AntWiki: The Ants --- Online
Jump to navigation Jump to search
Temporal range: 100.5–89.9 Ma Cenomanian to Turonian, Late Cretaceous
Scientific classification
Kingdom: Animalia
Phylum: Arthropoda
Class: Insecta
Order: Hymenoptera
Family: Armaniidae
Dlussky, 1983
Type genus

The taxonomic placement of the fossils belonging to this group, while relatively short, is complex and currently subject to speculation more than well-justified argumentation. They have generally been considered a family outside the ants (largely because of the lack of a known worker caste) or a subfamily within the Formicidae. However, they were recently placed within the subfamily Sphecomyrminae by Borysenko (2017) based on an analysis of antennal and petiolar morphology.

Armaniidae is an extinct family of ant-like hymenopterans known from male and winged female (not worker) Cretaceous fossils found in Asia and Africa (Grimaldi et al., 1997). Armaniidae is treated by some authors as one of the stem-group subfamilies within the Formicidae (as Armaniinae, for example by Bolton (2003), Ward (2007)), however most detailed fossil studies treat it as a family closely related to but distinct from true ants (LaPolla et al. 2013).

The group is known exclusively from impression fossils which have a limited preservation quality, leading to the uncertainty of what features are present in the described species. Overall armaniids have a poorly developed petiole which is broadly attached to the thorax, short scapes on the antennae, and queen-like looking females. The mandibles are vespid like, with possibly only one or two teeth, though this may be an artifact of preservation. The short scape is a feature that is also seen in Sphecomyrminae members, and does not exclude armaniids from Formicidae. Similarly the petiole is a feature that is seen in both the true formicids, and in the extinct chrysidoid wasp family Falsiformicidae, which is not related to formicids at all. The two defining features of the true formicids are considered to be the presence of females which are divided into adult workers and queens. Currently no worker-like armaniid specimens are known for the described species. The presence of metapleural glands in some fossils has been reported by Dlussky (1999) but the veracity of the presence is uncertain (Engel, 2005).

History and classification

Armaniidae is sometimes treated as the most basal of the ant subfamilies, and classed as a stem-group which is more distant in relation to modern ants than the next stem group, members of the Sphecomyrminae. More often the group, treated as "ant-like wasps", is elevated to the rank of family, and considered as a possible sister group to Formicidae. It has been suggested by Engel and Grimaldi that the group may be paraphyletic. This position is in contrast to the original hypothesis of Russian paleoentomologist Gennady Dlussky, who first described the family. Dlussky considered the group, when erected in 1983, have been an intermediate family bridging the families Scoliidae and the true formicids (LaPolla et al. , 2013). In contrast to both the treatment as a separate family and as a distinct subfamily, entomologist E. O. Wilson, in a 1987 paper, suggested that the then known armaniids and Sphecomyrma represented a single species. Wilson, in synonymizing the groups, made the hypothesis that the different described genera were actually fossils of different castes of the same species, with Sphecomyrma freyi being workers, Armania robusta being queens, and "Paleomyrmex" zherichini as the winged males. This view was rejected as new fossils and species were described.


Cretaceous of Botswana (1 collection), Kazakhstan (1 collection), the Russian Federation (4 collections).

Genera and Species


The following information is derived from Barry Bolton's Online Catalogue of the Ants of the World.

  • †ARMANIIDAE [junior synonym of †Sphecomyrmini]
  • †Armaniidae Dlussky, 1983: 66. Type-genus: †Armania Dlussky, 1983: 67.

Taxonomic History

  • †Armaniidae as family of Formicoidea: Dlussky, 1983: 66.
  • †Armaniidae as junior synonym of Formicidae: Wilson, 1987: 49; Carpenter, 1992: 491.
  • †Armaniidae as family: Dlussky, 1983: 66 [†Armaniidae]; Dlussky & Fedoseeva, 1988: 77; Grimaldi, et al. 1997: 7; Dlussky, 1999a: 63; Rasnitsyn, 2002: 249; Engel & Grimaldi, 2005: 4; LaPolla, et al. 2013: 618 (in text).
  • †Armaniinae as subfamily of Formicidae: Bolton, 1994: 187 [†Armaniinae]; Dlussky, 1996: 83 [†Armaniinae]; Bolton, 2003: 73, 259 [†Armaniinae].
  • †Armaniidae as stem ant subfamily of Formicidae: Ward, 2007a: 555 [†Armaniinae].
  • †Armaniidae as junior synonym of †Sphecomyrmini: Borysenko, 2017: 17 (in text).

Taxonomic References

Dlussky, 1983: 67 (genera and species key); Bolton, 1995b: 9 (catalogue); Bolton, 2003: 73, 259 (diagnosis, synopsis); LaPolla, et al. 2013: 618 (discussion); Borysenko, 2017: 17 (synonymy).

Taxonomic Notes

Bolton (2003):

  • (i) Characters used in the diagnosis are entirely plesiomorphic with respect to all other subfamilies of Formicidae.
  • (ii) Grimaldi, Agosti & Carpenter (1997) excluded *armaniines from Formicidae primarily because no metapleural gland is visible, but considering that all *armaniine fossils are merely impressions in rock this is hardly surprising. While it cannot conclusively be proved by individual characters whether the *armaniines should or should not be included in Formicidae, their descriptions and illustrations are of decidedly ant-like entities. Because the females combine prognathous heads and ant-like venations with the presence of a differentiated petiolar segment and the possibility of mass nuptial flights it seems reasonable to regard them as a subfamily of admittedly very basal ants, perhaps the (currently paraphyletic) sister-group of all other formicids.
  • (iii) The presence or absence of a trochantellus in this group remains equivocal; it is absent in all extant Formicidae.


Bolton (2003) - Mandible bidentate. Head prognathous. Scape very short, funiculus long and flexuous. Venation ant-like (note 1). Waist of one poorly separated segment (petiole), posteriorly very broadly articulated with abdominal segment III (first gastral). No constriction present between abdominal segments III and IV (i.e. segment IV without differentiated presclerites) (note 2). Sting present.


  • (1) Cross-veins 3rs-m and 2m-cu are absent from the forewing. Basically the *armaniine forewing has exactly the same pattern of veins and cells as do generalised extant species of Ponerini, Platythyreini and Myrmeciini, and some species of *Sphecomyrmini.
  • (2) Discussions of fossil taxa, and sometimes of extant forms, often refer to the presence or absence of a constriction between abdominal segments III and IV. When a constriction is present it superficially appears to be between the two segments, but in fact is between the presclerites and postsclerites of abdominal segment IV and happens to be visible because the posterior margin of segment III overlaps the presclerites of IV, but not the postsclerites or the girdling constriction that separates them. Thus, "constriction present" is usually another way of saying that abdominal segment IV has strongly differentiated presclerites.