Strumigenys flavianae, Chaul, 2023

Chaul, Júlio Cezar Mário, 2023, A New Ant Species of Strumigenys Smith, 1860 (Hymenoptera: Formicidae) from the Brazilian Atlantic Forest, Taxonomy (Basel, Switzerland) 3 (2), pp. 221-231 : 225-230

Strumigenys flavianae Chaul sp. nov.

( Figures 1–3)

Type material. Holotype worker. BRAZIL, Minas Gerais, Viçosa, Mata do Paraíso , -20.803959 -42.855107, 13.ii.2017, Berlesate (Borlini, P.) ( CELC, unique specimen identifier: ANTWEB1032460 ). Paratype workers: BRAZIL, Minas Gerais, Viçosa, Fragmento Florestal , 1994 (Sperber, Lopes and Louzada) ( CELC, unique specimen identifier: ANTWEB1032112 ) ; BRAZIL, Minas Gerais, Viçosa, Mata do Paraíso , -20.803146 -42.856782, 21.viii.2017, winkler sample (Moura, M. N.; Micolino, R.) ( CELC, unique specimen identifier: ANTWEB1032422 ) .

Diagnosis. Differing from all other Strumigenys by the combination of a slender body and long legs; pilosity composed of subflagellate, flagellate, or wire-like setae; basal lamella of mandible triangular, entirely covered by anterior clypeal margin when mandible is closed, and without diastemmic gap between it and the basal tooth of the masticatory margin; mandible outer margin confluent with clypeal lateral margin, not bulging; spongiform tissue ventrally on the petiole, notched at about its midlength; sides of mesosoma mostly sculptured, with only a small smooth patch in between upper mesopleuron and upper metapleuron.

Geographical Distribution. Minas Gerais, Brazil.

Description. HW 0.42–0.43, ML 0.08, ML2 0.18–0.21, HL 0.81–0.84, SL 0.39–0.4, ON 12, HD 0.28–0.31, WL 0.83–0.85, PetL 0.45–0.46, A3+A4L 0.72–0.8, PrW 0.29–0.305, DPetW 0.14–0.155, DPetndL 0.2–0.23, PosPetW 0.25–0.27, GW 0.5–0.53, MtfrL 0.635–0.68, MttbL 0.44–0.47, MtbtL 0.48–0.52, TL 3.02–3.16, LegL 1.55–1.67, CI 51.19–51.85, SI 92.86–93.02, MI 9.7, LegI 51.15–52.85 (n = 3). Head: head elongate ( Figure 1 A). In full-face view, mandibles small, elongate triangular, its dorsal surfaces smooth, covered with small, simple, appressed hairs. Outer margins of mandibles confluent with lateral clypeal margins. Basal lamellae of mandibles triangular, its basal side slightly tilted dorsally. Masticatory margin with 12 teeth, the first originating without a diastemmic gap between it and the basal lamella. The 6 basal teeth are large and acute, growing in size until tooth 3 (the largest on the mandible) and then becoming gradually smaller until tooth 6. Tooth 7 is even smaller than 6, and can be triangular or truncate. Teeth 8 to 11 are small and truncate, and tooth 12, the apical, is slightly larger than the previous, and triangular (but still smaller and less acute than the basal 7 teeth) ( Figure 2 B). Tips of mandibles downcurved, so that the row of small truncated teeth are best seen with the head in anterodorsal view. Basiventral glands on mandibles not visible. Basal half of labrum (excluding the apical lobes) with margins diverging and converging back, and then mildly convex along the second half of the labrum body. Lobes short and thick, about a quarter the length of the labrum disc, converging slightly towards each other and separated by a wide notch. Each lobe has three simple hairs and a longer, flattened and sinuous seta arising from the apex ( Figure 2 D, G). Clypeus longer than wide, reticulate punctate, its anterior margin projected strongly convex over mandibles and covering basal lamellae of mandibles. Antennae 6- segmented. Scape apicoventral gland present. Scape covered with appressed fine hairs and a row of freely projecting long, curved to coiled and wire-like setae on its anterior margin, most of them not obviously curved basally or apically ( Figures 1 A and 2 C). Apical and preapical antennomeres, taken together, slightly longer than scape; preapical antennomere as thick as apical, not swollen. Dorsal head reticulate punctuate with superimposed fainting rugulae, and ground pilosity composed of simple, irregularly coiled to subflagellate hairs. A pair of long, wire-like to flagellate apicoscrobal setae. Vertexal standing setae are absent, although ground pilosity are somewhat longer posteriorly on the head. In lateral view, scrobe not conspicuously deep, its dorsal margin well-marked, apically and ventrally poorly marked; ventral margin fades anterior to the eye level without reaching the anterior end of the head; in full-face view, dorsal margin of scrobe forms a thin translucent cuticular projection. Compound eyes with about 12 ommatidia, not seen in full-face view. Preocular carina ends much anterior than the level of the eyes in profile. Mesosoma: slender and elongate ( Figure 2 A). Dorsally, promesonotal suture a feeble ridge. Pronotum reticulate punctate with superimposed fainting rugulae, mesonotum reticulate punctate. Metanotal impression, a shallow groove, anteriorly marked by thin feeble traversal carina and with a short median longitudinal carina across it. Dorsum of propodeum, mesopleura, metapleura and lateral propodeum reticulate punctate, with only a small smooth patch in between upper mesopleuron and upper metapleuron. A pair of thin triangular lamellate projections laterally on mesonotum, on top of mesothoracic spiracles, directed laterally and upwards ( Figure 2 F). Humeral pair of long, wire-like setae arising from a pair of small indented protrusions. Similar, but slightly smaller and thinner, setae are also present anterolaterally on the pronotum posterior to the humeral pair, anterolaterally on the mesonotum, and mediolaterally on the mesonotum (arising from the tip of the mesonotal lamellate projections). Ground pilosity on dorsum of mesosoma, as on the head, composed of thin, long subflagellate hairs. Propodeal teeth sharp and long, in profile posteriorly and slightly dorsally inclined, subtended by a thin propodeal lamella which descends with regular width until the level of the metapleural gland reservoir, where the lamella forms a small round lobe ( Figure 2 A). Mesothoracic ventral excavation not densely hairy. In lateral view, anterior margin of mesopleuron with a translucid lamella, which is lobate on the upper section, pointing towards the mesothoracic excavation ( Figure 2 F). Legs elongate, reticulate, covered in hairs similar to the head and mesosomal ground pilosity. All pairs of legs with tiny, rounded to oval, apicodorsal femoral glands; hind femur with the largest and middle femur with the smallest ( Figures 1 D and 2 E). Slit-shaped, well-developed apicodorsal glands on hind and middle tibiae, and tiny oval gland on protibia. Among the tarsomeres, those with visible glands are probasitarsus, mesobasitarsus and mesotarsomeres 2, and metabasitarsus and metatarsomeres 2 and 3. Metasoma: petiole elongate, node longer than high, evenly and gently curved in profile view. The anterior face of node gradually ascending from peduncle and merging into dorsal faces, these faces not obviously distinguishable ( Figure 2 A). Postpetiole disc with longitudinal rugae which are well-marked laterally and shallow to fainting medially ( Figure 1 C). Spongiform tissue strongly developed ventrally on the petiole as a thin and tall curtain, notched at about its midlength ( Figure 2 A). Posterior edge of node of petiole with a thin strip of spongiform tissue connecting two pyramidal spongiform posterolateral lobes. Spongiform tissue well-developed ventrally on postpetioles, with two large and dense lateral outgrowths linked in the middle by a thinner strip. Postpetiole disc with a thin, transversal, spongiform anterior strip and large spongiform lobes posterolaterally, which gradually becomes thinner where they link posteromedially. Well-developed spongiform transverse strip anteriorly on first gastral tergite. Basigastral ventral patch of specialized setae reduced. Dorsum of metasoma covered in long, wire-like to flagellate hairs. Gaster smooth and shiny. Gastral sternites as well as dorsally around the sting with thin, curved and simple hairs. Basigastral costulae well-marked but small, about half the length of postpetiole disc ( Figure 1 C).

Etymology. The specific epithet honors the late Flaviana Heloisa da Silva Caetano, a young Brazilian undergraduate student from the Universidade Federal de Viçosa, whose untimely death during the COVID-19 pandemic saddened all her colleagues in the biology course. The name was created by adding the singular Latin genitive case suffix -ae to the first name of a female person. The orthography of an eponym is unchangeable and does not depend on the generic name for which the epithet is used.

Biology. Nothing is known about the biology of living colonies of S. flavianae ; however, its slender body and legs suggest that it forages on the upper stratum of the litter or maybe within cavities inside rotten logs rather than amidst the deeper and denser litter and soil stratum. The new species has been found three times in secondary growth forest remnants, once in a pitfall trap (specific forest remnant not determined on label) and twice by winkler extractors (both in “Mata do Paraíso” forest remnant). Despite years of sampling in the region, the species is still rare. It might be the case that its populations are low in the most often sampled remnant, Mata do Paraíso (most sampling efforts throughout the years have been undertaken there). Further explorations of other forest remnants (dozens exist in the municipality) as well as in larger, more preserved nearby forests, such as the Parque Estadual da Serra do Brigadeiro or Parque Nacional do Caparaó, might reveal other, hopefully more abundant, populations.

Comments. The new species stands out among all species occurring in the Brazilian Atlantic Rainforest for its slenderness (long head, slender mesosoma and petiole and long legs) and wire-like pilosity. In this biome, species with short mandibles (either gripping or short trap-jaw mechanisms) are never too slender and never have the same type of pilosity. Some species of long trap-jaw mandibles in the mandibularis group are as slender as S. flavianae , but in that case, they cannot be confused with the newly discovered species because of their conspicuous linear mandibles (and also due to the lack of wire-like pilosity). A few species in the probatrix- and hyphata -groups, mostly Amazonian, are similar to S. flavianae ; the ones in the probatrix -group are even more slender; however, due to other traits, they also cannot be confused with S. flavianae (see Discussion).

4. Discussion

Identification. While keying specimens of S. flavianae , the coiled setae on the anterior margins of the scapes make the decision ambiguous as to whether the setae curve to the base or the apex of the scapes, an important couplet in the key (couplet 31 of “Key to Neotropical Pyramica species ” [ 9]). An attempt to insert S. flavianae into the key can be made by modifying couplet 31 to a triplet, as follows (the first two lugs reproduced exactly as in Bolton, 2000):

—With the head in full-face view, the leading edge of the scape has a row of conspicuous projecting curved hairs, of which one or more, distal to the subbasal bend, distinctly curves towards the base of the scape. These hairs may be simple, spatulate, remiform, spoon-shape or wire-like ........................................................................................................... 32

—With the head in full-face view, the leading edge of the scape lacks projecting hairs that distinctly curve towards the base of the scape. The scape’s edge may have elongated, simple, straight or flagellate hairs present, may have entirely anteriorly or apically directed short hairs (hairs simple, narrowly to broadly spatulate or spoon-shaped) or may be hairless ............................................................................................................................................. 56

—With the head in full-face view, the leading edge of the scape has a row of conspicuous projecting wire-like hairs which have tips curving towards the base or the apex of the scape (or even ventrally or dorsally curved) ( Figure 2 C). Head elongate ( Figure 1 A). Spongiform tissue notched ventrally on the petiole ( Figure 2 A) ... S. flavianae

The ambiguity of this species to pass smoothly through the original couplet 31 of the key, and the fact that many other species have been described since 2000 highlight the necessity of an updated identification work that incorporates all species described into a new identification tool. This could be carried out either by one large key that accounts for all Neotropical species or by a key that leads to the species groups, with each group having its own specific key updated separately.

Morphologically similar Neotropical species. Strumigenys flavianae is an odd Neotropical representant, not fitting perfectly into any of the species groups delimited for the genus [ 9]; however, it is similar to the members of the hyphata -group, S. hyphata (Brown, 1953) [ 4] and S. cincinnata (Kempf, 1975) [ 30], and to S. doryceps (Bolton, 2000) [ 9], a member of the probatrix -group.

The new species overall slenderness suggests that it could be a member of the probatrix - group, as it is very similar to S. doryceps . It differs from S. doryceps by having a short and triangular basal lamella of the mandible, which is not visible when mandibles are fully closed, rather than a long, subrectangular one that is visible with the mandibles closed. It also differs from that species due to not having the outer mandible margin basally bulging in relation to the lateral clypeal margin just above it; instead, in S. flavianae , the margin is confluent with the lateral clypeal margin. The dentition in S. flavianae is not confined to the apical third of the mandible ( Figure 2 B) as in S. doryceps . The habitus between the two is similar, however, S. doryceps having a more slender mesosoma and much larger smooth pleural patches ( Figure 3 A, B, D, E). Two interesting similarities between S. flavianae and S. doryceps are the well-developed and medially notched spongiform tissue located ventrally on the petiole and the short but well-marked basigastral costulae. Strumigenys flavianae differs more radically from S. probatrix , the other member of the probatrix -group, in many obvious aspects. Except for slenderness, they do not have a similar habitus: S. flavianae has freely projecting setae on the anterior clypeal margin, which are not tidily arranged and spatulated as in S. probatrix ; it has dense and abundant spongiform tissue located ventrally on the petiole (moderate in S. probatrix ); and S. flavianae is ferruginous brown rather than dark brown. However, in terms of group-defining characters, all dentition differences between S. flavianae and S. doryceps are also present between it and S. probatrix (while S. probatrix matches S. doryceps in all of these dentition traits).

When it comes to the hyphata -group, a relationship between S. flavianae and the two members could be hypothesized due to the overall similar habitus, but not particularly the slenderness. In this case, the shapes of the basal lamellae (small and triangular) of S. hyphata and S. cincinnata match that of S. flavianae . Still, the dentition that immediately follows the basal lamella is not exactly equal; in the hyphata -group, there is a diastema between the basal lamella and the teeth which follow it ( Figure 3 C). Moreover, the labrum in the hyphata -group is unique, with the main body being roughly hourglass-shaped (imaged under specimens ANTWEB1032490 and UFV-LABECOL-000095 on Antweb.org), not resembling that of S. flavianae ( Figure 2 G). At the species level, S. flavianae cannot be confused with either of the two members of the hyphata -group. It can be easily differentiated from S. cincinnata by its much larger size, and from S. hyphata by size and by the absence of hypertrophied specialized clypeal setae, among other aspects.

In a recent phylogenetic study of the genus, the species S. probatrix and S. cincinnata were included [ 8]. The two, despite being recovered as members of the same subclade within a larger Neotropical clade, are not particularly closely related, with S. probatrix being the sister of a clade containing members of the paradoxa -group and the rogata -group and S. cincinnata being the sister of the exquisite S. mirabilis Mann, 1926 [ 8, 31]. With such a robust phylogenetic backbone available, investigating what position S. flavianae would assume once incorporated into it, is surely a goal that must be attained.

Rarity. Despite relatively constant sampling efforts in the past thirty years at the type locality, only three workers are known for the new taxon. Geographically, S. flavianae is far from its morphologically similar species, occurring in the southeastern to eastern South America Dry Diagonal (eSADD, sensu Luebert [ 32]), while the other species in the hyphata- and probatrix -groups all occur northwest of the eSADD or, at most, within it ( S. hyphata occurs in the Goias state of Brazil and in Bolivia) ( Figure 4). In a large study of Strumigenys in the Brazilian Atlantic Forest, no species similar to S. flavianae were recorded [ 33]. Moreover, Strumigenys specimens of various collections containing Atlantic Forest ants were examined by the author and myrmecologists working on large ant collections in Brazil confirmed the absence of specimens matching S. flavianae (Thiago R. Silva and Esperidião Neto, personal communication).

It is not uncommon that Strumigenys species are represented by only one or a few records, even in the case of species described long ago and/or in regions which have received considerable sampling efforts since their descriptions. Some examples of such species in the Neotropics are S. mandibularis Smith, 1860 [ 34], S. dentinasis (Kempf, 1960) [ 35], S. substricta (Kempf, 1964) [ 36], S. browni (Bolton, 2000) [ 9], S. xochipili (Bolton, 2000) [ 9], S. tlaloc (Bolton, 2000) [ 9] and S. warditeras (Bolton, 2000) [ 9]. Strumigenys doryceps itself, the species discussed just above, had been known by the Ecuatorian type series for almost two decades, but was only recently recorded again in Colombia [ 37].

Putative male sex. The male morphospecies Strumigenys ufv-66 ( Figure 5 and additional images under specimen UFV-LABECOL-009731 on Antweb.org), sampled in a Malaise trap, is possibly conspecific to S. flavianae . The morphospecies Strumigenys ufv-66 occurs in the type locality of S. flavianae , where six other unassociated male Strumigenys morphospecies are known (totaling seven male morphospecies out of twenty species known by workers in that locality). All of these males are databased and photographed, and can be accessed at Antweb.org. These are: ufv-42 (UFV-LABECOL-008242), ufv-45 (ANTWEB1032841), ufv-46 (ANTWEB1032780), ufv-62 (UFV-LABECOL-009966), ufv-65 (ANTWEB1038468) and ufv-111 (ANTWEB1041072). None of them appear to better match S. flavianae than Strumigenys ufv-66. It is larger than other male morphospecies, which is in accordance with the new species (also a relatively large Strumigenys ). Moreover, it shows similar development of the spongiform tissue on the waist segments, including a mild notch on the ventral petiole spongiform tissue that partially matches the corresponding state seen in workers of S. flavianae . Finally, it has a conspicuous propodeal spine, also in accordance with the workers of the new species.