Leptothorax athabasca Buschinger & Schulz, 2008, sp.n.

Taxonomy of Leptothorax athabasca sp.n.

Etymology. The name refers to the type locality. The specific epithet is to be treated as a noun in apposition. Type locality. On 27 July 1993, seven colonies have been collected in rock crevices on the south-exposed river bank just upriver Athabasca Falls of Athabasca River, Jasper National Park. The site is at 52° 39' 55" ' N, 117° 52' 58" W, at an elevation of c. 1200 m a.s.l. (as indicated by Google Earth).

The habitat is the river bank that sometimes is evidently flooded. A horizontally split, schist-like sandstone cliff is exposed there. The ants were exclusively found in rock crevices, whereas above the flood line a number of related species of Leptothorax were dwelling dead wood, or a layer of conifer needles and debris beneath small flat rocks. Mainly Leptothorax retractus Francoeur, 1986 could be found there. Close to the river and in the collecting site the coniferous forest was comparatively open.

Type material. Holotype gyne, 11 paratype gynes, 16 paratype workers, 10 paratype males.

Measurements of worker (n = 13). HL 697 ± 30 (656 - 760); HW 598 ± 23 (570 - 637); SL 457 ± 14 (437 - 475); MW 416 ± 23 (390 - 456); PSL 119 ± 15 (95 - 143); PEL 239 ± 13 (219 - 257); PEW 180 ± 10 (162 - 190); PPW 255 ± 14 (228 - 285); ML 864 ± 41 (808 - 931); PEH 264 ± 14 (247 - 290); HS 647 ± 26 (613 - 698); SL / HS 0.706 ± 0.019 (0.676 - 0.737); HW / HL 0.858 ± 0.020 (0.822 - 0.884); MW / ML 0.482 ± 0.008 (0.466 -0.495); PSL/ML 0.138 ±0.013 (0.113 -0.158); PEH / PEL 1.107 ± 0.057 (1.000 - 1.217); PEW / PEL 0.755 ± 0.047 (0.667 - 0.833); PEW / PPW 0.707 ± 0.028 (0.667 - 0.760). PSI 1.3 - 1.6 (n = 3).

Description of worker. Total length 3.2-3.3 mm; 11 antennomeres, as characteristic for the genus. Head (Fig. 1) subrectangular, with evenly rounded occipital corners. Head width equal in front of and behind the eyes. Eyes comparatively small, situated at midpoint of head. Frontal triangle distinctly delimited and well depressed. Scapes short, not reaching occipital corners. Mesosoma flat, outline in lateral view over a long distance straight, with shallow mesometanotal depression. Mesonotum at place where wings ofalate would insert with slight, but in dorsal and lateral aspect well visible crests; more prominent than in congeners. Propodeal spines broadly attached, longer than broad, in lateral view more or less straight caudad oriented, with pointed tips. In dorsal view, spines slightly convex, distally only minimally divergent. Propodeal spine index (cf. Buschinger1966) 1.3 - 1.6, the spines thus being comparatively short. Petiole short and high, node triangular, its anterior face markedly concave (Figs. 3, 4, 6) in lateral view, its posterior face also slightly concave, straight, or appearing a little bit convex with straight or slightly concave outlines. Top of node acute. In dorsocaudal view, upper part of node distinctly convergent, with sharp, medially distinctly impressed ridge. Anterior part of dorsal petiolar outline with prominent corners, well visible in dorsal or lateral view. Subpetiolar process well developed, pointing anterioventrad. Postpetiole rounded and lacking any conspicuous ventral appendage. Erect hairs relatively short (60 - 80 µm). Scapes only with subdecumbent, not erect hairs. Tips of hairs blunt. Petiole with a total of c. 8, postpetiole with c. 10 erect setae. Sculpture of head, mesosoma and waist mainly densely reticulate, without any stronger rugae or other sculpture elements. Color of body evenly dark brown, without any lighter spots, appendages distinctly lighter, mainly light brown to orange-brown.

Measurements of gyne (n = 10. Holotype in square brackets). HL 713 ± 18 (675 - 732) [675]; HW 600 ± 19 (570 - 618) [570]; SL 457 ± 13 (437 - 475) [447]; ED 166 ± 19 (133 - 190) [147]; MW 495 ± 13 (475 - 513) [475]; PSL 114 ± 8 (105 - 124) [109]; PEL 264 ± 16 (233 - 285) [233]; PEW 184 ± 9 (171 - 200) [171]; PPW 267 ± 16 (238 - 285) [238]; ML 1009 ± 31 (944 - 1042) [944]; PEH 279 ± 9 (266 - 295) [266]; HS 656 ± 17 (622 - 670) [622]; SL / HS 0.696 ± 0.015 (0.667 - 0.718) [0.718]; ED / HS 0.253 ± 0.028 (0.199 - 0.286) [0.237]; HW / HL 0.843 ± 0.016 (0.818 - 0.867) [0.845]; MW / ML 0.491 ± 0.011 (0.474 - 0.507) [0.503]; PSL / ML 0.113 ± 0.008 (0.100-0.122) [0.116]; PEH/PEL 1.060 ± 0.040 (1.033 - 1.143) [1.143]; PEW/PEL 0.698 ±0.022 (0.655 -0.735) [0.735]; PEW/PPW 0.689 ± 0.033 (0.650 - 0.750) [0.720]. PSI 1.3 - 1.6 (n=4).

Description of gyne. Total length 3.0 - 3.8 mm; head (Fig. 2) as in worker, outline below eyes more parallel sided than in worker. Occipital corners distinctly rounded. Mesosoma(Figs. 5, 7) slender, not bulky, particularly flat as compared to other congenerics. Dorsal outline straight. Propodeal spines as in worker. Petiole and postpetiole similar to worker, but in dorsocaudal view ridge medially only slightly impressed. General sculpture as in worker; but frons of head with many fine striae, only with scattered reticulate ground sculpture. Occipital corners nearly unsculptured and shining. Mesonotum longitudinally diffusely striate, with some reticulation and larger unsculptured parts. Scutellum reticulate, unsculptured and shining in medial part. Color and hairs as in worker.

Measurements of male (n = 7). HL 609 ± 28 (570 - 656); HW 633 ± 29 (589 - 675); SL 238 ± 11 (219 - 247); 2FL 221 ± 9 (209 - 238); ED 249 ± 24 (209 - 276); MW 677 ± 58 (580 - 732); PEW 214 ± 19 (190 - 238); PPW 269 ± 23 (238 - 304); ML 1250 ± 96 (1102 - 1359); HS 621 ± 28 (580 - 665); HW / HL 1.039 ± 0.023 (1.015 - 1.077); SL / 2FL 1.079 ± 0.076 (0.960 - 1.182); SL / HS 0.383 ± 0.026 (0.329 - 0.410); PEW / HS 0.345 ± 0.032 (0.286 - 0.378); PPW / HS 0.433 ± 0.041 (0.371 - 0.474); MW / ML 0.543 ± 0.048 (0.441 - 0.579); PEW / PPW 0.797 ±0.059 (0.741 -0.920).

Description of male (Fig. 8). Total length 3.7-4.1 mm; 12 antennomeres, as characteristic for the genus. Head wide, behind eyes distinctly wider than anterior. Eyes large and slightly oval. Mesosoma dorsally rounded, pronotum low and small. Mesonotum and scutellum strongly vaulted, metanotum short, propodeum rounded, without any angles or spines. Petiole variable, mainly flat and long, sometimes higher and shorter, node evenly rounded. Postpetiole shorter than petiole, with rounded semicircular dorsal outline. Waist segments without ventral appendages. Color of body totally dark brown to black, antennae dark brown, legs light brown to orange-brown. Body surface mostly smooth and shining except for head being roughly and irregularly rugoreticulate; scutellum and dorsal surface of propodeum slightly reticulate, with shiny parts; in general, lateral surfaces of mesosoma very diffusely reticulate, but always shiny.

Differential diagnosis. A detailed differential diagnosis cannot be provided because of the desolate condition of Leptothorax taxonomy in North America (see Discussion). A comparison is possible only with a few wellknown species and with the sympatric forms in the vicinity of the type locality.

The measured values are by no means unusual among related Leptothorax species. Leptothorax athabasca is a bit larger than the European L. muscorum (Nylander, 1846), and the syntopic L. retractus, but smaller than L. acervorum (Fabricius, 1793) and the North American Leptothorax "sp. B" sensu Heinze & Buschinger (1987), Heinze (1989b) and Loiselle & al. (1990).

Propodeal spine length (PSL, Gusten & al. 2006) has not been measured in other species of Leptothorax. The propodeal spine index (PSI; Epinotal Spine Index in Buschinger1966) can be compared among a few species. In L. athabasca gynes it is 1.3 - 1.6; in L. acervorum (Europe) 1.87 ± 0.06 (Buschinger 1966); in L. muscorum (Europe) 1.73 ± 0.09 (Buschinger 1966); in L. gredleri Mayr, 1855 (Europe) 1.48 ± 0.06 (Buschinger 1966); in L. scamni Ruzsky, 1905 (Caucasus and northern Turkey) c. 2.0 (Heinze & al. 1993); in L. pocahontas (Canada) 1.7 - 1.8 (Buschinger 1979); and in L. faberi (Canada) 1.5 -1.8 (Buschinger 1983).

The color of L. athabasca gynes and workers is evenly dark brown (see Figs. 1 - 7), similar to L. acervorum (but lacking the lighter areas on the body of that species), lighter than in the sympatric Leptothorax "sp. B", and darker than in L. retractus from the same site.

Sculpture and pilosity are quite similar to other North American species of Leptothorax. Some negative characters may be helpful for identification, though:

Leptothorax pocahontas has long, tapering hairs and, in the typical case, a smooth and shiny surface (however, "dull" gynes with shorter hairs have been found and provisionally attributed to this species, cf. Buschinger & Heinze 1993).

Leptothorax retractus is characterized by a small but clearly visible notch in the anterior margin of the clypeus. This notch lacks in L. athabasca.

Leptothorax sp. C (sensu Heinze & Buschinger 1987, Heinze 1989b and Loiselle & al. 1990), supposed to be the host species of L. pocahontas, is much lighter in coloration. Higgins (year unknown) suggested that Leptothorax sp. C from Jasper NP is identical to Leptothorax muscorum var. septentrionalis Wheeler, 1917, though this taxon still is considered a junior synonym of L. muscorum according to antbase.org and Bolton & al. (2007).

What remains as characteristic for L. athabasca is the very sharp crest on top of the petiole (Fig. 4), the flat mesosoma in the female castes (Figs. 3, 5), and the dorsolateral small outgrowths of the worker mesonotum.

Life history data of Leptothorax athabasca sp.n.

Two among the seven colonies that were collected were polygynous with two and four reproductive queens (checked by dissection; spermathecae full of sperm and ovaries well developed, yolky oocytes present); the species thus is probably facultatively polygynous like most other congenerics. The colonies collected on 27 July 1993 had sexual pupae in various states of pigmentation, and a few adult males. Colony size was about 50 to 100 adults.

Records of other interesting Formicoxenini from Jasper National Park

In addition to L. athabasca the visit in Jasper NP in July 1993 revealed a couple of colonies of very rare, unexpected or recently described species. Leptothorax faberi, an inquiline that had been described from Mt. Edith Cavell, could not be rediscovered. However, the inquiline ant, L. wilsoni was found, which as yet had been known only from eastern USA, and Formicoxenus quebecensis, a guest ant of Myrmica alaskensis Wheeler, 1917, also as yet only known from eastern North America (Quebec).

Discussion

Leptothorax athabasca is described here because among the many samples that Buschinger and collaborators have collected in several sites in Canada and in the USA, and among the described species as well as a few unnamed ones depicted on internet-sites (e.g., antweb.org year unknown, Longino 2005) none could be found with the very characteristic acute petiolar profile and the flat mesosoma of L. athabasca.

Our differential diagnosis is provisional, but at present it is impossible to thoroughly compare L. athabasca with all the more or less well or insufficiently described species and especially with the many undescribed forms, nor the subspecies of the " Leptothorax muscorum complex" listed by Creighton (1950) that all had been synonymized with Leptothorax muscorum by Brown (1955). The genus evidently comprises many more species in North America than in Europe where only three Leptothorax species are known that lack erect hairs in femora and antennal scapes, i.e., the "true" L. muscorum, Leptothorax gredleri, and Leptothorax scamni. All three species apparently do not occur in North America. Heinze (1989b) gives an impression of the variety of North American species with respect to biochemical markers: apart from L. crassipilis Wheeler, 1917 and L. retractus he had studied eight forms named " Leptothorax muscorum A" through " Leptothorax muscorum H" plus " L. muscorum AxB?", finally suggesting that the complex in addition to these eight or nine forms should consist of at least three or four further taxa. Loiselle& al. (1990), studying karyotypes of a number of taxa, also have failed to disentangle the group; the forms described until then as species could be confirmed (including the European species L. muscorum and L. gredleri), but a number of other, morphologically different samples and populations still remained under the name " L. muscorum ".

The flat and slender mesosoma in the female castes probably represents an adaptation to life in narrow rock crevices. This character corresponds well with the fact that practically all other Leptothorax species both in Europe and in North America preferably are nesting in dead wood or bark (L. acervorum in some places also in rock crevices), where nest entrances and galleries usually are tubular.

The new species seems to have an extremely limited range as far as is known, a phenomenon, however, that appears to be not unusual among the Formicoxenini. Intensive search in Alberta did not reveal any other site where L. athabasca would occur.

We point out that quite a number of ant species in fact are extremely rare, or better to say, inhabit very restricted ranges, or the ranges may be subdivided into a few widely scattered small plots. For example, Leptothorax pocahontas had been collected in August 1977, a few kilometers upriver of the L. athabasca site, in Maligne Canyon, in one very small site. One of the authors (A.B.) had found it there again in 1979 (Buschinger 1979) and 1993, J. Heinze had collected it in the very same place in 1988, and also S. Lindgren took specimens from this locality in July 2002 and 2003 (Lindgren year unknown). In all instances L. pocahontas was only found in the type locality and nowhere else in spite of intensive search in similar places along the Athabasca River. Leptothorax faberi had been found only once, in 1979, in Jasper NP, and on all later occasions (see data for L. pocahontas) we were unable to rediscover this species. One host colony with one queen of L. faberi had been found, not far from the L. athabasca site. Temnothorax fragosus is a clearly separate species. A number of complete colonies had been found near Jasper in 1979, in quite an ordinary habitat at the foot of a moraine, beneath pebbles. No colonies were found in any other place, neither in the year of its detection (1979), nor in subsequent years.

Hence, at least four species of formicoxenine ants have been found as yet exclusively in a comparatively small part of Athabasca River valley near Jasper. Moreover, other species were detected in this same area that had been recorded from only a few but very distant localities. One of these is Leptothorax wilsoni, a social parasite living in nests of " Leptothorax sp. B" (cf. Heinze & Buschinger1987, 1988). Leptothorax wilsoni had been described from eastern North America: USA: New Hampshire, Cheshire Co., and from Canada: Quebec, Parque National des Grandes Jardins, and New Brunswick, Westmorland Co. (Heinze 1989a). It has been discovered several times, in July 1993, in Jasper NP, and in USA: Montana, south of Glacier National Park (Buschinger & Schumann 1994). Most recently one specimen had been detected in Alaska (P.S.Ward, in litt.).

What about the reasons for rarity in Leptothorax athabasca and the other exceptional species found in the Athabasca River valley? As far as we could recognize, there were numerous places along Athabasca River that were looking quite exactly like the type locality ofL. athabasca. It is mere speculation to assume that slight differences in exposition, in the flooding regime, and in competition with syntopic Leptothorax species may be responsible for the very local occurrence ofthis species.

Perhaps, at the end of the Ice Age, a comparatively rapid colonization of the area may have occurred, along the valleys that commonly extend in North-South direction. We may speculate that the populations in the comparatively short time have not yet finally settled; that certain species are still increasing, others diminishing due to competition. Leptothorax athabasca thus may be secondarily restricted to one or a few sites where they escape competition due to nesting in rock crevices where they can survive flooding. It may be suspected that larger populations of L. athabasca can be found farther north in the North American Rocky Mountains.

Anyhow, as is true for most ant genera, much more field research will be necessary to establish the ranges of the North American species of Leptothorax.