| Genus List |

The Genus Thaumatomyrmex in Costa Rica

The Neotropical ant genus Thaumatomyrmex is a myrmecologist's delight, being highly distinctive and rare. The mandibles are like pitch-forks, each mandible composed of three long tines joined at the base. Thaumatomyrmex workers are rarely encountered, and only a handful of specimens exist in the world's museums. They are most often collected as isolated workers in samples of leaf litter from the forest floor, extracted using Berlese funnels or Winkler sacks. Brandao et al. (1991) discovered that Thaumatomyrmex contumax and atrox in southern Brazil is a specialist predator on millipedes in the order Polyxenida. Polyxenid millipedes are covered with detachable barbed setae which entangle potential predators. The Thaumatomyrmex workers used their long, specialized mandibles to capture polyxenids and subsequently strip them of their setae.

Kempf (1975) last revised the genus. He recognized three species groups: (1) the mutilatus group, from extra-Amazonian Brazil, was characterized by a subopaque, silky surface sculpture, and a pair of closely-spaced setae on the clypeus; (2) the cochlearis group, endemic to Cuba, was characterized by rugose and punctate surface sculpture, and a high density of setae, particularly on the head; and (3) the ferox group, found from Amazonian Brazil northward through Central America, was characterized by a highly polished appearance, the entire surface being very smooth and shiny. In an attempt to identify Costa Rican specimens, I made some observations on the ferox group and attempted a revision (Longino 1988). Species distinctions had often been made on head shape, which varied from relatively quadrate, about as long as wide, to very broad, much wider than long. I examined a scatterplot of HW vs. HL and found a continuum, suggesting a single allometric relationship in which head shape changed with size (figure). The one discrete character I found was the presence or absence of a small basal tooth on the mandible, in addition to the three long tines. I proposed that the presence or absence of this tooth was a significant character for species recognition, and head shape was not. I recognized two species: (1) atrox had the basal tooth and occurred from Panama southward to Brazil and Bolivia, (2) ferox lacked the basal tooth, and occurred from Honduras south to Amazonian Brazil. All the specimens I had from Costa Rica lacked the basal tooth, and I identified them as ferox.

At the time of the revision there were no known cases of sympatry of multiple forms of Thaumatomyrmex. Soon after publication, however, I received a letter from Stefan Cover, who had just examined a new set of Thaumatomyrmex specimens from Barro Colorado Island, Panama. He had obtained 14 specimens from Sally Levings' 1976-78 ecological sampling. Stefan wrote "they neatly sorted into two species: a large one with a distinctly trapezoidal head, no prominent hairs on the petiole, and eyes with a sort of flat side to them, and a small one with a much less trapezoidal head, 8-14 prominent recurved hairs on the upper surface of the petiolar node, and eyes lacking a flat side... Both these 'forms' have the basal mandibular tooth." Subsequently, Brandao et al. (1991) reported a case of sympatry in Brazil. Most recently, the ALAS project at La Selva Biological Station in Costa Rica produced the first large-headed Thaumatomyrmex specimen from Costa Rica. In this case, the large-headed specimen has a basal mandibular tooth, while the more common small-headed specimens lack it. These new observations reveal that (1) multiple species of Thaumatomyrmex co-occur locally, and (2) size and head shape may be a significant species-level character after all. In fact, the plot of HW vs. HL suggests what may be a slightly offset line of allometry for the larger specimens (figure). This makes many of the conclusions in my 1988 revision suspect, and makes it clear that more work (and more material!) is needed.

For the time being I will use the system erected in my 1988 revision to identify the two Costa Rican species as ferox and atrox, but one should bear in mind that this nomenclature may not be stable, given the uncertainties alluded to above.

Thaumatomyrmex species known from Costa Rica

Thaumatomyrmex atrox Weber 1939

Thaumatomyrmex ferox Mann 1922

Literature Cited

Brandao, C. R. F., J. L. M. Diniz, and E. M. Tomotake. 1991. Thaumatomyrmex strips millipedes for prey, a novel predatory behaviour in ants and the first case of sympatry in the genus. Insectes Soc. 38:335-344.

Kempf, W. W. 1975. A revision of the Neotropical ponerine ant genus Thaumatomyrmex Mayr (Hymenoptera: Formicidae). Stud. Entomol. 18:95-126.

Longino, J. T. 1988. Notes on the taxonomy of the neotropical ant genus Thaumatomyrmex Mayr (Hymenoptera: Formicidae). In J. C. Trager (ed.), Advances in Myrmecology, pp. 35-42. E. J. Brill, New York.

Page author: John T. Longino longinoj@evergreen.edu

Date of this version: 10 October 2005.