Agroecomyrmecinae

Both genera in this subfamily are monotypic, rarely collected and poorly known (Tatuidris; Tatuidris tatusia and Ankylomyrma; Ankylomyrma coronacantha). The former is found in the Neotropics: northern Mexico to central Brazil, French Guiana and Amazonian Peru. Most specimens and collections are currently known to occur in localities west of the Andes, with more collections tending to occur towards Central America and Mexico (Donoso 2012). The latter is known from the single type specimen, a canopy fogging sample from Ghana, and a subsequent handful of vegetation beating and litter samples from a number of surrounding countries.

Identification
Bolton (2003) provided the following diagnosis for the subfamily. At the time, the best evidence suggested this was a myrmicomorph subfamily but subsequent molecular analysis showed it to be a member of the poneriods. Additional morphological analysis of the ponderiods (Keller 2011) has in turn caste this in doubt. What is clear is this is an enigmatic taxon.

Mandibles large, masticatory margins oppose at full closure but do not overlap (note 1). Eye at extreme posterior apex of deep antennal scrobe (note 2). Clypeus very broadly triangular, broadly inserted between frontal lobes. Antennal sockets (and frontal lobes) strongly migrated laterally, far apart and close to lateral margins of head (note 3). Mesotibia and metatibia each with a pectinate spur. Alitrunk very short and compact. Petiole sessile; in posterior view the tergite and sternite not forming a circle. Abdominal segment III (postpetiole) without tergosternal fusion; segment large and very broadly articulated to segment IV. Helcium in frontal view with sternite bulging ventrally and overlapped by the tergite, sternite attached to tergite some distance up the inner tergal surface (note 4). Abdominal segment IV with complete tergosternal fusion (note 5), with stridulitrum on pretergite. Sternite of abdominal segment IV reduced, tergite much larger than sternite and strongly vaulted (note 6).

Notes (1) Mandibles that do not overlap at full closure are also characteristic of the dacetine tribe group and the isolated genus Lenomyrmex. These are certainly convergences as Tatuidris is otherwise very different morphologically and lacks Myrmicinae and dacetine tribe group apomorphies.

(2) Among the Myrmicinae eyes also occur at the apex of the scrobe only in some species of Cephalotes.

(3) Antennal sockets that are close to the lateral margins of the head also occur in the dacetine tribe group and in Cataulacini. In the former this is the result of narrowing of the anterior portion of the head, not of outward migration of the sockets which remain relatively close together. In Cataulacus the position of the sockets has been acquired by a parallel evolutionary process from a very different basic cephalic morphology. (4) In other words the helcium here resembles that of the poneromorphs, which is also the generalised state prevalent throughout the Formicidae; see notes under dorylomorph subfamilies.

(5) With the exception of the strange amblyoponine genus Adetomyrma, tergosternal fusion of abdominal segment IV is otherwise universal only in the poneromorphs. Fusion of this segment in agroecomyrmecines is regarded, perhaps incorrectly, as independently evolved here (its presence is of course speculative in the fossil genera). Its only other occurrence is in the myrmicine Ankylomyrma (Ankylomyrmini), where the condition is certainly uniquely derived.

(6) For other reductions of abdominal sternite IV see notes under formicomorph subfamilies.

Males Boudinot (2015) - Uniquely identified globally by petiolation of abdominal segment III (post petiole) and supraaxial helcium. Furthermore, the petiolar comformation seems to be globally unique, with the posterior petiolar foramen raised completely dorsad the anterior foramen. Identification may be confirmed with the following combination of characters: mandibles reduced, edentate; antennal toruli situated distant from anterior clypeal margin; antenna 12-merous; meso- and metatibia with one ventroapical spur each; forewing with five closed cells; jugal lobe absent; petiolar tergum and sternum distinct; abdominal segment IV pre- and postsclerites separated by cinctus; abdominal tergum IV vaulted; abdominal tergum VIII not spiniform; abdominal sternum IX apex rounded.

Nomenclature
AGROECOMYRMECINAE [subfamily of Formicidae]
 * Agroecomyrmicini Carpenter, 1930: 34 [as tribe of Myrmicinae]. Type-genus: †Agroecomyrmex.
 * Agroecomyrmecini: Brown & Kempf, 1968: 184 [emended spelling].
 * Agroecomyrmecini: Baroni Urbani & De Andrade, 2007: 75 [as tribe of Myrmicinae].
 * Agroecomyrmecinae: Bolton, 2003: 51; Ward, 2007a: 555 [as subfamily of Formicidae].

The subfamily was described by Carpenter (1930) as a new tribe to place the newly described fossil species Eulithomyrmex rugosus into a higher order clade. The original description is listed below (also see Ankylomyrma and Tatuidris).

Female - Head subquadrate; mandibles small; clypeus large; antennal scrobes present; antennae short, 12-segmented, with a two-jointed club; epinotum not armed; petiole and postpetiole short and compressed, the forewing with two cubital cells; head, thorax, and pedicel, coarsely sculptured.

Male - Antennae 13-segmented; scape short, but a little longer than the second segment; sculpturing weaker than that of the female; forewing with two cubital cells.

Worker - Very similar to the female, apparently differing only in the smaller size.