Formica rufibarbis group
Based on Seifert and Schultz (2009)
- Formica anatolica
- Formica clara
- Formica clara sinae
- Formica cunicularia
- Formica glabridorsis
- Formica orangea
- Formica rufibarbis
- Formica tarimica
- Formica tianshanica
Formica (Serviformica) rufibarbis group
Within the Palaearctic members of the subgenus Serviformica, the Formica rufibarbis group is diagnosable by the following character combination: mesosoma showing 25 to 100% reddish pigmentation; moderate to large body size (nest means of CS 1.075 - 1.731 mm); moderate eye size (nest means of EYE / CS1.4 0.274 - 0.312); posterior margin and underside of head normally without setae (nOCC and nGU may occasionally achieve 1.5 and 3.0 in Formica anatolica), mesosoma with zero to numerous setae; petiole scale rather wide (nest means of PEW / CS1.4 0.364 - 0.523), with a convex or bluntly angulate dorsal crest; gaster tergites with a dense, usually silvery pubescence (sqPDG1.4 2.7 - 3.9) and with more or less dense transverse micro-ripples (RipD1.4 3.8 - 9.2), thus appearing more matt at low magnifications. Range West Europe to East China. Moderately to strongly thermophilic; avoiding the boreal zone; in the temperate climate zone only in open, sun-exposed habitats; in warmer climate zones some species also occur-ring in woodland. Primary habitats are open grassland, and most species invade rural or suburban areas. Monodomous colonies with single to few queens. Simple, sometimes extended soil nests, frequently under stones, rarely with a flat mound of mineral soil particles or even some organic material. Predacious and trophobiotic.
The Palaearctic Formica rufibarbis Fabricius, 1793 group comprises species inhabiting different kinds of warm, well-drained, sun-exposed habitats. They occur along a habitat gradient extending from moderately dry meagre meadows and pastures to very dry, vegetation-reduced semi-deserts but they avoid bogs and are rare in woodland. Despite their comparably low numbers per unit area these mainly zoophagous, trophobiotic and nectarivorous ants attract the attention of the observer by their swift and erratic movements over the ground. Their subjection by the spectacular slave raids of dulotic ants put them on the world's literary scene very early (Huber 1810).
The name-giving species Formica rufibarbis and the related Formica cunicularia are exemplary for the taxonomic problems in this group – type specimens are not available for both species and the original descriptions allow nothing more than concluding on medium-sized Formica ants with some reddish pigment. Further deficiencies, caused by the tradition of subjective colour and structure assessment, are inadequate and erroneous species delimitations and failures to detect cryptic species. A first attempt to demonstrate a cryptic species in this group by means of numeric character description was done by Seifert (1997) but this basically successful approach had only a regional scope, its character system was not sufficiently developed and questions on synonymies remained unsolved.
Here we want to clarify the nomenclatural situation in the F. rufibarbis group for the whole Palaearctic and will show the existence of nine species by means of numeric morphology-based alpha-taxonomy (Numobat), assisted by zoogeographical and ecological arguments. A credible demonstration of cryptic species is only possible if a sufficiently large sample is available for each considered species. This condition was only given in the worker caste with a total of 3567 Numobat-studied individuals belonging to 827 nest samples. The rarity or lack of gynes and males in the collections does not allow using these castes for reliable statements on the validity of characters. Hence, the scope of this revision is restricted to the worker caste. We apply the unified species concept (USC) of De Queiroz (2007). It considers a separately evolving metapopulation lineage as the only necessary conceptual property of species and recognises the species criteria of other species concepts (e.g., reproductive isolation, niche separation, phenotypic and genetic cohesion and clustering) only as operational criteria. The oversplitting bias of the USC in case of single-source decisions requires that each discipline must find its own remedy against this fault. In case of Numobat data we applied a confidence threshold of p > 98%.
The delimitation of the species groups in this paper is purely phenotypic and serves primarily to allow an entry into the right species determination key. Nevertheless, there is much hope that this phenotypic clustering on the basis of non-cladistic Numobat data (see Seifert & Schultz 2008), also reflects to a great degree the true phylogenetic relationships. Three species groups within the subgenus Serviformica Forel, 1913 showing reddish pigmentation on mesosoma, varying in size from small patches to complete surface coverage, are distinguished:
(a) the European to East Asian Formica rufibarbis group, characterised by the reduction of setae on posterior vertex and underside of head, smaller eyes and wider petioles;
(b) the West to Central Asian Formica subpilosa group, charaterised by intermediate setae numbers on posterior vertex and underside of head, smaller eyes and wider petioles; and
(c) the Eurocaucasian Formica cinerea group, characterised by large setae numbers on posterior vertex and underside of head, large eyes and narrow petioles.
The criterion of reddish mesosomal surface patches is usually reliable to distinguish the F. rufibarbis group from the dark pigmented species related to Formica fusca, Formica lemani, Formica picea, Formica gagates, Formica fusca tombeuri (which we plan to raise to species status and to establish as a senior synonym of Formica decipiens elsewhere, based on morphometric evidence; B. Seifert & R. Schultz, unpubl.), or Formica pyrenaea. As exception from this rule, light yellowish-reddish brown mesosomas occasionally occur among the blackish species in some specimens of the Iberian F. pyrenaea and F. fusca tombeuri and entirely dark mesosomas exceptionally occur in the F. rufibarbis group member F. cunicularia. As in many other ant groups with unsettled taxonomy and confused determination by traditional subjective taxonomy, an uncritical carryover of literature statements on distribution is not possible. As a consequence, zoogeographic knowledge is badly deficient and the maps presented here should not be interpreted to show a complete distribution picture. They only show the distribution of sampling sites from which we got reliably Numobat-determined material.