Hagensia

A small South African genus (two described species and four subspecies), notable for having gamergate workers.

Identification
Schmidt and Shattuck (2014) - Diagnostic morphological apomorphies of Hagensia workers include the presence of a dorsal mandibular groove, sharp pronotal margins, and a squamiform petiole with anterior and posterior faces meeting along a sharp edge. This combination of characters is unique to Hagensia. The dorsal mandibular groove is approximated by the condition in Euponera wroughtonii, and a sharp pronotal margin is present in some African Centromyrmex species as well as some members of Pachycondyla and Neoponera, but these genera lack the other diagnostic characters of Hagensia.

Distribution
Hagensia is restricted to the southern and eastern coastal areas of South Africa.

Biology
Like most ponerines, Hagensia are generalist predators and scavengers. Duncan and Crewe (1994a) studied the foraging behavior of Hagensia havilandi and observed exclusively diurnal solitary foraging, with no chemical recruitment (though with occasional tandem running). Foraging occurred in leaf litter, with prey consisting of a diversity of invertebrates and some plant matter; foragers apparently navigated using visual cues. Arnold (1951) reported diurnal foraging in Hagensia peringueyi, but crepuscular and nocturnal foraging in H. havilandi, in contrast to both Villet (1992a) and Duncan & Crewe (1994a). Hagensia forms subterranean nests, with colony sizes of about 10 to 50 workers in H. havilandi (Arnold, 1951; Villet, 1992a; Duncan & Crewe, 1994a). Reproduction occurs via a single mated gamergate worker, the queen caste having been completely lost (Peeters, 1991a). Villet (1992a) found no evidence of a dominance hierarchy among workers of H. havilandi.

Nomenclature

 *  HAGENSIA [Ponerinae: Ponerini]
 * Hagensia Forel, 1901f: 333 [as subgenus of Megaloponera (sic)]. Type-species: Megaloponera (sic) (Hagensia) havilandi, by monotypy.
 * Hagensia subgenus of Megaponera: Wheeler, W.M. 1910g: 135; Emery, 1911d: 61.
 * Hagensia subgenus of Euponera: Forel, 1917: 237.
 * Hagensia raised to genus: Arnold, 1926: 202; Schmidt & Shattuck, 2014: 89.
 * Hagensia junior synonym of Pachycondyla: Brown, in Bolton, 1994: 164.
 * Hagensia revived from synonymy: Schmidt & Shattuck, 2014: 89.

Schmidt and Shattuck (2014):

Hagensia was originally erected by Forel as a subgenus of' Megaponera to house the newly described species Megaloponera [sic] havilandi Forel. Forel considered the absence of a preocular carina in this species to be of primary significance but otherwise felt that it was close toMegaponera (sensu stricto). Arnold (1915) synonymized Hagensia with Euponera (subgenus Mesoponera), based on characters of the males, and Forel (1917) then raised Hagensia to subgenus status under Euponera. Arnold (1926) later raised Hagensia to generic status based on several characters of both the worker and male castes, and revised the species-level taxonomy of the genus (Arnold, 1951). As part of his unfinished revision of ponerine taxonomy, W. L. Brown (in Bolton, 1994) synonymized Hagensia under Pachycondyla without phylogenetic justification.

We are reviving Hagensia to full genus status, given its morphological distinctiveness and phylogenetic age, which is consistent with that of other recognized ponerine genera. Schmidt's (2013) molecular phylogeny of the Ponerinae places Hagensia havilandi (the type species) with strong support within the Odontomachus group. The sister group of Hagensia is unresolved in this phylogeny, and there is no evidence of a close relationship toMegaponera, Euponera, or Mesoponera, though a sister relationship with any of them could not be rejected. Hagensia is not closely related to Pachycondyla.

Hagensia’s dorsal mandibular groove, sharply margined pronotum, and sharply edged squamiform petiole are apparently autapomorphic within the Odontomachus group, and hence do not assist in determining its phylogenetic position. Morphologically, Hagensia resemblesMegaponera and Ophthalmopone most closely. The workers of these genera share large body size, dark coloration, finely punctate sculpturing, sparse or scattered pilosity, dense pubescence (sparser in minor workers ofMegaponera), a distinct metanotal suture, narrow propodeal dorsum, slit-shaped propodeal spiracles, a simple metapleural gland orifice which opens posterolaterally, a weak or obsolete gastral constriction between A3 and A4, and toothed tarsal claws (absent in some Ophthalmopone). In many cases these similarities are likely plesiomorphic, but some are potentially synapomorphic for these three genera, such as the toothed tarsal claws (though toothed tarsal claws do occur in some other members of the Odontomachus group).

In addition, Hagensia shares with Megaponera and Ophthalmopone the loss of alate queens (Peeters, 1991a). Both Hagensia and Ophthalmopone have lost the queen caste entirely (reproduction is performed by gamergates), while in Megaponera the queen is ergatoid. It is tempting to consider this a possible synapomorphy for these taxa, but ponerine reproductive strategies seem to be quite fluid in evolutionary time and may not be good phylogenetic markers. Both gamergates and ergatoids have evolved in many other members of the Odontomachus group, such as at least some species of Streblognathus, Leptogenys, Bothroponera, and Pseudoneoponera.

Wheeler & Wheeler (1971a) examined Hagensia semipupae and found them to be quite different from those ofMegaponera, though the phylogenetic significance of these differences is uncertain. Still, given their morphological and behavioral similarities, we believe that the most probable phylogenetic placement for Hagensia is as sister to Megaponera and Ophthalmopone. A close relationship with Streblognathus is also plausible, though the morphological evidence for this relationship is not particularly compelling.

Worker
Large (TL 10.5–13.0 mm; Arnold, 1915) ants with the standard characters of Ponerini. Mandibles triangular, with a deep dorsal transverse groove. Eyes moderately large and located near the head midline. Pronotum with sharp dorsolateral margins. Mesopleuron partially divided by a transverse groove. Metanotal groove shallowly to deeply impressed. Propodeum strongly narrowed dorsally. Propodeal spiracles slit-shaped. Tarsal claws with a single preapical tooth. Metatibial spur formula (1s, 1p). Petiole squamiform, the anterior and posterior faces meeting along a sharp edge. Head and body finely punctate, with sparse pilosity and dense pubescence. Color black.

Male
See descriptions in Arnold (1915, 1926). Both winged and wingless males occur in H. peringueyi saldanhae (C. Peeters & H. Robertson unpubl.)

Larva
Pupae of Hagensia peringueyi were described by Wheeler & Wheeler (1971a).