Leptanilloides mckennae Longino

Longino, J. T., 2003, A new Costa Rican species of Leptanilloides (Hymenoptera: Formicidae: doryline section: Leptanilloidinae)., Zootaxa 264, pp. 1-6 : 2-5

Leptanilloides mckennae Longino

Plate 1

Holotype worker: Costa Rica, Puntarenas Prov., Monteverde , 1300m, 10°18'N 84°48'W, July 1995, in moist forest litter (D. McKenna) [Instituto Nacional de Biodiversidad, Costa Rica]. Specimen code INBIOCRI001281138 .

Paratypes, all same data as holotype, each pin with 2 workers and unique specimen code: INBIOCRI001281139 [The Natural History Museum, London, U.K.]; INBIOCRI001281140 [John T. Longino, personal collection, Olympia, WA, USA]; INBIOCRI001281141 [Los Angeles County Museum of Natural History, Los Angeles, CA, USA]; INBIOCRI001281142 [Museum of Comparative Zoology, Cambridge, MA, USA]; INBIOCRI001281143 [Museu de Zoologia da Universidade de São Paulo, São Paulo, Brazil]; INBIOCRI001281144 [University of California, Davis, CA, USA]; INBIOCRI001281145 [National Museum of Natural History, Washington, DC, USA]; INBIOCRI001281146 [Naturhistorisches Museum, Basel, Switzerland]; INBIOCRI001281147 [American Museum of Natural History, New York, NY, USA].

Additional workers from the type series (all nestmates) are in 95% ethanol and have been deposited at USNM and stored in liquid nitrogen to preserve DNA. These workers were stored at ambient temperatures prior to deposition at USNM.

Holotype measurements: HL 0.707, HW 0.559, SL 0.492, WL 0.942, PL 0.237, PPL 0.172.

Additional paratype measurements (two workers): specimen one: HL 0.738, HW 0.635; specimen two: HL 0.753, HW 0.631.

Etymology: named for Duane McKenna, who collected the type series. It is used here as a noun in apposition and thus invariant.

Diagnosis (worker): genal teeth lacking; dorsal face of propodeum much longer than posterior face; postpetiole smaller than petiole in lateral view; gaster with constrictions between segments.

Description (worker): head subrectangular, with gently convex sides; mandibles in side view strongly curved ventrally; masticatory margin concave, with approximately 10 evenly spaced, extremely minute denticles; masticatory margin smoothly rounding into basal margin; dorsal surface of mandible sublucid, faintly striate, with scattered piligerous puncta; clypeus short, with broadly triangular translucent anterior lamella; antennal sockets fully exposed in face view; frontal carinae closely approximated, forming a low median keel; gena lacking lateral teeth overlapping mandible bases; entire head capsule, dorsally and ventrally, uniformly and densely punctate; eyes absent; scapes when laid back reach 3/ 4 distance from antennal insertions to posterolateral vertex margins; antennae 12-segmented, second funicular segment one and a half times length of first and third segment, which are subequal in length; funicular segments gradually thicken distally to longer, fusiform apical segment.

Pronotal suture present and flexible; in profile pronotal and mesopropodeal sutures very weakly impressed, pronotum and mesonotum forming very low convexities, dorsal face of propodeum long and flat, gently sloping posteriorly before rounding into short posterior face, dorsal face twice as long as posterior face; flange over metapleural gland opening forming a blunt angle; pronotum and dorsal face of mesonotum punctate, interspaces subequal in width to puncta diameters, interspaces smooth and shiny; mesopleura and propodeum more densely, finely punctate, mat; lacking metatibial gland; middle and hind tibiae each with single pectinate spur, metatibial spur larger than mesotibial spur; claws simple.

Petiole and postpetiole very small; petiolar tergite with anterior node and long sloping posterior face; petiolar sternite with deep, rounded anteroventral process; anterior juncture of petiolar tergite and sternite forming an angular notch; anterior margins of tergite and sternite produced anteriorly with carinate rims, such that petiole has an anterior pocket that encloses the narrow presclerites that articulate with the propodeum; petiolar spiracle on anterior rim of tergite, very small and difficult to see; postpetiole somewhat barrel-shaped, tergite a half cylinder, with flat semicircular anterior face above very small neck of helcium, sternite also large, with large anteroventrally projecting tooth; postpetiolar spiracle on anteromedian side of tergite, beneath juncture of anterior and dorsal face of tergite.

Gaster elliptical, with three large, conspicuous segments (abdominal segments 4, 5, and 6), and with distinct constrictions between them; spiracle of abdominal segment 4 very small, located on lateral tergite one third distance from anterior to posterior margin, spiracles on 5th and 6th segments larger and closer to anterior margin of tergite; tergites and sternites smooth and shining.

Entire body and appendages covered with uniform length, moderately abundant, subdecumbent to appressed pubescence; somewhat longer setae restricted to mouth region and ventral margin of postpetiole; head capsule, mandibles, and pronotum red brown, grading to lighter yellow brown on propodeum, petiole, postpetiole, gaster, and appendages.

Comments: This new species is known only from the type material. Duane McKenna collected the ants in a leaf litter sample at 1300m in the Bajo del Tigre Reserve on 22 June 1995 at 7:00am. A few dozen workers were in sifted litter from a 0.5m2 plot. The Bajo del Tigre Reserve is a patch of moist forest on the Pacific slope just below Monteverde. It is an area of abrupt habitat change, from highly seasonal and somewhat xeric conditions a few kilometers downslope to cold, wet cloud forest conditions a few kilometers upslope.

Discussion

Brandão et al. (1999) proposed apomorphic characters for the subfamily and each of the two genera. The apomorphies for the subfamily, excluding the sting characters (which require dissection) were (1) presence of lateral blunt teeth on the genae, overhanging the mandibles; (2) lack of metatibial glands; and (3) an extremely small pygidium (abdominal segment 7) that is overhung and concealed by the tergite of abdominal segment 6. Apomorphies for the genus Leptanilloides were (1) dorsal face of propodeum at least two times longer than posterior face, and (2) gaster with constrictions between the segments (a constriction between abdominal segments 4 and 5 and between 5 and 6). Apomorphies for Asphinctanilloides were (1) postpetiole extremely reduced in size, smaller than the petiole in profile, and with spiracles at midlength; and (2) several derived sting characters. Asphinctanilloides has no constrictions between the gastral segments, a condition hypothesized to be plesiomorphic in the subfamily ( Brandão et al. 1999).

The new species described here exhibits a combination of characters that blurs the distinction between Leptanilloides and Asphinctanilloides and calls into question the monophyly of the two genera. Leptanilloides mckennae has the long propodeum and the gastral constrictions considered apomorphic for Leptanilloides . However, it also has a very small postpetiole and a somewhat posteriorly shifted postpetiolar spiracle, which are proposed apomorphic characters for Asphinctanilloides . The shape of the petiole and postpetiole is nearly identical to A. anae (Fig. 12 in Brandão et al., 1999). Also L. mckennae and A. anae both have the genal teeth extremely reduced or absent. In the phylogeny of Brandão et al. large genal teeth are plesiomorphic in the subfamily, and reduced genal teeth are a derived condition. Thus, this new species exhibits a mix of characters thought to be apomorphic in disparate lineages.

The subfamily Leptanilloides is entirely subterranean ( Brandão et al. 1999). Subterranean ants are notoriously difficult to collect, and they are almost certainly severely undersampled compared to other ants. Leptanilloidinae may be far more abundant than the sparse museum collections suggest, and it remains to be seen whether their distribution is a set of disjunct populations, a continuous sheet of parapatric forms throughout the Neotropics, or widespread communities of sympatric forms (two sympatric species occur near Manaus). At this stage in the inventory of the subfamily there are eight uniques (species known from one collection) and no duplicates (species known from two). The species accumulation curve is linear. This means there are many more species to be found and we have no idea how species-rich the subfamily may be. I will be comforted when someone makes a second collection of a known species. With this degree of uncertainty and strong undersampling of the taxon we can expect intra-taxon phylogenetic hypotheses to be unstable for the foreseeable future. This uncertainty leads me to eschew any formal generic level taxonomic changes in the subfamily until more species are discovered.

The discovery of L. mckennae in Costa Rica is a significant range extension for the subfamily, with the closest previous species being L. legionaria from the Sierra Nevada de Santa Marta in Colombia. Much of Costa Rica is geologically young and most of the biota is also young or recently arrived via dispersal (Coates and Obando 1996, Gentry 1982), but some old lineages do occur. Using evidence from the plant family Bignoniaceae, Gentry proposed a South America - Central America land connection at the beginning of the Tertiary, during which some lineages dispersed northward. A long period of isolation ensued, followed by the most recent land connection and a second wave of immigration. Brady (2003) has shown that at least some lineages in the doryline section date from the mid-Cretaceous and thus the Leptanilloidinae could be very old as well. Leptanilloides mckennae may have dispersed to Costa Rica following the closure of the Panamanian isthmus a few million years ago, or it could be an ancient faunal element that arrived at the beginning of the Tertiary or before.