Thaumatomyrmex Mayr

Thaumatomyrmex Mayr Fig. 33 Thaumatomyrmex Mayr, 1887: 530 (as genus). Type-species: Thaumatomyrmex mutilatus Mayr, 1887: 531; by monotypy. Thaumatomyrmex is a small (12 described species) Neotropical ponerine genus. These ants are notable for their pitchfork-like mandibles and other unusual cephalic characters and their specialized predation on polyxenid millipedes. Formerly considered a tribe separate from Ponerini, we confirm here their phylogenetic placement within the Pachycondyla group of Ponerini. Diagnosis. Thaumatomyrmex workers are among the most morphologically derived of all ponerines, and would be difficult to confuse with those of any other genus. Their pitchfork-like mandibles and widely separated frontal lobes are autapomorphic within Ponerini and immediately identify them as Thaumatomyrmex. Belonopelta and Emeryopone also have mandibles with long attenuated teeth, but their teeth are shorter than those of Thaumatomyrmex and their frontal lobes are closely approximated as is typical for Ponerini. Synoptic description. Worker. Small (TL 3.3–5.0 mm; Kempf, 1975) ants with the standard characters of Ponerini except that the antennal sockets are very widely separated by a broad posterior extension of the clypeus. Mandibles pitchfork-like with three very long and attenuated teeth, the mandibular articulations located on narrow anterolateral projections of the head. Clypeus generally greatly reduced except for a broad posterior extension. Frontal lobes of moderate size, semi-vertical, reaching or surpassing the anterior clypeal margin. Eyes large and very convex, located anterior of head midline. Metanotal groove absent to shallowly impressed. Propodeal dorsum moderately narrowed but rounded. Propodeal spiracles round. Metapleural gland orifice with a U-shaped cuticular flange posteriorly and a shallow groove laterally. Metatibial spur formula (1p). Petiole ranging from a thick broad scale with sharp lateral margins to a cuboidal node. Gaster with only a weak constriction between pre- and postsclerites of A4. Pretergite of A4 with a distinct stridulitrum. Head and body with variable sculpturing, ranging from smooth and shiny to finely shagreened to finely punctate and rugulose. Head and body with scattered pilosity and no pubescence. Color black. Queen. Kempf (1975) mentioned the existence of an alate queen of T. zeteki (= T. atrox), but it has neither been described nor confirmed. Gamergates occur in at least two species (Jahyny et al., 2002). Male. See description by Kempf (1975). Larva. Discussed in Kempf (1975) and described for T. mutilatus by Kempf (1954) and Wheeler & Wheeler (1964). Geographic distribution. Thaumatomyrmex is a strictly Neotropical genus whose known range extends from Mexico to Brazil on the mainland and also includes some islands of the Caribbean (Kempf, 1975). Ecology and behavior. Thaumatomyrmex displays an unusual suite of morphological, ecological and behavioral traits. Brandão et al. (1991) examined the feeding habits of T. atrox and T. contumax and found that they are highly specialized predators of polyxenid millipedes (confirmed by Delabie et al., 2000; see also the account in Hölldobler & Wilson, 1995). Polyxenids are covered with protective hooked setae which hunting Thaumatomyrmex workers must deal with before consuming their prey. The ants deal with the polyxenids by grasping them with their pitchfork mandibles, stinging them (presumably to minimize defensive struggles), and finally scraping off the defensive setae using their modified front tarsi, rendering the millipedes palatable (Brandão et al., 1991). Given the highly specialized mandibular structure present in all Thaumatomyrmex species and the observation of millipede predation in two different species groups, polyxenid predation is probably universal in the genus. Thaumatomyrmex workers forage individually in leaf litter and feign death when disturbed (Brandão et al., 1991). Though Thaumatomyrmex were long considered to be rare (Longino, 1988), improved sampling methods have demonstrated that their colony density can be very high (Delabie et al., 2000). Given the cryptobiotic foraging habits of Thaumatomyrmex, the function of the large well-developed eyes in the workers is a mystery (Baroni Urbani & de Andrade, 2003). Thaumatomyrmex nests have been observed under bark, in rotting wood, under leaves, and in abandoned wasp nests (Kempf, 1975; Brandão et al., 1991; Delabie et al., 2000). Jahyny et al. (2002) studied the reproductive system of two species, T. atrox and T. contumax, and found that they reproduce via gamergates and that their colonies are exceptionally small (fewer than five workers, on average, and never more than nine). Kempf (1975) also reported a small colony size for T. mutilates and the existence of an alate queen of T. zeteki (= T. atrox). Delabie et al. (2000) documented aggressive interactions between a putative gamergate and her nestmates in a colony of T. contumax. Phylogenetic and taxonomic considerations. Mayr (1888) erected Thaumatomyrmex to house the single species T. mutilatus Mayr. Several additional species were subsequently described. Given the bizarre morphological traits of these ants, there has never been any doubt that Thaumatomyrmex constitutes a valid genus. There has been uncertainty, however, about its higher taxonomic placement, with authors variously placing it in Ectatommini (e.g., Emery, 1895d), Ponerini (e.g., Forel, 1895), Cylindromyrmicini (Ashmead, 1905), or most commonly and most recently in its own tribe, Thaumatomyrmecini (e.g., Emery, 1901; also sometimes spelled Thaumatomyrmii). Its close relationship with Ponerini was confirmed by the discovery of the first male specimen (Kempf, 1954), which had vestigial mandibles as in members of Ponerini. Bolton (2003) continued to treat it as a tribe separate from Ponerini, given its lack of supposed apomorphies of Ponerini, though he suggested that the vestigial male mandibles were likely synapomorphic for the two tribes. In the molecular phylogeny of Formicidae published by Brady et al. (2006), Thaumatomyrmex was inferred to be nested within a non-monophyletic Ponerini. Schmidt's (2013) molecular phylogeny of Ponerinae confirms this placement, with Thaumatomyrmex resolved as sister either to Simopelta or more likely to the remainder of the Pachycondyla group minus Simopelta; a sister relationship to Ponerini was statistically rejected. These results show that Thaumatomyrmex is simply a highly aberrant member of tribe Ponerini. The placement of the genus in its own tribe outside Ponerini (e.g., Bolton, 2003) was always suspect, as the apomorphies of Ponerini which are absent in Thaumatomyrmex (close approximation of the frontal lobes and hence only a narrow posterior extension of the clypeus) are some of the very characters which are so highly derived in Thaumatomyrmex (along with the mandibles, which are poor phylogenetic markers). In all other respects these ants are morphologically fairly typical for Ponerini. Given the unequivocal results from molecular phylogenetic analyses and the above morphological consideration, we are synonymizing Thaumatomyrmecini under Ponerini.Published as part of 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, pp. 1-242 in Zootaxa 3817 (1) on pages 159-161, DOI: 10.11646/zootaxa.3817.1.