Centromyrmex sellaris

Specimen labels show Centromyrmex sellaris has been collected from termitaries of the genus Odontotermes.

Identification
This species and Centromyrmex angolensis are the most widely distributed and most commonly encountered species of the feae group in Africa. Superficially the two look very similar, but angolensis always has a longer, narrower head and only ever has a single stout spiniform seta at the apex of the anterior surface of the metatibia, about opposite the pectinate spur. In addition, the basal mandibular tooth in angolensis is generally absent, sometimes vestigially present, but there is never the basal tooth + diastema arrangement that is characteristic of sellaris. Care should be taken with this character because of variation in development in sellaris but it is usually a good indicator of the species. (Bolton and Fisher 2008)

Distribution
This taxon was described from Cameroon. It is also found in Ivory Coast, Mozambique, Kenya, Zimbabwe, Guinea, Uganda, Tanzania, Gabon, Zambia and Democratic Republic of Congo, Ghana and Nigeria.

Nomenclature

 *  sellaris. Centromyrmex sellaris Mayr, 1896: 230 (w.) CAMEROUN. Senior synonym of arnoldi, congolensis, constanciae, guineensis: Bolton & Fisher, 2008c: 22.
 * constanciae. Centromyrmex constanciae Arnold, 1915: 38, pl. 2, fig. 14 (w.q.) ZIMBABWE. Arnold, 1926: 199 (m.). Senior synonym of arnoldi: Arnold, 1926: 199 (in text). Junior synonym of sellaris: Bolton & Fisher, 2008c: 22.
 * arnoldi. Centromyrmex arnoldi Santschi, 1919b: 229, figs. a-d (w.m.) MOZAMBIQUE. Subspecies of constanciae: Santschi, 1920b: 8. Junior synonym of constanciae: Arnold, 1926: 199 (in text); of sellaris: Bolton & Fisher, 2008c: 22.
 * congolensis. Centromyrmex congolensis Weber, 1949b: 5, figs. 3, 4 (w.) DEMOCRATIC REPUBLIC OF CONGO. Junior synonym of sellaris: Bolton & Fisher, 2008c: 23.
 * guineensis. Centromyrmex arnoldi r. guineensis Bernard, 1953b: 186, fig. 1 (w.) GUINEA. Subspecies of constanciae: Bolton, 1995b: 140. Junior synonym of sellaris: Bolton & Fisher, 2008c: page.

Worker
Bolton and Fisher (2008) - TL 4.8-6.1, HL 0.84-0.98, HW 0.90-1.13, CI 108-118, ML 0.58-0.72, MI 68-75, SL 0.64-0.76, SI 67-73, PW 0.74-0.92, WL 1.54-1.80 (15 measured).

With characters of the genus and the feae group. Head capsule in full-face view always appears obviously broader than long, CI 108 or usually more. Mandibles smooth with scattered small punctures. Masticatory margin of mandible with 7–10 small, low blunt teeth that are usually broadly low-triangular but are often reduced to mere crenulations when worn. Basal tooth at basal angle of mandible and usually obvious, only rarely reduced and insignificant. Distal of the basal tooth there is usually a diastema before the next tooth on one or both of the mandibles, but sometimes this is not apparent as a denticle may be present within the diastema on one, or less often both, of the mandibles. Dorsum and sides of head with scattered punctures on smooth cuticle, and also with weak striation within the antennal fossae and on the sides, especially anteriorly. Extent of the striate component is variable. Metatibia with only normal setae dorsally but its anterior surface, at the apex and approximately opposite the pectinate spur, with 2 (3 in a single specimen) much stouter and usually more darkly coloured spiniform setae. Petiole node in dorsal view broader than long. Pronotal dorsum, and anterior mesonotum, with widely scattered broad, shallow punctures that may be almost effaced. Pronotum dorsally also with variable weak oblique or arched faint disorganised sculpture. Colour yellow to light brown.

Queen
Bolton and Fisher (2008) - TL 6.9, HL 1.01, HW 1.12, CI 111, OI 32, ML 0.70, MI 69, SL 0.80, SI 71, PW 1.11, WL 2.00. The queen of this species should run out correctly in the key to workers but care should be taken as the queens of longiventris and ereptor remain unknown. The form of the mandible described above for the worker is reproduced in the queen caste.

Type Material
Bolton and Fisher (2008) - Holotype worker, CAMEROUN: no loc. 1891 (Y. Sjöstedt) [examined] (see note 1).

Centromyrmex constanciae Syntype workers and queen, ZIMBABWE: Bembesi, 24.iii.1913 [worker examined].

Centromyrmex arnoldi Santschi, 1919: 229, figs. a–d. Syntype workers and males, MOZAMBIQUE: Amatongas Forest, ix.1917 (G. Arnold) [examined].

Centromyrmex congolensis Holotype worker, DEMOCRATIC REPUBLIC OF CONGO: Niangara, 1.iii.1948 (N.A. Weber) (not in or, presumed lost; see note 2).

Centromyrmex arnoldi r. guineensis Holotype worker, GUINEA: Mt Nimba, Nion, St.22, 700 m., 15.iv.1942 (Lamotte) [examined].

NOTES.

1 The unique holotype of sellaris was discovered by Dr Hege Vårdal in the NHRS collection, still preserved in alcohol after 117 years. The specimen was accompanied by three labels, two of which merely stated “Camerun” and “Sjöstedt” respectively. A larger and more informative label carried the information “Riksmuseets Entomologiska Afdelning. Centromyrmex sellaris Mayr n. sp. Typ. Kamerun, 1891. Colleg. Y. Sjöstedt”. The holotype has now been mounted upon a standard card point.

2 Although the holotype of congolensis appears to have been lost, it is possible that it may still be present but unrecognised in Weber’s material, either at AMNH or MCZC. Fortunately, the original description and figures are sufficient to allow identification of

Determination Clarifications
Bolton and Fisher (2008) - The species referred to as C. sellaris in Lévieux (1976, 1983) is correctly identified, as indicated by voucher specimens deposited in MCZC.