Trachymyrmex, Forel, 1893

Rabeling, Ch., Cover, S. P., Johnson, R. A. & Mueller, U. G., 2007, A review of the North American species of the fungus-gardening ant genus Trachymyrmex (Hymenoptera: Formicidae)., Zootaxa 1664, pp. 1-53 : 25-27

[[ Genus Trachymyrmex ]]

GENERAL DISCUSSION

The taxonomy of the genus Trachymyrmex is slowly being clarified ( Mayhé-Nunes & Brandão 2002, 2005, 2007), following years of confusion caused in part by insufficient collections of most taxa and by the failure to discover reliable species-level characters, a task made difficult by the notable variation in morphology and color seen within many Trachymyrmex species. The present study identifies nine Trachymyrmex species occurring in North America north of the Mexican border. Reliable delineation of species boundaries was possible due to the collection of large numbers of nest series in recent years by several collectors, including the authors. The new material has enabled us to re-evaluate the morphology of each species and to assemble an independent molecular dataset addressing between- and within-species variation.

The molecular data confirms the genetic distinctness of each of the nine species in the context of the North American fauna (Fig. 21). It also reveals that morphological similarity is not necessarily an indicator of a close genetic relationship. Of the Trachymyrmex covered in this study, there are two pairs of species ( carinatus / septentrionalis and pomonae / desertorum ) that show close overall morphological similarity. The molecular information reveals, however, that the species within each of the two pairs are not closely related.

We purposely analyzed our molecular data without choosing an outgroup for two reasons. Trachymyrmex may not be monophyletic (Schultz & Meier 1995), and our sampling is paraphyletic by definition because we studied only a fraction of the total number of species in the genus occurring at the northern limit of its distribution. A phylogenetic analysis of the tribe Attini based on several protein-coding genes by Schultz and coworkers(Schultz personal communication) suggests that Trachymyrmex is polyphyletic and paraphyletic with respect to Sericomyrmex . On the other hand Mayhé-Nunes and Brandão (2002) interpret the tubercular mesosomal projections as an autapomorphy of Trachymyrmex , suggesting monophyly of the genus. Our data cannot resolve these conflicting hypotheses; a more extensive study combining molecular and morphological data for a representative sampling of both Trachymyrmex and Sericomyrmex will be necessary to resolve this problem. North American Trachymyrmex are clearly thermophilic and are adapted to life in warm, arid climates. Seven of nine species occur in the deserts of southwest USA, including Arizona, New Mexico, Texas, and extending into northern Mexico. All these Trachymyrmex species occur in warm desert, desert-scrub, or seasonally dry mid-elevation habitats. Winters are short, mild, and usually dry. As one might expect, these Trachymyrmex show life-history features similar to those of many other dry climate ants. They construct moderately deep nests that buffer the ants and their gardens from surface temperature changes and the effects of prolonged dry spells. During periods of severe drought and low foraging success, these Trachymyrmex species often become inactive and seal their nests for extended periods. The ants also respond strongly to the return of favorable conditions, foraging and excavating vigorously while soil is moist. Lastly, mating flights are triggered by summer rains, when conditions become favorable for founding queens to start new nests. The two Trachymyrmex species found in the eastern US are also associated with warm, seasonally dry habitats. Trachymyrmex jamaicensis is a tropical species that extends into southern Florida from the Caribbean region. It occurs primarily in coastal scrub or hammock vegetation on soils derived from coralline rock. These soils are extremely well-drained and are very dry for much of the year. In this respect, they provide conditions broadly similar to those encountered by the Trachymyrmex in the southwest.

In T septentrionalis , however, we encounter something quite different. This ant is unique within the genus in that it is fully adapted for life in temperate climates. During the winter, the ants become completely dormant and brood production ceases. Likewise, the fungus garden assumes a dormant state and gradually decreases in size as the ants cease to add substrate. The ants may even collapse the gardens to a few fungus fragments that the ants hold in their mandibles and which are used as starter cultures for spring gardens. As far as we know, this phenomenon is unique among the Attini . Despite this unique response to cold conditions, T septentrionalis shows a strong preference for warm, well-drained soils and seasonally dry habitats; in this respect, T septentrionalis behaves like the desert-adapted Trachymyrmex discussed above.

Why are all North American Trachymyrmex associated with well-drained soils and relatively open habitats? This preference is clearly not characteristic of the genus as a whole, and appears to have been derived during the northward expansion of the genus throughout Mexico and into the US. In the Neotropics, in contrast, Trachymyrmex species reach their highest diversity in lowland rainforests with weak seasonality, relatively constant moisture, and water-retentive, clay soils (Rabeling, unpublished data). We offer the following tentative reflections on the biogeography of the North American Trachymyrmex species, and the evolutionary changes during the northward expansion from the Neotropics into subtropical and temperate zones:

1) If high temperatures are sufficiently stable for uninterrupted fungiculture and brood rearing, as is true for most of the lowland Neotropics, Trachymyrmex apparently have little difficulty coping with abundant soil moisture.

2) Trachymyrmex are successful in drier climates, as in the southwestern US, because they evolved behavioral adaptations to regulate the microclimate within their nest chambers (e.g. by placing chambers deep in the soil, as in T desertorum and T smithi ), thereby reducing mortality when soil moisture levels are low at the surface. It is even possible that the fungus cultivars evolved physiological adaptations for temperate conditions, like cold-hardiness or desiccation tolerance, enhancing survival during unfavorable seasons. Perhaps because of such adaptations, aridity has been no barrier to the radiation of the genus into subtropical and temperate regions, like northern Mexico and the southern United States.

3) As Trachymyrmex moved north and colonized areas with an ever more pronounced winter season, their preference for arid habitats and well-drained soils may have been an adaptive advantage, maximizing the length of the warm, active season while simultaneously minimizing the period during the year when soil temperatures are cool (or cold) and soil moisture is high.

4) Colonization of truly temperate areas with long, cold winters (e.g.: Illinois) was made possible by the evolution of the unique garden reduction seen in winter in the fungus gardens of T septentrionalis . This winter dormancy apparently serves to preserve the fungus culture when the ants are torpid themselves, and thus are unable to care for full-grown gardens. These behavioral adaptations of the ants and possible physiological adaptations of the cultivars during winter may enable the fungus to endure cold and wet conditions that might be lethal to actively growing gardens.

With this taxonomic revision, we hope to make the North American Trachymyrmex species accessible for further studies by other biologists. Experimental research testing the hypothesis of cold-hardiness and/or desiccation resistance of Trachymyrmex fungi would be particularly interesting areas of research; or to say it with the timeless words of W. M. Wheeler: "The study of the Attini […] has only just begun, and further advances in this fascinating subject will be more difficult for the mycologist than for the entomologist" (Wheeler 1910, p. 338).