Cerapachys sexspinus species group
This species group is distinctive enough to warrant raising the genus Yunodorylus to its status as a valid genus (Borowiec 2016).
Antennae with 11 or 12 segments.
Apical antennal segment not conspicuously enlarged, equal or little longer than two preceding.
Parafrontal ridges strongly reduced to absent (see note 1 below).
Two ridges present behind frontal carinae with medial impression between them.
Mandibles triangular, crenulate or with denticles, or elongated, with teeth.
Eyes in worker completely absent.
Ventrolateral margins of the head capsule reduced to weakly pronounced ridge.
Ocelli in workers absent.
Pronotum not marginate anterodorsally, pronotal collar not separated from dorsal surface.
Mesosoma, petiole and postpetiole not marginate dorsolaterally.
Waist of a single segment, abdominal segment IV with very broad presclerites and weak constriction between pre- and postsclerites (see note 2 below).
Fourth abdominal tergite not folding over sternite, and anterior portions of sternite and tergite are equally well visible in side view.
Spur formula 2(1s,1p),2(1s,1p) (see note 3 below).
Middle and hind basitarsi not widening distally, circular in cross-section.
Posterior flange of hind coxa not produced as raised lamella.
Metatibial glands with no discernible orifice, visible as a lighter, round to elongate patch of cuticle devoid of pubescence.
Pretarsal claws simple.
Polymorphic (see note 4 below).
Characters in italics are hypothetical autapomorphies of the group. They could not be seen in any other Cerapachys or other species of the Cerapachyinae available at the time of this study and are apparently unique for the whole subfamily.
Note 1. Absence of parafrontal ridges is characteristic of this group but can be observed in at least one species outside the group, namely Australian Ooceraea australis. However, this unrelated species is easily distinguished by having 9 antennal segments and small abdominal segment III, differentiated as postpetiole. Many Sphinctomyrmex have parafrontal ridges reduced to absent. Since monophyly of both Cerapachys and Sphinctomyrmex is questioned (Brown 1975, Brady et al. 2006), at this point it is not possible to say how many times this feature appeared or disappeared in the subfamily.
Note 2. In Cerapachys the abdominal segments III and IV are externally separated by at least a girdling constriction between pre- and postsclerites of segment IV. Additionally, abdominal segment III may be of various sizes, in some species of fragosus and typhlus species-groups sensu Brown (1975) being much smaller and forming what can be called a postpetiole. The species-group discussed here is an exception, having a very weak impression between pre- and postsclerites IV. Illustrated description of this character throughout the Cerapachyinae was provided by Bolton (1990).
Note 3. In the original description of Yunodorylus sexspinus, the tibial spur condition is described as single pectinate on middle and hind tibiae. The true condition is different, as all the specimens examined possess one simple and one broadly pectinate spur on both middle and hind tibiae. Bolton (2003) lists the following character states for spur formula in Cerapachys: 2p,2p; 2(1s,1b-p),2(1s,1p); 1s-b,1p; 1p,1p. The formula 1p,1p is the most common, occurring in, for example, the antennatus-group (containing the type species of the genus), the speciose dohertyi-cribrinodis species-group (sensu Brown 1975), and in most species with sides of petiole margined (the ‘Phyracaces lineage’ of Brown ). The spur formula 1s-b, 1p is characteristic for apparently arboreal species related to Lioponera longitarsus. The unique spur formula 2p,2p is present in the aberrant crawleyi-group (see general discussion below).
Note 4. Of the species examined here, at least Yunodorylus eguchii, paradoxus, and sexspinus exhibit considerable range of body size among workers. Only one specimen of Yunodorylus doryloides has been examined, so nothing can be said about its polymorphism. A polymorphic worker caste is also present in some apparently not related, hitherto undescribed Cerapachys (author’s unpublished data). Due to a limited amount of material, quantitative analysis of allometry and polymorphism is not attempted here but there seems to be continuous size variation, and allometry of various body parts is weakly pronounced.
No gynes or males are known thus far and the diagnosis is based solely on worker characters.