, and Flinders Island, Tasmania. We have seen specimens from New Caledonia. There are no known records from sub-continental India, Western Australia, mainland Southeast Asia, China, Japan or New Zealand. One-quarter of the known taxa are Australian and the highest known species-density is in tropical north Queensland.

Biology
From the very first collections, Metapone workers have been found in association with termites. The type species of the genus, Metapone greeni, was found in “galleries in a decayed branch, which was also infested by two species of termites” (Forel, 1911, quoting E.E. Green, the collector). Subsequently, Wheeler (1919, 1936) listed M. greeni as an inquiline of termites and suggested that all species of Metapone probably form small colonies and live in or near the galleries of termites in dead wood. Taylor (1991) has found several undescribed species of Metapone in association with termites in Australia. In Papua New Guinea, Metapone were found in rotten wood with termites of the genus Prorhinotermes (Leigh Miller, personal communication). Eguchi (1998) reported that Metapone quadridentata from Borneo “was collected from the galleries of a termite nest in rotten wood”. Gregg (1958) obtained from Alfred E. Emerson, the noted termite specialist, two new species of Metapone (Metapone emersoni and Metapone madagascarica) from a series of termites collected from Madagascar by Harold Kirby in 1935. Additional specimens of M. madagascarica were collected in association with Coptotermes truncatus and Cryptotermes sp. (Brian Fisher, written communication). Though the evidence is anecdotal, the pattern of obligatory association with termites is clear.



Taylor and Alpert (2016) - Metapone is distinctive largely due to features related to adaptation for life associated with termite hosts. Its species seem universally to be termite predators and all reported nests were located in logs or rotting wood, usually within, adjacent to, or contiguous with termite nests. Liomyrmex is also termitophilous (Rigato & Bolton, 2001). Wheeler (1914) reported Liomyrmex gestroi ( as Liomyrmex aurianus Emery) “living in entire amity” with its hosts in nests of the common southeast Asian/Philippines mound-building termite Macrotermes gilvus (Hagen). Specialized association with termites has never been reported for Podomyrma or Huberia, nor has it been observed by the senior author in many field encounters with both genera.

Association between Metapone and termites has frequently been observed by collectors and authors and was first reported with the original description of Metapone greeni, when Forel (1911) quoted Sri-Lankan field notes by the first known collector, M.E. Green. Alpert (2007) reported Metapone associations with Cryptotermes kirbyi (Kalotermitidae) and Coptotermes truncatus Wasmann (Rhinotermitidae) on Madagascar and Prorhinotermes (Rhinotermitidae) in Australia. A colony of the Australian Metapone tillyardi is reported below to have been collected in joint association with species of both Kalotermes and Glyptotermes (both Kalotermitidae) in a diseased living willow trunk (termite identifications by J.A.L. Watson). Two species of termites were also involved at the initial collection of Metapone greeni (Wheeler, 1919b).

It is not known (1) whether any individual Metapone species are obligate associates of particular termite species, let alone involved in more catholic associations at multispecies, genus or higher taxonomic level; (2) whether there is size correlation between Metapone species and their termite hosts; (3) whether the constructed nests of mound-building termites or subterranean-nesting grass-eating termites are ever parasitized by Metapone (though see below under Metapone tricolor); or (4) whether the nests of alien, introduced pestiferous termites can be parasitized by local resident Metapone species.

In the absence of known relations between particular Metapone species and individual termite taxa distributional correlations can sometimes imply possible associations, notably in the well-documented Australian termite fauna, especially where there are few candidate termite species known from Metapone collection sites, as discussed below under M. tricolor and Metapone mathinnae.

Putative associations involving the termite families Termopsidae, Kalotermitidae, Rhinotermitidae and Termitidae are cited in these pages, excluding only the northern Australian endemic family Mastotermitidae (namely, its sole extant species Mastotermes darwiniensis Froggatt).

It is notable that several of the Metapone worker or gyne specimens examined here have characteristically damaged antennae or limbs. Sections of legs or apical portions of funiculi or scapes + funiculi appear to have been cleanly severed and subsequently to have healed in a manner very seldom observed in any other ants we have studied. We presume such damage to be the result of termite mandibular attack.

The biology of several Madagascan species was reviewed by Hölldobler, Leibig et.al (2002) and Hölldobler, Oldham et.al (2002). They confirm the role of Metapone species as specialist predators nesting in association with termites, and demonstrate the presence of gamergates (mated reproductive workers) in colonies of Metapone madagascarica and Metapone vincimus, in which all investigated workers were found to have functional ovarioles and spermathecae. Some individuals had previously mated, as evidenced by sperm in their spermathecae. No morphological differences between mated and non-mated workers were noted. Field observations and laboratory studies indicated that queenless colonies of these species containing gamergates are almost certainly able to reproduce. Also, prey termites stung by Metapone workers became paralysed and remained alive for extended periods while stored in the ant nests. Workers deposited chemical trails using one of seven compounds identified in the venom.

Alpert (2007) recognised worker-queen intermorphs and ergatoid males present in low numbers in M. madagascarica nests which also contained normal members of both reproductive castes. He remarked that “The presence of both winged and ergatoid males in the same colony is exceptional in ants and warrants further study”. We suggest that the ergatoids could possibly be males of a second species inquiline in M. madagascarica nests.

Metapone collections sometimes include long series of alate gynes, usually taken at flight intercept traps. We assume that they were engaged at collection in mating-flight or post-mating dispersal activity. The problems faced by newly-fertilised females in locating suitable host termite nests for colony foundation might have been adaptively addressed in part by the seasonal release of very large numbers from parental colonies, contributing to the frequency of such collections. Specimen labels indicate that alates of both sexes are attracted to light. Gynes have been collected from pitfall traps, presumably indicating that their post-flight search for termite host colonies might be partly pedestrian.

Workers have never been reported foraging epigaeically, no doubt because they seldom if ever leave the confines of their own nests or those of host termites, and are almost always encountered only when such nests have been breached by happenstance or by entomological collectors. Records of workers or dealate gynes are unknown from Berlese funnels, Winkler extractors or other soil/litter sampling devices.

Metapone nests are rarely encountered by myrmecologists. In a lifetime tally of almost two years of accumulated daily field work in north Queensland, New Guinea, West Malaysia and Borneo, featuring dissection of many rotting logs, the senior author encountered only four colonies: Metapone truki in the Gogol Valley, Papua New Guinea; Metapone hoelldobleri and Metapone mjobergi (twice) in tropical north Queensland—the last three records during 11 days of intense ant collecting in 1977 in company with former CSIRO colleague John Feehan.

Castes
All species of Metapone are known to have winged queens and workers. Metapone madagascarica is a highly unusual species in having both normal winged males and worker-like ergatoid males without wings. Both forms have been found together in the same nest.

Morphology
Bolton (2003:71) cites the following diagnostic apomorphies for the genus within subfamily Myrmicinae: (1) Procoxa smaller than mesocoxa and metacoxa; (2) Metafemur extremely anteroposteriorly compressed, extremely deep in anterior view; (3) metatibia and basitarsi of all legs with traction spines; (4) presclerites of abdominal segment IV very large; and (5) articulation of abdominal segments III and IV very broad.

Nomenclature

 *  METAPONE [Myrmicinae: Metaponini]
 * Metapone Forel, 1911h: 447. Type-species: Metapone greeni, by monotypy.