A new species of the ant genus Strumigenys F. Smith 1860 from the Philippines is described. This new species possesses four-segmented antennae, simple setae on the head dorsum, and a smooth first gastral sternite. Only 13 valid species of 882 species possess four-segmented antennae. Species that possess four-segmented antennae form two informal species groups: the emmae group, composed of species with spatulate to orbiculate setae on the head dorsum and the leading edge of antennal scape; and the eurycera group, composed of species with simple setae on the head dorsum and the leading edge of antennal scape, as well as possessing a distinctive anteriorly flat propodeum that slopes sharply downward to the propodeal tooth and a longitudinally costulate first gastral sternite. Unlike members of the emmae group, Strumigenys silangan sp. nov. possesses simple setae on the head dorsum and the leading edge of the antennal scape. Unlike the members of the eurycera group, S. silangan sp. nov. possesses a propodeum that gently slopes down to the propodeal tooth and a smooth first gastral sternite. The worker and dealate queen are described.
INTRODUCTION
The myrmicine genus Strumigenys F. Smith 1860 is a very large cosmopolitan genus containing 882 extant species and four fossil species (AntWiki, 2025; Bolton, 2023). Although Bolton (2000) treated Strumigenys separately from Pyramica Roger, 1862, Pyramica was shown by Baroni Urbani and De Andrade (2007) to be a junior synonym of Strumigenys. The molecular phylogenetic study by Booher et al. (2021) supported the morphologic findings of Baroni Urbani and De Andrade (2007). For convenience, Bolton's (2000) separate keys for Strumigenys and Pyramica are maintained because these keys still work well in species identification. Since 2000, many new species have been discovered and described (Longino, 2006; Bharti and Akbar, 2013; Dong and Kim, 2020; Booher, 2021; Booher and Hönle, 2021; Tang and Guénard, 2023; and others). Booher and Hönle (2021) proposed a new monotypic species group for their new species because of the nonmonophyletic nature of many of Bolton's (2000) species groups (Booher et al., 2021). At present there are 117 species groups in Strumigenys, several of which contain only a single species (AntWiki, 2025).
Strumigenys ants nest in soil, leaf litter, rotten wood, and sometimes in pockets of rotten plant matter in tree forks. They forage in the leaf litter and hunt springtails and other soft-bodied arthropods (Hölldobler and Wilson, 1990).
In the Philippines, the genus is represented by 37 species distributed throughout the archipelago (AntWiki, 2023b). These ants are usually found in natural areas, although one introduced species, S. eggersi Emery, 1890, was first collected on the campus of the University of the Philippines in Quezon City, a highly urbanized city (General, 2017). Of interest, the type locality of two Philippine species described by Brown (1957), S. uichancoi Brown, 1957 and S. phytibia Brown, 1957, is unknown. These ants were intercepted on fern plants (uichancoi) and orchids (phytibia) by the U.S. Plant Quarantine Office in Honolulu, Hawaii. These two species have not been recollected.
The typical Strumigenys species possesses antennae with six segments, arranged in order from the antennal attachment: the long scape, followed by three small funicular segments, and finally the enlarged antennal club of two segments. Only 15 Strumigenys species (two valid species and one undescribed species from the Neotropics and 11 valid species and one undescribed species from Australasia) are known to possess four antennal segments, where the first three funicular segments appear to have fused into a single small segment. Two valid Neotropical species, namely, S. minuscula (Kempf, 1962) and S. simulans (Santschi, 1931), bear four-segmented antennae but possess triangular mandibles, armed with numerous teeth along the masticatory margin, that close tightly without a gap. Another Neotropical species, DBB079, likely related to S. minuscula, also possesses a four-segmented antenna (D. B. Booher, personal communication). Booher et al. (2021) found that geography, not mandibular morphology, provided the strongest signal on the phylogeny of Strumigenys. Because S. minuscula, S. simulans, and DBB079 are all Neotropical species, their possession of four-segmented antennae may simply be due to convergence and are unrelated to the Australasian species.
The remaining valid 11 Australasian species possessing linear mandibles, with only two or three apical teeth, that form a large gap when fully closed belong to two informal species groups: the eurycera group and the emmae group. These two groups appear to be related because all the members bear mandibles of the S-TRAP (short mandible trap jaw mechanism) type. Booher et al. (2021) found that, in Australasia, Asia, Madagascar, and even Fiji, the S-TRAP mandible type evolved from the L-TRAP (long mandible trap jaw mechanism) and secondarily shortened.
The eurycera group is known only from New Guinea and nearby islands (AntWiki, 2024b). The four valid species of the eurycera group are characterized by antenna with only four segments, propodeal dorsum relatively flat anteriorly but posteriorly abruptly angled downward and sloping steeply to the propodeal spine; the absence of flagellate, spatulate, or orbiculate pilosity; and a costulate base of the first gastral sternite.
The emmae group is predominantly Australian in distribution, with six endemic Australian species and the pantropical tramp species, S. emmae (Emery, 1890) (AntWiki 2024a). Six of the seven valid species of the emmae species group also possess four-segmented antennae (except S. pnyxia Bolton [2000], which has the plesiomorphic six-segmented antennae) but bear spatulate or orbiculate pilosity and an unsculptured base of the first gastral sternite.
The geographic ranges of these two small groups, both with S-TRAP mandible types, are relatively close to each other but not overlapping, perhaps representing range extension and followed by subsequent extinction and speciation, but this speculation remains to be tested.
In 2003–2004, we sampled ants in both the wet and dry seasons along the same transect line in a private reforestation project just outside the Mt. Isarog Natural Park in Panicuason Village, Naga City, Camarines Sur Province, Luzon Island, Philippines. Using the same transect, we sampled the ants during the dry season (March) and wet season (June through October) of 2003 and the dry season of 2004.
Among the Isarog collections was a singleton belonging to the genus Strumigenys, with the number of antennal segments reduced from six to four, quite uncommon in this hyperdiverse genus. A subsequent transect study in a nearby abandoned farm conducted by a student collected the conspecific dealate queen. Other specimens were subsequently found in the collections (from Polillo Island) of the University of the Philippines Los Baños Museum of Natural History Entomological Collection and from the bycatch collection (from Samar Island) of the Acari laboratory of Dr. Leonila Corpuz-Raros, also in the University of the Philippines Los Baños.
In this contribution we describe an uncommon ant species that does not belong to either of the groups of species with four-segmented antennae.
MATERIALS AND METHODS
This study site was a 27-year-old narra reforestation project (Pterocarpus indicus Willd., Fabaceae) on private property adjacent to the Mt. Isarog Natural Park. The area was previously a corn and upland rice farm. Eventually, the landowner decided to plant narra trees (P. indicus Willd.). When we conducted the surveys, the narra trees had grown large enough to close the canopy, limit insolation of the forest floor, and shade over the sparse undergrowth. There was a considerable amount of leaf litter and downed woody debris on the ground.
The transect was 250 m long, with 25 transect stations each with a radius of 5 m. At each station we performed the following collecting techniques: sifting leaf litter from a randomly selected 1 m2 of forest floor on a plastic sheet and manual collection of ants; 24-hour pitfall trapping; twig breaking; breaking of dead wood (pieces that could not be broken by hand); beating of low vegetation using a 1-m2 beating sheet; and timed opportunistic collecting. In another transect study conducted nearby in 2006 the dealate queen was collected from leaf litter sifting.
The specimens from the islands of Polillo and Samar were collected by leaf litter sifting.
Worker specimens were examined and measured with a Wild M5A stereomicroscope equipped with an ocular micrometer. The dealate gyne was examined and measured with a Leica MZ16 stereomicroscope equipped with an ocular micrometer. Images of the gyne were created with a Canon 7D digital camera attached to a Leica MZ16 stereomicroscope. The map was created in QGIS 3.14 (QGIS 2023).
The following measurements and indices are reported: total length (TL), the length of the ant when stretched out; the sum of mandible length (ML), head length (HL), Weber's length (WL), lengths of the waist segments (petiole and postpetiole), and the gaster length (GL); HL, in full-face view, the distance between an imaginary line across the apices of the occipital lobes and the anterior margin of the clypeus; head width (HW), in full face view, the maximum width of the head; ML, in full-face view, the length of the mandible from the anterior margin of the clypeus to the apex of the mandible; scape length (SL), the maximum length of the antennal scape, excluding the basal neck and condyle; pronotal width (PW), in dorsal view, the maximum width of the pronotum; WL, in lateral view, the diagonal distance between the anterior edge of the pronotum, excluding the pronotal collar, and the posterior margin of the propodeal lobe; petiole length (PL), in lateral view, length of petiole from posterior margin of propodeal lobe obscuring the petiolar base to the posterior margin of petiole, excluding any spongiform lobes; postpetiole length (PPL), in lateral view, length from anterior margin to posterior margin of postpetiole; GL, in lateral view, length from the base of the first gastral tergite to the apex of the gaster; cephalic index (CI), HW/HL × 100; mandible index (MI), ML/HL × 100; scape index (SI), SL/HW × 100.
Repositories
MCZC, Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts; NHM, Natural History Museum, London; PNM, National Museum of Natural History, National Museum of the Philippines, Manila, Philippines; UPLB, University of the Philippines Los Baños Museum of Natural History, Los Baños, Laguna, Philippines.
RESULTS
Taxonomy
Class. Insecta Linneaus 1758
Order. Hymenoptera Linneaus 1758
Family. Formicidae Latreille 1809
Subfamily. Myrmicinae Lepeletier de Saint-Fargeau 1835
Genus. Strumigenys F. Smith 1860
Tribe. Attini F. Smith, 1858
Type species. Strumigenys mandibularis F. Smith 1860
Strumigenys silangan sp. nov.
Diagnosis. Four-segmented antenna; setae on dorsum of head simple; apiscrobal seta absent; antennal scape dorsoventrally flattened, expanded in distal 4/5 of length; apical segment of antenna not spindle shaped, broad basally where it articulates with preapical segment; postbuccal groove broad and developed; pronotal humeral seta absent; propodeal spine subtended by broad lamella, with posterior margin (free) concave in upper half; spongiform lobes on waist present; first gastral sternite smooth.
Distribution. Philippines, Islands of Luzon, Polillo, and Samar (Fig. 1).
Material examined
Holotype. PHILIPPINES: Luzon Island, Camarines Sur Province, Naga City, Panicuason Village, ABT1 Wet Season Transect, ex leaf litter (21L), 550 m above sea level (masl), 10viii2003, leg. J. Sabiña (deposited at PNM [PNM 15700]) (Figs. 2–5).
Paratypes. PHILIPPINES: Luzon Island, Camarines Sur Province, Naga City, Panicuason Village, ex leaf litter, 23xi-23x2006, leg. J. Caceres-Plopenio (n = 1 dealate queen) (deposited at PNM [PNM 15701]) (Figs. 6–9); Quezon Province, Polillo Island, Municipality of Burdeos, ex leaf litter berlesate, near Mapanghi Cave, 27iv2003, leg. O. L. Eusebio (n = 3 workers) (deposited at UPLB [UPLBMNH-HYM 002089, UPLBMNH-HYM 002090], NHM); Samar Island, Eastern Samar Province, Municipality of Dolores, Osmeña Village, ex leaf litter berlesate, 125 masl, 23 × 2003, leg. W. S. M. Gruezo (n = 1 worker) (deposited at MCZC [MCZ type XXXX]).
Measurements (mm)
Holotype. TL 2.14, HL 0.54, HW 0.49, ML 0.25, SL 0.30, PW 0.28, WL 0.54, PL 0.19, PPL 0.14, GL 0.49, CI 91, MI, 47 SI 62.
Paratypes (range of values, n = 4 workers). TL 2.11–2.20, HL 0.53–0.54, HW 0.48–0.49, ML 0.23–0.25, SL 0.30–0.31, PW 0.26–0.28, WL 0.54–0.56, PL 0.18–0.20, PPL 0.13–0.14, GL 0.48–0.53, CI 90–93, MI 43–47, SI 62–66.
Paratype gyne (n = 1). TL 2.69, HL 0.54, HW 0.48, ML 0.24, SL 0.28, PW 0.31, WL 0.63, PL 0.25, PPL 0.14, GL 0.60, CI 88, MI 44, SI 58.
Description
Worker. Head: Mandibles slender, narrower at base than at apex, inserted very close together in full-face view and without a trace of a lamella on the inner margin proximal of the preapical tooth. Preapical tooth spiniform, 2× as long as width of mandible and directed medially. Apical fork of mandible with two spiniform teeth, without any intercalary denticles. Apical antennal segment stout, not spindle shaped or constricted basally, broadly articulated to the preapical segment. Scape dorsoventrally much flattened, broadened in distal 4/5 of length. Scrobe behind level of eye short and only feebly impressed. Transverse preocular groove present. Postbuccal groove broad and developed. Compound eye visible in full-face view.
Mesosoma: In lateral view, pronotum moderately arched; metanotal groove broad; propodeum slightly elevated behind metanotal groove and gently sloping toward propodeal spine. Propodeal spine subtended by a broad lamella, with the posterior margin concave in upper half. Spongiform tissue present ventrally on petiolar peduncle, lateroposteriorly on petiolar node and postpetiole.
Sculpture: Dorsum of head, mesosoma, and petiolar node densely punctate; disc of postpetiole shallowly punctate; mesopleuron and metapleuron entirely smooth; first gastral tergite basally costulate but rest of tergite smooth; first gastral sternite entirely smooth.
Pilosity: Setae on leading edge of scape, consisting of alternating simple and narrowly spatulate setae, curved toward the scape apex. Apicoscrobal seta absent. Dorsolateral margin of head and pronotal humeri without specialized projecting setae. Cephalic dorsum with very short blunt erect pubescence. A row of longer erect simple setae close to the occipital margin. Mesosomal dorsum with sparse erect pubescence. Erect setae on petiole, postpetiole, and first gastral tergite short and stiff, truncated or very slightly expanded apically.
Color: Head and mesosoma dark orange with a slightly darker gaster.
Gyne. Mandibles similar to that of worker. Head capsule similar in size and shape to worker head capsule, but with three ocelli present. Compound eye visible in full-face view. Pronotal humeral seta absent. In lateral view, mesonotum weakly convex; mesoscutum rather bulbous; flight sclerites present; wing scars present. Propodeal spine, spongiform tissue on petiole and postpetiole as in worker. First gastral sternite completely smooth. Pilosity and color as in worker.
Male. Unknown.
Comparative notes
This species fails to key out at couplet 3 of the online key to the Strumigenys of East Asia (Bolton 2000, adapted by AntWiki 2023a). The key to find the new species provides the comparison with the emmae and eurycera species groups. Even though the new species does not fit in either of these groups of species possessing four-segmented antennae, we refrain from proposing a new species group to contain S. silangan sp. nov. at this time.
Etymology
The species epithet, silangan, (meaning “east” in Filipino), is an invariant noun in apposition and refers to the fact that all the specimens were collected from the eastern part of the Philippines. The Filipino language, of the Austronesian language family, is genderless.
Figure 1.
Distribution map of Strumigenys silangan sp. nov. The type locality, Mt. Isarog, is indicated by a gold diamond. Elevation is indicated by false colors where blue is lowest, yellow intermediate, and red is highest.
Figures 2–5.
Holotype worker of Strumigenys silangan sp. nov. (2) Full face view; (3) lateral view; (4) dorsal view; (5) labels. Images copyright by Museum of Comparative Zoology, Harvard University and the President and Fellows of Harvard College, and noncommercial use allowed under CC BY-NC-SA 4.0 Deed (Creative Commons 2024).
Figures 6–9.
Paratype dealate queen of Strumigenys silangan sp. nov. (6) Full face view; (7) lateral view; (8) dorsal view; (9) labels. Images copyright by the National Museum of the Philippines (PNM), and noncommercial use allowed under CC BY-NC-SA 4.0 Deed (Creative Commons 2024).
Key to Find S. silangan sp. nov. among the Species Groups with Four-Segmented Antennae
1. Setae on dorsum of head spatulate to orbiculate; anterior clypeal margin very broad; leading edge of scape with spatulate setae Strumigenys emmae group
1′. Setae on dorsum of head simple, never spatulate or orbiculate; anterior clypeal margin broad; leading edge of scape at least with simple setae 2
2(1). In lateral view, dorsal margin of propodeum anteriorly flat but steeply sloping downward to propodeal tooth; first gastral sternite longitudinally costulate Strumigenys eurycera group
2′. In lateral view, dorsal margin of propodeum slightly elevated and gently sloping down to propodeal tooth; first gastral sternite entirely smooth Strumigenys silangan sp. nov.
DISCUSSION
Strumigenys silangan sp. nov. is the 38th species of the genus known from the Philippines, suggesting a high diversity of forms in the archipelago (AntWiki 2023b). This is also the 16th species of Strumigenys known to possess four-segmented antennae. Undoubtedly, many more species will be encountered since many islands and natural habitats remain unexplored for ants.
ACKNOWLEDGMENTS
We are indebted to the senior author's uncle, the late Rodolfo C. General, who permitted us to conduct the transect studies in his private reforestation project. We are very grateful to the students of the Ateneo de Naga University (AdeNU) who assisted in the field and laboratory work and the Institute for Environmental Conservation and Research of AdeNU for the use of their facilities and equipment during our transect studies. We are also grateful to Barry Bolton for sharing his valuable insights on the placement of this species in a monotypic species group. We also thank Douglas B. Booher and an anonymous reviewer for their constructive comments that improved this contribution. We also thank Charles Whittemore Farnum of the MCZ and Perry Buenavente of the PNM for creating the images of the holotype worker and paratype queen, respectively. We are also grateful to Perry Buenavente of PNM for creating the image of the labels.
Finally, the senior author is grateful to the Harvard University MCZ Grant Committee for providing an Ernst Mayr Travel Grant for his visit to the MCZ in 2023 and to Crystal Maier and her staff for allowing him access to the MCZ Ant Room.
© The President and Fellows of Harvard College 2025.
English (US) ·