Typhlomyrmex Mayr

Lacau, S., Villemant, C. & Delabie, J. H. C., 2004, Typhlomyrmex meire, a remarkable new species endemic to Southern Bahia, Brazil (Formicidae: Ectatomminae)., Zootaxa 678, pp. 1-23 : 1-3

[[ Genus Typhlomyrmex Mayr ]]

Typhlomyrmex Mayr, 1862 (Formicidae: Ectatomminae : Typhlomyrmecini ) is a small genus of Neotropical ants, which included six valid species and one nomen nudum prior to this report (Brown, 1965; Kempf, 1972; Brandao 1991; Bolton, 1995 and 2003). It is today the single genus of the tribe Typhlomyrmecini Emery, 1911, since Prionopelta Mayr, 1866 and Rhopalopone Emery, 1897 (junior synonym of Gnamptogenys ) have been combined in the Amblyoponini (Brown, 1960) and Ectatommini (Brown, 1958), respectively. In his Typhlomyrmex revision, Brown (1965) described a new species, T. prolatus , and suggested that the taxon pusillus certainly included several species. The genus is morphologically homogenous, but there is interspecific variation in petiole shape, head shape, and body size (Brown, 1965; Lacau et al., unpublished data). The terricolous species are the smallest. Bolton (2003: 46) proposed the following potential autapomorphies for the genus: "workers eyes vestigial to absent," and "male hypopygium with an elongate upcurved and median digitiform process." The first character has been reported as homoplasic for the entire family by Baroni Urbani et al. (1992): the eye reduction occurs in many other ant genera with hypogaeic habits, in several subfamilies (Hoelldobler & Wilson, 1990). All Typhlomyrmex species share a cryptic ecology and reduced eyes, but it is unknown if this character represents an autapomorphy for the genus in the clade Ectatomminae and/or a convergent adaptation between different Typhlomyrmex species. The digitiform process on the male hypopygium, proposed as autapomorphic by Bolton, is uniformly present in Typhlomyrmex but also occurs in some Gnamptogenys . The digitiform process is possibly homologous in the two taxa (Lacau, unpub.) and should be considered homoplasic. Thus, no clear autapomorphy distinguishes Typhlomyrmex from other ectatommine genera and its position in this subfamily remains unclear. According to Brown (1965), a range of characters position the Typhlomyrmecini nearer to the Amblyoponinae (formerly Amblyoponini ), while another brings them nearer the Ectatommini . The cladistic analyses of Lattke (1994) and Keller (2000) showed a close phylogenetic relationship between Typhlomyrmecini and Ectatommini , but these results cannot be corroborated as long as the monophyly of the Ponerinae sensu Bolton (1990) remains uncertain inside the poneroid group (Ward, 1994; Grimaldi et al, 1997; Keller, 2000; Ward & Brady, 2003; Lattke, 2003). Recently, Bolton (2003) reclassified the whole family Formicidae, combining Ectatommini and Typhlomyrmecini in the new subfamily Ectatomminae , which has been recently confirmed by Saux et al. (2004, in press) based on nuclear 28S rDNA sequences.

Compared to other poneromorph genera (sensu Bolton, 2003), Typhlomyrmex biology remains largely unknown but the rare available data reveal a strong ecological diversity between species. Although the distributions of individual species are highly variable, the genus has one of the largest distributions among the New World endemic poneromorph genera (Kempf, 1972). For example, the type species, T. rogenhoferi Mayr , 1862, which was described from Amazonas State (Brazil), is spread from northern Argentina to southern Mexico (Kempf, 1972; Longino, 1999; Lattke, 2003; Lacau et al, in progress). In contrast, others species have a much more reduced distribution and several are known only from a single locality. The new species described here is only found in a narrow part of the cocoa producing region of southern Bahia (Brazil). Typhlomyrmex prolatus Brown , 1965 and Typhlomyrmex foreli Santschi, 1924 are two other endemic species respectively described from San Jose (Costa Rica) and Rio Negro (Brazil, Parana State) regions (Brown, 1965). These restricted geographic ranges must be considered with circumspection owing to the scarcity and the disparity of the data concerning the diversity and the distribution of the Neotropical Formicidae according to the regions (Kempf, 1972). In fact, ants remain undersampled in large areas of South and Central America. Many species currently regarded as rare or endemic could be relatively common in poorly known regions. This is particularly true for Typhlomyrmex species because they are all cryptobiotic, nesting and foraging within soil or rotting wood. The autoecology of Typhlomyrmex species appears as variable as their distribution. Typhlomyrmex rogenhoferi is an epigaeic species that colonizes large dead trunks lying on the rain forest floor (Lacau et al., 2001), while the others species are subterranean and terricolous. In particular, the small colonies of T. pusillus nest in minute soil chambers (Brown, 1965; Lacau, pers. obs.); its biological cycle remains almost completely subterranean and only the winged gynes and males periodically emerge for mating.

The genus Typhlomyrmex is generally considered to be rare (Brown, 1965) because these ants are very difficult to find in the field. The terricolous species are rarely collected with extraction traps such as Winkler or Berlese-Tullgren eclectors because workers only occasionally forage in the litter (Ward, 2000; Lacau, pers. obs.). Also they are not found in woody macro-elements of litter, such as little branches or dried fruits, nor in logs (see Carvalho& Vasconcelos, 2002; Delabie et al. 1997; Lacau, pers. obs.). The best technique to find living colonies is to look for individuals by careful hand fragmentation of soil elements.

Such laborious field collecting explains the scarcity of Typhlomyrmex specimens in museum collections, especially when compared with those of other poneromorph genera such as Pachycondyla and Hypoponera (Lacau et al., unpublished data). Furthermore, the morphology of the different castes is rarely known. For example, T. prolatus and T. foreli are only known from their winged queen holotype. In the other taxa, the castes were frequently described separately, from material collected in different localities so that complete series including workers, queens and males coming from the same colony or even from a single location are extremely rare in collections (Lacau et al., unpublished data).