Formica exsecta Nylander, 1846

Seifert, Bernhard, 2000, A taxonomic revision of the ant subgenus Coptoformica Mueller, 1923 (Hymenoptera, Formicidae), Zoosystema 22 (3), pp. 517-568 : 525-532

Formica exsecta Nylander, 1846

Formica exsecta Nylander, 1846

TYPE LOCALITY. — Helsinki, Finland.

TYPE MATERIAL. — Syntypes 1 male, 1 female, 3 workers ( FMNH) [investigated].

Formica exsectopressilabris Forel, 1874 . Synonym.

TYPE LOCALITY. — Samedan, Switzerland.

TYPE MATERIAL. — Lectotype 1 queen (designated by Agosti 1989); paralectotypes 2 queens ( MHN) [investigated].

Formica exsecta var. rubens Forel, 1874 . Synonym. TYPE LOCALITY. — Apples V. Fermaur, Switzerland. TYPE MATERIAL. — Syntypes 4 workers ( MHN) [investigated].

Formica exsecta var. etrusca Emery, 1909 . Synonym. TYPE LOCALITY. — Praccia, Italy.

TYPE MATERIAL. — Syntypes 4 workers ( MCSN), 1 worker ( NHM Basel) [investigated].

Formica dalcqi Bondroit, 1918 . Synonym. [Synonym; study of topotypical material and description of topotypical population by Agosti (1989)].

TYPE LOCALITY. — Mt Canigou, S France.

TYPE MATERIAL. — Syntypes 3 workers (probably in MNHN), not seen.

Formica exsecta var. sudetica Scholz, 1924 . Synonym.

TYPE LOCALITY. — Sudety, SE Poland.

TYPE MATERIAL. — Syntypes 1 worker ( ZMHU), not seen [type description of Agosti (1989)].

Formica kontuniemii Betrem, 1954 . Synonym.

TYPE LOCALITY. — Inari, N Finland.

TYPE MATERIAL. — Syntypes not traceable ( RNH) [investigation of topotypical material].

Formica nemoralis Dlussky, 1964 . New synonym.

TYPE LOCALITY. — Voronesh Reservat, S Russia.

TYPE MATERIAL. — Paratypes 9 workers ( ZMLU) [investigated].

GEOGRAPHIC ORIGIN OF THE MATERIAL STUDIED. — The numerically evaluated 220 workers, 71 queens, and 22 males came from Norway 5, Sweden 48, Finland 37, Denmark 2, the Netherlands 3, Germany 49, Poland 11, France 5, Switzerland 77, Austria 37, Spain 4, Italy 12, Slovenia 4, Bulgaria 6, Turkey 2, Russia 6, and NE China 5. Total number of specimens seen> 1700.

DESCRIPTION

Worker ( Figs 2 ; 10)

Maximum size larger than in other species (CL 1419 ± 82, 1200-1641; CW 1362 ± 83, 1131- 1574). Head shape of average Coptoformica type (CL/CW 1.042 ± 0.023, 0.979-1.099); however, long-headed and short-headed specimens may occur within the same nest. Rather long scape (SL/CL 1.008 ± 0.022, 0.931-1.063). Clypeus at least in anterior third, but normally also in median and posterior portions with standing setae (ClySet 3.54 ± 1.08, 2-5). Lateral semierect setae in the ocellar triangle usually present (OceSet 92%). Eye hairs at least in a fraction of the nest population strongly developed, often hook-shaped (EyeHL 27.0 ± 6.5, 0-45; Fig. 10). Pubescence in the occellar triangle frequently dilute, but enormous intraspecific and intranidal variation occurs (sqrtPDF 5.69 ± 0.90, 3.78- 9.30). Region of occipital corners with semierect to subdecumbent pubescence (however, in specimens of the etrusca population almost appressed). Craniad profile of forecoxae with semierect setae (nCOXA 8.86 ± 3.89, 0.5-23). Dorsal mesosoma and propodeum occasionally with few standing setae, lateral metapleuron and ventrolateral propodeum more frequently setaceous (nMET 1.86 ± 2.11, 0-9). Outer edge of the hind tibial flexor side conspicuously hairy (nHTFL 10.97 ± 2.82, 5.0-23.0), with two size classes of setae, and subdecumbent pubescence ( Fig. 2 ). Semierect setae on gaster tergites as a rule beginning on the 1st tergite (TERG 1.19 ± 0.46, 1-3); nest sample means of TERG always <2.4. Pubescence density on first gaster tergite with extreme intranidal and intraspecific variation (sqrtPDG 6.82 ± 1.19, 3.93-9.88).

Queen

Size definitely larger than in other species (CL 1636 ± 44, 1514-1741; CW 1721 ± 42, 1629- 1809; ML 2878 ± 116, 2613-3115). Head broad (CL/CW 0.950 ± 0.022, 0.900-1.008), scape long (SL/CL 0.956 ± 0.023, 0.893-1.004). Clypeus at least in anterior third, but normally also in median and posterior portions with standing setae. Lateral semierect setae in the ocellar triangle normally present. Eye hairs normally long and numerous, often hook-shaped (EyeHL 45.6 ± 7.6, 31-69); samples with less numerous eye hairs may occur. Pubescence in the occellar triangle relatively dense (sqrtPDF 4.28 ± 0.49, 3.34-5.75), less variable than in workers. Occipital corners of head normally with semierect smaller setae and pubescence, morphs with almost reduced and such with very developed occipital hairs may occur (OccHD 46.9 ± 22.5, 7-107). Brilliance of dorsal head surface variable, but relatively matt and weakly sculptured surfaces dominate (GLANZ 1.71 ± 0.36, 1.0-2.5). Craniad profile of forecoxae with semierect setae (nCOXA 12.95 ± 4.09, 3.5-23.0). Promesonotum normally with standing setae that clearly differ from semierect pubescence, in weakly haired specimens the differentiation between shorter semierect setae and longer semierect pubescence can be lost (MnHL 181.8 ± 40.0, 0-256). Outer edge of the hind tibial flexor side conspicuously hairy (nHTFL 12.81 ± 3.20, 8.0-22.0), with two size classes of setae and subdecumbent pubescence. Semierect setae on gaster tergites always beginning on the first tergite (TERG 1.00 ± 0.00). Pubescence density on first gaster tergite with extreme intraspecific variation (sqrtPDG 6.17 ± 1.16, 3.83-9.25).

TAXONOMIC COMMENTS AND

DIFFERENTIAL DIAGNOSIS

Interspecific setae and pubescence differences between Coptoformica species are normally correlated with differences in body measures and indices at least in the queens. Investigations of the enormous setal and pubescence variability of the exsecta samples seen during this study could not convincingly show such correlations and could not satisfactorily separate entities of possible taxonomic significance. The same local population may show a wide range of pubescence and pilosity variance and there is no clear indication that certain setae morphs could be associated with a certain distribution, habitat selection, or biology. As a consequence and in agreement with the view of Agosti (1989), Formica exsecta is considered here as polymorphic, relatively eurytopic species with a large range, similar to the situation in Formica truncorum , pratensis , and Scandinavian lugubris . This conception of F. exsecta contrasts the situation in the other Coptoformica species that are constant, monomorphic entities with a defined zoogeography and more specific habitat selection.

Three queen syntypes of Formica exsecta var. exsectopressilabris Forel, 1874 stored in MHN and labelled “ F. exsectopressilabris f Samedan ” are very close to the European population mean of exsecta in almost all characters except for subaverage scape length and more developed coxal setae. Their sample means are: CL 1620, CW 1714, ML 2870, CL/CW 0.945, SL/CL 0.922, EyeHL 40.3, GLANZ 1.87, OCCHD 66.0, MnHL 158.0, nCOXA 18.7, TERG 1.0, nHTFL 11.0, sqrtPDF 4.74, sqrtPDG 6.23. Their synonymy is not in question.

Four worker syntypes of Formica exsecta var. rubens Forel, 1874 , stored in MHN and labelled “ exsecta (M) variét? (Forel) avec esclav fusca V. Fermaur \ V. rubens Forel ” are rather large, have subaverage setae numbers, higher TERG, and a more developed reddish pigmentation. These characters, however, are within the range known for exsecta , the overall resemblance is too large, and a synonymy is most probable as the following sample means show: CL 1560, CW 1478, SL/CL 0.997, CL/CW 1.056, EyeHL 33.8, TERG 2.3, nCOXA 3.2, nHTFL 8.75, nMET 0.12, sqrtPDF 6.58, sqrtPDG 7.99, Clyset 2.5, OceSet 0.50.

Five worker syntypes (four in MCSN and one in NHM Basel) of Formica exsecta var. etrusca Emery, 1909 , all with locality label “ Praccia VII.90 S. ”, deviate from the exsecta standard in having a much lower gastral and frontal pubescence distance and an almost appressed pubescence at the occipital corners. Three samples with 12 workers from the terra typica in the Toscanese Alps (from Praccia, Abetone, and Giulia) have the following data (mean ± SD): CL 1491 ± 116, CW 1430 ± 112, SL/CL 1.002 ± 0.020, CL/CW 1.043 ± 0.009, EyeHL 33.8 ± 4.3, TERG 1.00 ± 0, nCOXA 5.88 ± 3.28, nHTFL 8.16 ± 2.00, nMET 0.25 ± 0.40, sqrtPDF 4.85 ± 0.71, sqrtPDG 4.74 ± 0.86, ClySet 2.50 ± 0.80, ClyPub 4.12 ± 0.96, OceSet 92%. The mean values of sqrtPDG, sqrt PDH, and EyeHL of the population from the Toscanese Alps significantly differ from those of the exsecta population from outside this region (p <0.001). Etrusca is assumed here to represent a deviating local population of exsecta . A more extensive population study and search for diagnostic genetic markers are needed to test if etrusca already represents a separate evolutionary line worth to carry a taxonomic name. The high phenotypic similarity between etrusca and mesasiatica Dlussky should be considered in such an investigation.

Type specimens of Formica dalcqi Bondroit, 1918 (type locality: Mt Canigou /E Pyrénées, 1 800 m) were not studied. However, three topotypical workers in NHM Basel, labelled “ Canigou 2200 m, Pyrénées or. Weill.” were available and had the following sample means: CL 1392, CW 1385, SL/CL 1.015, CL/CW 1.005, EyeHL 27.0, TERG 1.0, nCOXA 10.0, nHTFL 11.5, nMET 0.7, sqrtPDF 5.46, sqrtPDG 7.41, Clyset 4.0, ClyPub 3.0, OceSet 100%. All these data are within the usual range of variation known for exsecta . Agosti (1989) studied the population at the locus typicus and stated that all characters presented by Bondroit as diagnostic for dalcqi were variable in this population as they were in exsecta in general. As a consequence there is no suggestion for a heterospecifity of dalcqi .

The type worker of Formica exsecta var. sudetica Scholz, 1924 from the Sudety mountains / SW Poland was not seen. The pigmentation characters presented by Scholz (1924) as characteristic for this form are not known to have a diagnostic value in any known species and also habitat selection, nest construction, polycaly, and aggressive behaviour do not show deviations from the exsecta standard. Agosti (1989) examined the type specimen and stated that the sculpture characters presented by Scholz for sudetica (“the usually rather expressed longitudinal carina and lateral longitudinal carinulae of the propodeum”) may also occur in other populations. Hence, the synonymisation of sudetica is most probable.

The type specimens of Formica kontuniemii Betrem, 1954 from Inari / N Finnland are not traceable in the Museum of Leiden (Agosti 1989). As diagnostic characters of this very hairy form were presented by Betrem (1954) the presence of standing setae on gula, prosternum, pronotum, and propodeum. The sympatric occurrence of very hairy forms matching the description of kontuniemi and of less hairy forms was observed by the author throughout Fennoscandia (see also Agosti 1989) and a distinct cluster of very hairy ants could not be demonstrated by discriminant functions in the studied material. Thus, the synonymy of kontuniemi with exsecta seems most probable though the problem should be investigated with more sophisticated methods.

Nine paratype workers of Formica nemoralis Dlussky in the collection of ZMLSU, collected from the same colony as the holotype, and labelled “Voron.Sap. 29.VII.1962 No 221 Dlussky” are in any character within the range of variability known for nest sample means of exsecta and are almost identic with the type workers of rubens . Six measured paratype workers of nemoralis had the following mean values (in brackets the means of the rubens types): CL 1360 (1560), CL/CW 1.057 (1.057), SL/CL 0.977 (0.997), EyeHL 27.5 (33.8), TERG 2.17 (2.33), nCOXA 3.2 (3.2), nHTFL 7.5 (8.5), nMET 0.0 (0.1), sqrtPDF 6.53 (6.58), sqrtPDG 7.77 (7.99), ClySet 2.5 (2.5), ClyPub 3.1 (no data), Oceset 0.83 (0.50). The rather low TERG value in the nemoralis types is explained by an abnormal reduction of setae on first and second gaster tergite in one portion of the specimens, possibly caused by a mutation. Queen types of nemoralis were not available to the author, but decriptions of females from this colony (Dlussky & Pisarski 1971; Agosti 1989) give some diagnostic characters. The queens have a dark, shining body surface (according to Agosti within the range known for exsecta ) and their metric measures are equal to the means known for exsecta . As average of the data of Agosti, Pisarski, and Dlussky is calculated a thoracic length of 2900, an overall head length of 1920, a CW of 1690, and a SL of 1600. As a consequence, Formica nemoralis should represent a less hairy morph of exsecta and should be kept in synonymy.

BIOLOGY AND DISTRIBUTION

Geographic range

South to Central Spain , to the N Appennine and to the Balkans at 40°N. Found in high Anatolia and Caucasus ; apparently absent from the dryest Pontic and Caspian steppe zones. West to SW England and the Scottish Highlands. Northern range troughout Fennoscandia up to North Cape .

Distribution in the east in European Russia, across Siberia, Mongolia, NE China (Beishan National Park, 37°N, 102°E) and east to the lower Amur river. The northern distribution in the continental parts of Eurasia is limited by the -8 °C isotherm of soil in a depth of one meter (achieved at 67°N in

W Siberia at the Ob river and at 62°N in E Siberia at the Lena river) and the southern distribution coincides with the southern border of foreststeppe (Dlussky 1967). Vertical distribution: in Switzerland and Austria 300-2250 m, bimodal, with very low frequencies from 800-1200 m ; Bulgarian mountains 1100-2200 m .

Habitat selection

F. exsecta is more a generalist with a wider habitat selection than usual within Coptoformica species. Very different open or slightly shaded habitats, which must have medium-term stability at least, are inhabited. A high cover percentage of grasses within the field layer is typical but not essential. As for all species of the group, sites with rapid plant succession, with high nitrogen input, or sites within the inundation plains of rivers are not colonised. Formica exsecta and all other members of the group cannot increase nest temperatures by metabolic heat production independent of environmental temperatures as known for populous nests of some wood ant species. The dependence of Coptoformica upon direct insolation is thus increased and nests cannot be constructed in fully shaded woodland habitats. Habitat types used by exsecta are subalpine and boreomontane pastures, clearings and margins of woodland, sunny forests, semidry to xerothermous grasslands, heathland, and dryer spots of bogs and fens.

Status as threatened species

Red List Germany: three (threatened). In Central Europe it is probably the least endangered Coptoformica species though its populations have significantly declined since 1950. The decline is caused by afforestation of clearings and meadows, vanishing of coppice wood management, decline of sheep pasturing, intensifying of cattle pasturing, intensive use of mineral fertilisers and liquid manure, and high athmospheric nitrogen immission. Dewes (1993) recorded F. exsecta colonies in the Nature Park Saar-Hunsrück preferentially in sites of historic coppice wood management where oak bark for tanneries was harvested.

Colony foundation

Flight or ground dispersal of single queens is followed by socially parasitic colony foundation in nests of the subgenus Serviformica . Formica fusca (Dlussky, Collingwood, Pisarski, Agosti, Seifert) and Formica lemani (Collingwood, Agosti, Hellrigl, Seifert) are reported as host species. According to Pisarski (1982), socially parasitic colony foundation is only possible in queenless host nests. The same author stated that monogynous nests of exsecta never accept alien queens, with exception of very minute nests. To become polygynous, they must have lost the queen and can then adopt several queens simultaneously. Once shifted to polygyny, huge polycalic colony systems may develop by nest splitting.

Nest construction

Nest mounds are often more structured than in related species. Materials and type of nest construction depend upon insolation, ground water level, soil type, and composition of vegetation. The following type is frequent on mineralic soils: a calotte-like outer hull of the dome, which can often be lifted without breaking (Agosti 1989), is constructed with more dense, strongly adhesive materials (usually finely cut grass pieces). Rather voluminous upper brood chambers (during warm weather the sites for pupae) may be situated between this hull and inner mound material that consists of more lofty plant pieces. The lower mound core is a mixture of humified plant material and mineralic soil. Galleries may reach 150 cm down into the soil. In polygyous nests and during summer, numerous clearly separated chambers of 2 cm width are found below the level of soil surface down to 80 cm. These chambers usually contain a queen, eggs, small larvae, and few workers. In habitats deficient of grasses, other materials as coniferous needles, rabbit pellets, or small pebbles may be used. Nest diameters may reach 200 or even 300 cm (Agosti 1989; Dlussky 1967). In semi-shaded boreal forests, the author observed nest mounds of 150 cm height and 180 cm diameter, the base of which had been abandoned by the ants and covered by mosses ( Polytrichum ).

Development and microclimatic requirements Alates develop from eggs laid in early spring. Oviposition starts in Central European lowland habitats in late March but is considerably delayed with growing altitude and latitude. In Finnish monogynous colonies, less than 5% of the spring eggs develop to workers (Chapuisat et al. 1997), but this ratio may reach 50% in Swiss polygynous societies (Schneider pers. comm.). The main source of worker production is late spring or summer eggs. The optimum temperature for “brood development” is 28-30 °C (Grinfeld 1939, cited in Dlussky 1967). The supercooling point of winter-adapted Siberian workers is -20 °C; the long-term minimum at which 50% of workers survive is -8 °C (Berman, Shigulskaja & Leirich 1987; Leirich 1989). The conditions where soil temperatures in a depth of one metre do not decline below -8 °C are given in E Siberia up to 62°N, which coincides with the northern distributional border of exsecta .

Demography of nests and colonies

The ratio of monogynous nests vs polygynous/ polycalic colonies differs locally and geographically. Patchiness and general availability of suitable habitats and a payoff between the costs of single queen colony foundation and the costs of reproductive competition in polygynous nests are probable factors influencing this ratio. In S Finland monogyny/monodomy clearly predominates. Big polycalic colonies are known from the Alps, Central Europe, and European Russia. The mean longevity of queens in monogynous nests in S Finland was over 20 years (Pamilo 1991). According to Sundström et al. (1996), 39% of the queens in 57 monogynous S Finnish nests were multiply-mated (polyandrous). The sex ratio of produced alates is strongly dependent upon the number of queens in a colony and the number of mates per queen; in highly polycalic colonies in the Swiss Alps it was about 15: 1 (Schneider pers comm.), in Finnish polyandrous/ monogynous colonies it was 3.76: 1, but in Finnish monoandrous/monogynous nests it was 1: 2.2. Workers of the latter colonies increased their inclusive fitness by selectively killing male larvae before pupation and (most probably) feeding them to the now rapidly growing female larvae. Ecological and demographic factors (recource limitation!) are believed to interfere with genetic factors (optimisation of kinship value), i.e. males should survive in higher numbers in monoandrous/monogynous nests if there is plenty of recources (Chapuisat et al. 1997). Monogynous nests mainly produce the large male morph (macraner) but polycalic colonies mainly micraners (Pamilo & Rosengren 1984). The micraners can develop from worker-laid eggs, mature later, and show narrower activity peaks. Both micraners and macraners are normally haploid (Agosti 1989).

No complete population census has been perform- ed so far in exsecta . The “Aussendienst” population of four monogynous nests of 368 ± 100 cm 2 basal area was censused by Pisarski (1982) as 2750 ± 1000 workers, i.e. 7.5 workers/cm 2 basal nest area. That means a total population of 18.8 workers/cm 2 basal nest area if assuming 40% Aussendienst workers. The total population of exactly censused polygynous bruni nests was 39.3 workers/cm 2 basal nest area (Schneider pers. comm.). Both values seem realistic in view of the higher worker density and smaller worker size in polygynous nests. A very large polygynous exsecta nest of 150 cm diameter should then contain> 300 000 workers. Polycalic exsecta colonies may be huge and can dominate a site as known for polyctena . Dewes (1993) described a supercolony comprising 408 nests (the smallest not counted!) spreading over an area of 2 ha. If assuming only 25 000 workers for an average polygynous exsecta nest, the whole population of this colony should be> 10 000 000.

Swarming

Mature alates are found in the nest 28.2 July ± 20.2 d (10 June-4 Sept, n = 33). In contrast to R. Rosengren, M. A. Schneider did not observe micraners to fly higher and farther than macraners. In Swiss polycalic colonies about 30% of females fly, the others are inseminated at the nest mounds. Swarming is restricted to the the first half of the day (between 5.30 and 12.20 h) and starts as soon as the first direct sunlight hits the mound surface. Completely cloudy sky or strong air movement prevents the flight and beginning sunshine in the second half of the day can not release it (Schneider pers. comm., and my own observations).

Food sources

F. exsecta can use a wide range of food sources. Trophobiosis with epigaeic and subterranean Aphidina (and more rarely Cicadina and Coccina) is observed and obviously covers a major portion of the energy needs. Lachnidae are the main trophobionts in coniferous and deciduous forests. Zoophagy may be important and is then comparable to that of Formica polyctena with the difference of a smaller average size of prey items (Wesselinov & Horstmann 1968); all kinds of dead or living Arthropoda and Lumbricidae are consumed. Very populous supercolonies effectively displace different species of predatory arthropods from their territory.

Intraspecific behaviour

Monogynous colonies are highly aggressive against conspecific aliens. Polygynous/monodomous colonies show reduced aggressivity but only polycalic colonies do not establish territorial boundaries against other conspecific polycalic colonies (Pisarski 1982).

Figures

FIG. 2. — Hind tibiae in workers of Formica exsecta, suecica, fennica, bruni, forsslundi, and pressilabris. The view is not fully directed on the large tibial diameter; there is a slight tilt for better visualisation of setae on outer margin of flexor profile. Scale bar: 400 µm.