Proceratiinae

Members of the Proceratiinae are seldem encountered because of their small size and cryptic habits. Workers forage below the ground surface or in leaf litter and nests are small, containing fewer than 100 workers. Most species are thought to be specialist predators of arthropod eggs, although direct observations are few. They are most often encountered in leaf litter samples.

Identification
Members of this subfamily have the frontal lobes reduced to a narrow, sharp ridge between the antennal sockets and the antennal insertions are clearly visible when viewed from the front. In addition, most species (members of Discothyrea and Proceratium) have the upper plate of the second segment of the gaster (second gastral tergite) strongly arched so that it forms the rear-most part of the gaster when viewed from the side, the remaining segments being pushed forward underneath the first segment of the gaster. The remaining species (in Probolomyrmex) have a single petiolar node and an elongate gaster with a constriction between the first and second segments. However, the reduced frontal lobes and exposed antennal insertions will separate these species from those belonging to the otherwise similar Ponerinae.

Males: Boudinot (2015) - All proceratiine genera share the following characters which are required for identification: oblique mesopleural sulcus present; mesotibia with one or no apicoventral spurs, metatibia with one apicoventral spur; propodeal lobe present; three to five closed cells present on forewing; jugal lobe absent; petiolar tergum and sternum distinct; abdominal sternum IX unpronged and edentate. Two conditional sets of characters are required in conjunction with those indicated above: 1) if mandibles triangular then antennal toruli situated well-posterad anterior clypeal margin and crossvein 1m-cu absent; and 2) if mandibles reduced then antennal toruli situated at or produced anterad anterior clypeal margin; 1m-cu may be present or absent. Additionally, proceratiine males may or may not have vaulted fourth abdominal terga, and the eighth abdominal tergum is never spiniform.

Nomenclature
PROCERATIINAE [subfamily of Formicidae]
 * Proceratii Emery, 1895j: 765 [as tribe of Dorylinae]. Type-genus: Proceratium.
 * Proceratiinae: Bolton, 2003: 48 [as subfamily of Formicidae].

Bolton 2003:

The poneromorph subfamilies

Subfamilies Amblyoponinae, Ectatomminae, Heteroponerinae, Paraponerinae, Ponerinae, Proceratiinae.

Diagnosis Orifice of metapleural gland never concealed by a dorsally located cuticular flange or flap. Propodeal lobes present. Waist of one segment (petiole) that is separated posteriorly from abdominal segment III (first gastral) at least by a constriction (note 1). Helcium sternite retracted, overlapped by the tergite (note 2) (also in male). Abdominal segments III and IV with tergosternal fusion (also in male) (note 3). Abdominal segment IV with presclerites and usually a girdling constriction present between the presclerites and postsclerites (note 4) (also in male). Spiracles of abdominal segments V - VII concealed by posterior margins of preceding tergites. Sting present, usually strongly developed. [Synopsis, p. 153.]

Notes (1) In almost all poneromorphs abdominal segment III varies from slightly larger than IV to slightly smaller than IV. However, in Paraponerini and a few species of Proceratiini segment III is markedly reduced with respect to IV and may be termed sub-postpetiolate. See also notes under myrmicomorphs. (2) Helcium sternite is convex and not overlapped by the tergite only in Discothyrea (Proceratiini). In this respect Discothyrea resembles the dorylomorphs but otherwise their morphologies are very different; the similarity of the helcium is presumed to be by convergence. (3) For distribution of tergosternal fusion of abdominal segment III throughout the family see under dorylomorphs (note 3). Of all the poneromorphs only the monotypic Malagasy amblyoponine genus Adetomyrma lacks tergosternal fusion on abdominal segments III and IV. Whether this is plesiomorphic or a reversal from a previously fused state remains under debate (see discussion in Ward, 1994). It is by no means definite that tergosternal fusion of abdominal segment IV represents a poneromorph synapomorphy. Outside the poneromorphs this fusion is restricted to Tatuidris (Agroecomyrmecini) and Ankylomyrma (Ankylomyrmini). (4) The girdling constriction is usually apparent but in the amblyoponine Adetomyrma sharply differentiated presclerites are absent on abdominal segment IV. In Ponerini the character is variously reduced or lost in such genera as Asphinctopone and Phrynoponera, and in some individual species or species groups within larger genera such as Leptogenys, Anochetus, Odontomachus, Pachycondyla, and also in Simopelta.

Comments (i) The traditional large subfamily Ponerinae is abandoned here and its former components are regrouped as six independent subfamilies. This radical reassessment is because it has become apparent in recent years that the old and long-established concept of a single "subfamily Ponerinae" is no longer defensible. Regarding all the poneromorphs as a single subfamily has probably held back the generation of an accurate phylogeny in this part of Formicidae because "Ponerinae" as a terminal taxon could not be defined in a precise manner. (ii) Despite the lack of an unequivocal synapomorphy the six subfamilies together are treated here under the unofficial group-name poneromorph, to distinguish them from other obvious and often better delimited assemblages of subfamilies, such as the dorylomorphs and formicomorphs. Subfamily Ponerinae is now restricted to tribes Ponerini + Platythyreini + Thaumatomyrmecini.

PROCERATIINAE

Diagnosis With characters of poneromorph subfamilies. Antennal sockets mostly to entirely exposed, close to anterior margin of head (i.e. clypeus reduced) (note 1); sockets sometimes on a shelf-like frontoclypeal structure that overhangs the mandibles. Antennal sockets horizontal, in plane of transverse axis of head. Torulus not fused to frontal lobe. Promesonotal suture fused, vestigial to entirely absent; pronotum and mesonotum not capable of movement relative to each other. Metacoxal cavities either fully closed, or closed but with a suture in the annulus. Mesotibia and metatibia each with 1 spur, or mesotibia without spur. Pretarsal claws simple. Metapleural gland orifice simple, lateral. Petiole without laterotergites. Helcium projects from about the midheight of the anterior face of abdominal segment III; no high vertical anterior face to abdominal segment III above the helcium. Stridulitrum absent from pretergite of abdominal segment IV. Jugal lobe absent from hindwing of alates. Male mandibles opposable. Male with cerci absent. [Synopsis, p. 178.]

Notes (1) The location of the antennal sockets and their degree of exposure is regarded as having evolved independently of the dorylomorphs; see comments below.

Comments (i) The feature (1) above approximates, but does not correspond in detail, to what is generally observed in dorylomorphs, except for those cerapachyines in which horizontal frontal lobes are retained. Otherwise the lack of dorylomorph apomorphies in proceratiines, coupled with the general agreement of the latter with the poneromorph diagnosis and tendency to form a frontoclypeal shelf (never seen in dorylomorphs), reinforces the probability of convergence in these anterior head features between the two groups. (ii) Modification of the frontal lobes is evident throughout the Proceratiinae. Most Proceratium species have discrete frontal lobes, the outer margins of which are elevated rather than transverse, and divergent posteriorly. Some species have the frontal lobes reduced to very slender erect carinae or absent anteriorly. In at least one Proceratium species there is an erect thin single lamella, formed by fusion of the lobes, between the antennal sockets. This last adaptation is universal in Probolomyrmex and also occurs in about half of Discothyrea species. The remainder of Discothyrea have a small to moderate raised platform behind the level of the antennal sockets, the sides of which are strongly convergent anteriorly. A few Discothyrea species exist that are intermediate between these two forms. (iii) In all Probolomyrmex, all Discothyrea and at least one Proceratium species the open antennal sockets are borne on an anteriorly projecting shelf-like frontoclypea1 structure that overhangs the mandibles. This frontoclypeal shelf has probably evolved separately in each proceratiine genus, but in each as a continuation of the process begun by clypea1 reduction.