Leptanilloides chihuahuaensis

Joe A. Macgown, Terence L. Schiefer & Michael G. Branstetter, 2015, First record of the genus Leptanilloides (Hymenoptera: Formicidae: Dorylinae) from the United States, Zootaxa 4006 (2), pp. 392-400: 394-399

publication ID

http://doi.org/ 10.11646/zootaxa.4006.2.10

publication LSID

lsid:zoobank.org:pub:C9E0FAA3-324C-404D-8A1C-3B4AC6DE663E

persistent identifier

http://treatment.plazi.org/id/FA729970-A22E-FFAE-189F-86F3C13ACB24

treatment provided by

Donat

scientific name

Leptanilloides chihuahuaensis
status

 

Leptanilloides chihuahuaensis  

( Figs. 1, 2)

Holotype male: [ United States] TEXAS, Jeff Davis Co., Davis Mts. St. Pk., 5118 ', 30 ° 35 ' 27 "N 103 ° 56 ' 26 "W, 22– 25 July 2014, T.L. Schiefer, Malaise trap on hillside on highland grassland, W.H. Cross Expedition. [Holotype deposited in Harvard Museum of Comparative Zoology ( MCZC), Cambridge , MA]. Specimen code MEM 207971. GoogleMaps   Paratype males: Two point-mounted specimens with the same data as the holotype but with unique specimen code: MEM 207972 and MEM 207973 [both specimens in Mississippi Entomological Museum ( MEM), Mississippi State, MS, USA]   ; and three specimens, one point-mounted and two in 95 % ethanol, with the following collection data: [ United States] TEXAS, Jeff Davis Co., Davis Mts. St. Pk., 4915 ', 30 ° 36 '08"N 103 ° 54 ' 55 "W, 21–25 July 2014, T.L. Schiefer, Malaise trap in riparian area in desert scrub/grassland, W.H. Cross Expedition with unique specimen codes: MEM 207974 [point mounted specimen deposited in MEM], MEM 207975 [preserved in 95 % ethanol for DNA study, deposited in MEM] and MEM 207976 [preserved in 95 % ethanol for DNA study, deposited in MEM] GoogleMaps   .

Holotype measurements and indices: HW 0.24, HL 0.24, EL 0.11, MDL 0.13, SL 0.08, PedL 0.06, FlaLI 0.04, FlaLII 0.04, FlaLXI 0.12, MH 0.29, ML 0.46, PrW 0.20, PW 0.09, PL 0.07, AIIIW 0.14, AIIIL 0.07, AIVW 0.20, AIVL 0.09, FFeW 0.05, FFeL 0.23, HFeL 0.24, HTiL 0.26, FWL 1.46, CI 100, MI 63, PI 128.

Additional paratype measurements and indices (two males): HW 0.24, HL 0.22, EL 0.09–0.11, MDL 0.13, SL 0.09, PedL 0.06–0.07, FlaLI 0.04, FlaLII 0.04, FlaLXI 0.11–0.12, MH 0.30–0.31, ML 0.45–0.46, PrW 0.18, PW 0.07–0.08, PL 0.06–0.07, AIIIW 0.15, AIIIL 0.07–0.10, AIVW 0.18–0.19, AIVL 0 0.10, FFeW 0.05, FFeL 0.21, HFeL 0.23, HTiL 0.26, FWL 1.38, CI 109, MI 65–69, PI 114–117.

Etymology. Named for the Chihuahuan Desert region where specimens were collected.

Diagnosis (male). Of the six males that have been described in the Leptanilloides   genus-group, L. chihuahuaensis   is most similar to "Leptanilloidinae male 1 " from southern Mexico described by Borowiec & Longino (2011). Leptanilloides   chihuahuaensis differs from "Leptanilloidinae male 1 " by having finer and sparser pilosity, in dorsal view the petiole being widened anteriorly and posteriorly, parameres being narrower and incurved, and the hindwing having three hamuli (instead of two). Males of L. chihuahuaensis   differ from males of Amyrmex golbachi Kusnezov, 1953   by lacking submarginal cells and a stigma in the forewing, possessing a free M vein, and having a larger paramere (about as long as petiole); from L. mckennae Longino 2003 by lacking submarginal cells and a stigma in forewing and having a much shorter paramere (in side view) (Ward 2007); from L. nubecula Donaso, Vieira & Wild, 2006 by being light to medium brown instead of dark brownish-black and being much smaller ( HL 0.22–0.24 in L. chihuahuaensis   vs. HL 0.32 in L. nubecula), and from both "Leptanilloidinae male 2 and male 3 " (Borowiec & Longino 2011) by the much smaller size ( HL 0.22–0.24 vs. HL 0.30–0.33 of "Leptanilloidinae male 2 and male 3 ") and by lacking submarginal cells and stigma in the forewing. Five other male morphotypes unassociated with workers were briefly discussed by Ward & Brady (2009). Leptanilloides   chihuahuaensis differs from each of these by lacking submarginal cells and a stigma in the forewing. Geographically, L. chihuahuaensis   is the only species thus far know to occur in the United States.

Description (male): Body size minute. Color light yellowish-brown, head and margins of abdominal segments IV–VII darker, appendages (antennae, mandibles, legs) lighter. Integument mostly smooth and shiny, with simple curved to strongly curved suberect to decumbent setae; pilosity on head, mesosoma, petiole, and first gastral tergite ( AIII) scattered and not obscuring the shininess of the integument; pilosity denser on remainder of gaster and appendages.

Head in full-face view excluding eyes about as long as wide, widest above eyes; posterior corners of head evenly rounded. Eyes large, bulging, occupying almost half the side of head. Ocelli small, protruding slightly, arranged in almost an equilateral triangle, but with distance between lateral ocelli slightly greater than distance between lateral ocellus and median ocellus; distance between median ocellus and eye approximately the length of eye. Clypeus short, tranverse, lacking lamelli; lateroclypeal teeth and hypostomal teeth lacking. Mandible slender, falcate; apex blunt; masticatory margin edentate; external margin of mandible mostly evenly curved along its length; mandible tips overlap at closure.

Antennal sockets circular and exposed, located at the anterior clypeal margin with anterior edge of sockets slightly overlapping anterior clypeal margin. Antenna 13 -merous; scape, pedicel, and each flagellomere longer than wide. Scape length about twice the length of the first flagellomere, and about the combined length of the first and second flagellomere; scape subequal to the length of ultimate flagellomere; pedicel thickened, length about 1.5 times width; flagellomeres 1-11 each at least twice as long as wide; first and second flagellomeres subequal in length.

Pronotum U-shaped in dorsal view, reduced anteromedially to a thin horizontal strip, set below the level of the dorsally protruding mesonotum; pronotum triangular in lateral view, with pointed posterior apex directed toward wing base. Mesoscutum lacking notauli; parapsidal lines not discernable. Axillae depressed, not meeting medially, connected by a narrow furrow. Tegula, or tegula-like structure present at wing base, minute, inconspicuous, with 2- 3 erect setae. Mesopleuron lacking oblique transverse sulcus, not divided into anepisternum and katepisternum. Mesoscutellum prominently bulging, as seen in lateral view. Metapleural gland not discernable. Propodeum with dorsal and declivious surfaces not differentiated, evenly rounded. Propodeal spiracle small, ovate, positioned slightly below midheight of propodeum and slightly posterior to the midlength. Legs slender, mesotibia and metatibia each with two simple spurs, pretarsal claw lacking preapical tooth.

Wings with reduced venation; clear; fringed with short to longer fine setae; with numerous short, fine microsetae evenly distributed across wing surfaces and with scattered longer setae present. Fore wing: Costal ( C) vein present, about ½ the length of wing, tubular basally for less than ¼ the length of wing, then becoming nebulous before fading out completely. Pterostigma not present. Subcostal and radial veins fused forming Sc+R; located just below and parallel to costal vein, approximately the same length as costal vein and similarly tubular for about ½ of its length before becoming nebulous. Medial ( M) and cubital (Cu) veins fused ( M +Cu), nebulous, slightly less than ¼ of the wing length. Anal ( A) vein present beneath M+Cu, about the same length as M+Cu, tubular for slightly more than half of its length before becoming nebulous; apically M+Cu connected to A by nebulous crossvein cu-a; subbasal cell present between M+Cu and A. Branching upward from the top of Cu-a is the tubular abscissa Rs+M (the joined veins M·f 1 and Rs+ M). Radial sector (Rs), tubular, branches upward from Rs+M, then becomes nebulous just past the nebulous/spectral crossvein 2 r-rs (1 r-rs is absent). M branches off of Rs+M just before Rs, appearing spectral basally then continuing as a nebulous vein that does not reach the wing margin. Cu is present as a nebulous vein that branches off near the apex of A; Cu does not reach the wing margin. Posterior margin of fore wing with narrow, conspicuous fold where hamuli attach.

Hind wing with Sc+R present, tubular basally before becoming nebulous, less than ¼ the length of wing. No other veins present. Anterior margin of hind wing past midlength with a thin, but conspicuous dark stigma. Three hamuli present in the region of the stigma. Jugal lobe absent.

Petiole (abdominal segment II) subquadrate in lateral view, about as long as high or wide, not constricted posteriorly; in dorsal view anterior corners widened, sides slightly constricted, appearing shallowly concave before widening posteriorly; petiole broadly joined to abdominal segment III; petiolar spiracle located on anterior third of the segment, near anterodorsal extremity. Abdominal segment III larger than petiole, not developed as postpetiole, and not separated from abdominal segment IV by a constriction. Abdominal spiracles III, IV, V, and VI located on anterior third of tergites (abdominal spiracle VII not visible in examined specimens). Abdominal tergite VIII (pygidium) small, simple and visible dorsally, not entirely covered by abdominal tergite VII. Pygostyli absent. Abdominal sternite IX (subgenital plate) with posterior margin broadly concave, but with a median subtriangular process. Paramere (basimere + telomere) small (about the length of petiole); basimere roughly triangular, widest at base, narrowing to incurved falcate telomere that terminates in a blunt apex. Volsella triangular, with slightly rounded apex; volsella not differentiated into digitus and cuspis. Penisvalva elongate triangular, extending to about as far as paramere apex.

Analysis of COI. A 658 bp region of the mitochondrial gene COI, the barcode region, was sequenced successfully and the resulting consensus sequence was submitted to GenBank ( Table 1). Five additional COI sequences for Leptanilloides   were found and downloaded from the BOLD database ( Table 1), including three for L.

gracilis Borowiec & Longino (southern Mexico, Honduras), one for L. nubecula Donoso et al. (Ecuador), and one for an undescribed species L. MAS001 (Costa Rica). Comparing L. chihuahuaensis   to the other specimens, pairwise genetic distances were all above 8.0%, ranging from 8.2 % (L. gracilis, Honduras) to 20.1 % (L. nubecula, Ecuador).

Discussion

Of the 16 species in the Leptanilloides   genus-group, males of only three have been given species names, and of those three, only males of L. nubecula Donoso, Vieira & Wild and L. mckennae Longino were associated with their respective worker castes (Donosa et al. 2006,Ward 2007). The description of Amyrmex golbachi   was based solely on the male caste (Kusnezov 1953). Three other male morphotypes were described by Borowiec & Longino (2011), but, because they were unassociated with workers, they could not be certain of their specific identities. Therefore, Borowiec & Longino (2011) simply referred to these male morphotypes as Leptanilloidinae male 1, male 2, and male 3. Borowiec & Longino (2011) stated that "Leptanilloidinae male 1 " was likely conspecific with L. gracilis Borowiec & Longino, 2011 based on various shared morphological features, the relatively small size, and the largely sympatric distribution of "Leptanilloidinae male 1 " and workers of L. gracilis. The status of Leptanilloidinae males 2 and 3 is less clear. These male morphotypes might represent the males of described species in the group or they may be new species. While searching unidentified material in the Bohart Museum of Entomology collection ( UCDC), Ward & Brady (2009) discovered several male Leptanilloides   genus-group males that represent at least five more species in the group. These appeared to belong to Amyrmex   and possibly Asphinctanilloides   , although it is not clear whether they represented new species or were the males of described species in the group.

All of the described species and even the undescribed males discussed by Ward & Brady (2009) in the Leptanilloides   genus-group have exclusively Neotropical distributions ranging from Chiapas, Mexico (Borowiec & Longino 2011) to southeastern Brazil (Silva et al. 2013). Based on reported specimen data, members of this genus group have been collected mostly in cloud forests (Borowiec & Longino, 2011; Brandão et al. 1999; Donoso et al. 2006; Longino 2003, Silva et al. 2013). The new species described here, however, was collected in semiarid grassland habitat in the southern foothills of the Davis Mountains in western Texas, approximately 2500 km north of the next closest collection of the genus-group in Chiapas, Mexico.

The specimens of L. chihuahuaensis   were collected in Townes style Malaise traps (Townes 1972) at two locations within Davis Mountains State Park. The first location was on the south (north-facing) slope of Keesey Canyon adjacent to a hillside drainage area. The vegetation at this site was dominated by short grasses but also included scattered forbs and a few low trees and shrubs. The noncalcareous soil, of igneous origin, was shallow, well drained, and extremely rocky (Turner 1977). The second location was in Limpia Canyon on the dry floodplain of the intermittent Limpia Creek. The soils here were similar to the first site, but the vegetation differed in having taller grasses and forbs, with grasses being less dominant, and with forbs, shrubs, and small trees being more numerous. Many rocky, unvegetated areas were also scattered through this habitat. The Malaise trap at this site was located only 20 m from the north (south-facing) slope of Limpia Canyon, which had similar habitat to the first site, so it is possible that the ants in this trap originated from the canyon slope rather than the floodplain.

Other ant species collected by MEM researchers in the same localities and during the same time period that L. chihuahuaensis   was collected were typical species for this arid region. Other species collected in the same Malaise traps included Dorymyrmex flavus McCook   , Forelius pruinosus (Roger)   , Formica   gnava Buckley, Myrmecina americana Emery   , and Myrmecocystus melliger Forel   , as well as unidentified males of Crematogaster   , Pheidole   , Solenopsis   , and Strumigenys   . General collections in the same area yielded species including Camponotus festinatus   (Buckley), C. semitestaceus Snelling  (found in higher elevations), Forelius mccooki (McCook)   , Liometopum apiculatum Mayr   , Odontomachus clarus Roger   , Neivamyrmex minor (Cresson)   , N. texanus Watkins   , Novomessor cockerelli   (André), Pheidole hyatti   Emery, P. titanis  Wheeler, P. spp., Pogonomyrmex barbatus (F. Smith)   , P. imberbiculus Wheeler   , P. rugosus Emery   , and Trachymyrmex septentrionalis (McCook)   .

The climate in the vicinity of Davis Mountains State Park is marked by hot summers and cool, dry winters (Bell et al. 2014), with over 80 % of annual precipitation occurring from May through October ( NOAA 2014). The average annual precipitation at nearby Fort Davis is 44.4 cm, compared to the 20.4 to 30.5 cm of rainfall received by the surrounding lower elevation areas of Trans-Pecos Texas ( NOAA 2014; Texas State Historical Association 2010). This difference in precipitation is a due to the pronounced orographic lifting caused by the Davis Mountains, which results in an approximately 5 cm increase in annual rainfall for each 305 m increase in elevation (DeBaca 2008; Turner 1977).

The Davis Mountain range is located within the Chihuahuan Desert ecoregion, which occupies almost 650, 0 0 0 square kilometers extending from central Mexico into western Texas, central and southern New Mexico, and southeastern Arizona (Commission for Environmental Cooperation 2006; Griffith et al. 2007; Hoyt 2007). The isolating effect of the complex topography of this extensive ecoregion, which is characterized by low-lying basins and isolated mountain ranges, contributes to the Chihuanhuan Desert being one of the most diverse arid regions in the world (Bell et al. 2014; Hernández & Gómez-Hinostrosa 2005). This diversity is threatened by overgrazing, water diversion, excessive groundwater pumping, introduced species, and overcollecting of animals and plants, such as cacti, for the commercial trade (Hoyt 2002). The collection of Leptanilloides   from this region is surprising both because of the remarkable range extension it represents for the genus-group as well as the significantly different type of habitat the specimens were collected in compared to other collections of the genus-group.

Despite the fact that males of all species in the Leptanilloides   genus-group have not been described and that males collected in Texas were unassociated with a colony, we believe there is sufficient evidence for describing L. chihuahuaensis   as a new species. First, L. chihuahuaensis   has a very disjunct distribution and occurs in a markedly different habitat than other species. Second, there are obvious morphological features that distinguish L. chihuahuaensis   from other described males. And lastly, COI sequence divergence, as compared to several other species with data in GenBank, including L. gracilis, exceeds 8 %. Although we do not think that this divergence level by itself is sufficient for species status - it has been shown that intraspecific divergence can be quite high (e.g. Meyer and Paulay 2005, Wiemers and Fiedler 2007) – we do believe that the result strengthens our argument. We also feel confident in using the generic name Leptanilloides   because specimens of L. chihuahuaensis   most closely match Borowiec & Longino's (2011) description and figures of "Leptanilloidinae male 1," which they believed to be conspecific with L. gracilis. Additionally, the taxonomic status of the genera in the Leptanilloides   genus-group remains unresolved with a strong likelihood that Amyrmex   , Asphinctanilloides   , and Leptanilloides   might be synonymized with one another. If these genera were synonymized, the name Leptanilloides   would have priority.

TABLE 1. Information for six Leptanilloides specimens that have available COI sequence data: species name, country of origin, GenBank and BOLD accession numbers, length of COI sequence fragment, and uncorrected (p) distance as compared to L. chihuahuaensis.

Species Country GenBank# BOLD# COI Size (bp) p-distance (%)
Leptanilloides   chihuahuaensis USA (Texas) KT007966 n/a 658 -
L. gracilis Mexico (Chiapas) n/a ASLAM1070-11 282 10.1
L. gracilis Honduras n/a ASLAM1144-11 658 10.3
L. gracilis Honduras JF863464 ASLAM381-11 525 8.2
L. nubecula Ecuador DQ353302 GBAH2069-06 844 20.1
L. MAS001 Costa Rica KF371235 ACGAG099-11 503 10.6
MCZC

USA, Massachusetts, Cambridge, Harvard University, Museum of Comparative Zoology

MEM

USA, Mississippi, Mississippi, Mississippi State University

UCDC

USA, California, Davis, University of California, R.M. Bohart Museum of Entomology