Tetramorium sabatra Hita, Hita, 2012

Hita Garcia, F. & B. L. Fisher, 2012, The ant genus Tetramorium Mayr (Hymenoptera: Formicidae) in the Malagasy region - taxonomic revision of the T. kelleri and T. tortuosum species groups., Zootaxa 3592, pp. 1-85: 76-79

publication ID


publication LSID


persistent identifier


treatment provided by


scientific name

Tetramorium sabatra Hita

sp. n.

Tetramorium sabatra Hita  Garcia & Fisher sp. n.

(Figs. 11, 15, 24, 25, 26, 135, 136, 137, 142)

Holotype worker, MADAGASCAR, Toliara, Rés. Andohahela, 11 km NW Enakara, 24.56667 S, 46.83333 E, 800 m, montane rainforest, pitfall trap, collection code BLF00490, 16.XI.1992 (B.L. Fisher) (CASC: CASENT0189241). Paratypes, three workers with same data as holotype (CASC: CASENT0218056; CASENT0218057; CASENT0270780); one worker from Toliara, Rés. Andohahela, 11 km NW Enakara, 24.56667 S, 46.83333 E, 800 m, montane rainforest, from sifted litter, collection code BLF00492, 17.XI.1992 (B.L. Fisher) (CASC: CASENT0189240); and one worker from Toliara, Rés. Andohahela, 10 km NW Enakara, 24.56667 S, 46.81667 E, 430 m, rainforest, sifted litter, collection code BLF00522, 22.XI.1992 (B.L. Fisher) (CASC: CASENT0189239).


Tetramorium sabatra  is easily distinguished from the other group members by the following character combination: antennal scapes comparatively short (SI 73-80); extremely long and massively constructed propodeal spines (PSLI 48-72); anterodorsal and posterodorsal margins of petiolar node situated at about same height; mesosoma with one or two pairs of standing hairs, restricted to dorsal pronotum; hairs on leading edge of antennal scapes usually strongly appressed; first gastral tergite with few standing hairs and very sparse, short pubescence; very dark brown to black colouration.


HL 1.00-1.12 (1.04); HW 1.02-1.13 (1.05); SL 0.76-0.90 (0.82); EL 0.20-0.23 (0.21); PH 0.48-0.54 (0.50); PW 0.71-0.79 (0.75); WL 1.27-1.41 (1.33); PSL 0.50-0.81 (0.60); PTL 0.34-0.37 (0.36); PTH 0.39-0.44 (0.41); PTW 0.27-0.31 (0.28); PPL 0.30-0.34 (0.32); PPH 0.40-0.46 (0.42); PPW 0.37-0.44 (0.39); CI 100-103 (101); SI 73-80 (77); OI 18-21 (20); DMI 55-58 (56); LMI 37-40 (38); PSLI 48-72 (57); PeNI 36-39 (38); LPeI 82-90 (87); DPeI 75-86 (80); PpNI 50-56 (52); LPpI 71-77 (75); DPpI 119-129 (125); PPI 134-143 (138) (ten measured).

Head as long as wide to weakly wider than long (CI 100-103); posterior head margin strongly concave. Anterior clypeal margin medially impressed. Frontal carinae strongly developed, diverging posteriorly, and ending at corners of posterior head margin. Antennal scrobes well-developed, moderately deep, but narrow, and without defined posterior and ventral margins. Antennal scapes short to moderate, not reaching posterior head margin (SI 73-80). Eyes small to moderate (OI 18-21). Mesosomal outline in profile flat, moderately marginate from lateral to dorsal mesosoma; promesonotal suture and metanotal groove absent; mesosoma comparatively stout and high (LMI 37-40). Propodeal spines massively developed with very broad base, extremely long, and acute (PSLI 48-72); propodeal lobes short and blunt. Petiolar node in profile rectangular nodiform, approximately 1.1 to 1.2 times higher than long (LPeI 82-90), anterior and posterior faces approximately parallel, anterodorsal and posterodorsal margins approximately at same height, dorsum slightly convex; node in dorsal view approximately 1.1 to 1.3 times longer than wide (DPeI 75-86). Postpetiole in profile subglobular, weakly anteroposteriorly compressed, approximately 1.3 to 1.4 times higher than long (LPpI 71-77); in dorsal view around 1.2 to 1.3 times wider than long (DPpI 119-129). Postpetiole in profile appearing less voluminous than petiolar node, in dorsal view approximately 1.3 to 1.5 times wider than petiolar node (PPI 134-143). Mandibles distinctly longitudinally rugose; clypeus longitudinally rugose, with three to five rugae; cephalic dorsum between frontal carinae with 9 to 12 longitudinal rugae, most rugae running unbroken from posterior head margin to anterior clypeus, few rugae interrupted or with cross-meshes; lateral and ventral head longitudinally rugose, rarely with cross-meshes. Mesosoma laterally and dorsally distinctly longitudinally rugose. Forecoxae with very distinct and pronounced longitudinal rugae. Waist segments longitudinally rugose, rugae on waist segments weaker than on head and mesosoma, especially dorsally. Gaster completely unsculptured, smooth, and shining. Ground sculpture generally faint to absent everywhere on body. Head with abundant standing hairs; hairs on mesosoma restricted to dorsal pronotum, usually one or two pairs of hairs present, rarely three pairs; waist segments and first gastral tergite with few to numerous standing hairs; first gastral tergite with very sparse, short, and appressed pubescence. Anterior edges of antennal scapes usually with appressed hairs (decumbent in one specimen). Body a uniform very dark brown to black colour.


Despite the comparatively small number of known specimens (15 in total), T. sabatra  seems to be widely distributed in the rainforests and montane rainforests of eastern Madagascar, as well as in Analavelona in the southwest. The southernmost locality is the type locality, Andohahela, and the northernmost locality is Montagne d'Akirindro. Inbetween it is only known from few more localities. Furthermore, T. sabatra  seems to inhabit forests at elevations of 430 to 1300 m, and was mainly collected from the ground. The scarcity of material and wide distribution range suggests that this species is fairly rare, uncommon, or just sampled inappropriately. Four out of the fifteen specimens were collected while visiting the flowers of Impatiens mandrakae Fischer & Rahelivololona  ( Balsaminaceae  ) at Mandraka. This suggests that T. sabatra  lives in the vegetation and is therefore rarely sampled from the ground. This could be true for T. latreillei  and T. smaug,  too, and might explain the rarity of these three species.

Inside the T. smaug  complex, T. sabatra  is unlikely to be misidentified with T. adamsi,  T. marojejy,  and T. nazgul.  The latter three have numerous standing hairs on the first gastral tergite, and are much hairier than T. sabatra.  Furthermore, T. adamsi  has a petiolar node shape with the posterodorsal margin situated higher than the anterodorsal, while both margins are at the same height in T. sabatra.  Tetramorium nazgul  also has much longer antennal scapes (SI 89-93), and T. marojejy  is of orange to light brown body colour. Nevertheless, T. sabatra  appears to be morphologically most closely associated with T. latreillei  and T. smaug  since they share the same morphometric range and have a very similar gestalt. They are very darkly coloured species with massively developed propodeal spines and reduced hairiness. However, T. sabatra  can be well separated from T. latreillei  due to the absence of standing hairs on the first gastral tergite in the latter, whereas a few to several standing hairs are always present in T. sabatra.  In addition, the latter species has only very sparse and inconspicuous pubescence on the first gastral tergite while T. latreillei  has moderately dense and distinct pubescence. Tetramorium smaug  mainly differs from T. sabatra  in the number of hairs on the mesosomal dorsum. The latter has just one or two pairs on the pronotal dorsum, whereas T. smaug  has 7 to 14 pairs throughout the whole mesosomal dorsum. Also, in T. sabatra  the hairs on the leading edges of the antennal scapes are usually appressed but subdecumbent to suberect in T. smaug. 

As noted in the above descriptions, T. latreillei,  T. sabatra,  and T. smaug  could all belong to one species with considerable variation in patterns of pilosity/pubescence. However, as previously discussed, we are of the opinion that these characters are generally of high diagnostic value at the species level, which leads us to treat the three as distinct species. The available material is limited, however, and more material could demonstrate that our current species delimitations are incorrect.


The species epithet is an arbitrary combination of letters.

Material examined

MADAGASCAR: Antananarivo, Mandraka, 18° 54' 46" S, 47° 53' 32" E, ca. 1200 m, 23.I.2006 (A. Erpenbach); Fianarantsoa, Parc National Befotaka-Midongy, Papango 27.7 km S Midongy-Sud, Mount Papango, 23.83517 S, 46.96367 E, 940 m, 15.XI.2006 (B.L. Fisher et al.); Toamasina, F.C. Didy, 18.19833 S, 48.57833 E, 960 m, 16.-23.XII.1998 (H.J. Ratsirarson); Toamasina, Montagne d'Akirindro, 7.6 km 341° NNW Ambinanitelo, 15.28833 S, 49.54833 E, 600 m, rainforest, 17.-21.III.2003 (B.L. Fisher, C. Griswold et al.); Toliara, Forêt Classée d'Analavelona, 33.2 km 344° NNW Mahaboboka, 22.64333 S, 44.17167 E, 1300 m, 22.-26.II.2003 (B.L. Fisher, C. Griswold et al.); Toliara, Rés. Andohahela, 11 km NW Enakara, 24.56667 S, 46.83333 E, 800 m, montane rainforest, 16.-17.XI.1992 (B.L. Fisher); Toliara, Rés. Andohahela, 10 km NW Enakara, 24.56667 S, 46.81667 E, 430 m, rainforest, 22.XI.1992 (B.L. Fisher).