Cardiocondyla

Cardiocondyla are tiny myrmicine ants which live in colonies consisting of several dozen to a few hundred workers. Queen number varies by species. Nests are commonly in the soil, less so under rocks and in just a few species are known to be made in vegetation. Open arid habitats are favored by many Cardiocondyla. Several species are well known tramp species (Seifert, 2003).

Identification
Eguchi, Bui and Yamane (2011) - Worker monomorphic; head in full-face view subrectangular; frontal lobe small and narrow; frontal carina and antennal scrobe absent; median portion of clypeus prominently extended forward, and fused to the flattened lateral portions to form a shelf which hides basal part of mandibles in full-face view but is elevated away from the dorsal surface of mandibles in lateral view; posteromedian portion of clypeus relatively broadly inserted between frontal lobes; median clypeal seta present; mandible triangular, with 5 teeth which decrease in size from apex to base; palp formula 5,3; stipes of maxilla with a transverse crest at about midlength; antenna 12-segmented, with 3-segmented club; eye generally large and conspicuous; promesonotal dorsum in lateral view flattened to slightly convex;promesonotal suture absent dorsally; metanotal groove absent or distinctly impressed dorsally; propodeum nearly unarmed to strongly bispinose; propodeal lobe roundly extended posteriad; petiole pedunculate anteriorly and with distinct node; subpetiolar process present but small; postpetiole in lateral view dorsoventrally flattened, in dorsal view very broad, much broader than petiolar node; gastral shoulder indistinct or distinct; dorsa of head, mesosma, waist and gaster lacking standing hairs.

The worker of Cardiocondyla is similar to Monomorium and Temnothorax, but in the latter two genera the postpetiole is as broad as or only a little broader than the petiolar node, and the dorsa of head, mesosoma, waist and gaster bear at least a few standing hairs.

Seifert (2003) - [slightly modified from Bolton, 1982] Small to minute, monomorphic myrmicine ants. Palp formula 5,3 (16 species examined). Mandibles with 5 teeth which decrease in size from apical to basal. Clypeus with flattened and prominent projecting lateral portions, which are fused to the raised projecting median portion to form a shelf which projects forward over the mandibles. Median portion of clypeus posteriorly broadly inserted between narrow frontal lobes. Antennal scrobes absent. Eyes relatively large, situated in front of mid length of the head sides. Antennae with 11 - 12 segments, usually with a distinct 3-segmented club. Dorsal mesosoma without sutures; pronotal corners broadly rounded to bluntly angular. Propodeum unarmed to strongly bispinose. Metapleural lobes low and rounded. Petiole nodiform, with a moderate to long, usually slender, anterior peduncle. Postpetiole dorsoventrally flattened, in dorsal view always much broader than petiole. Sting large, knife blade-like in profile, without lamelliform appendages. Pilosity on dorsal body sparse to absent.

Cardiocondyla species groups

Biology
The biology of most Cardiocondyla species has not been studied. Seifert (2003) revised the holarctic species of Cardiocondyla and the following is based on his excellent treatment of the biology of the genus. References to other publications and more details about what is reported here can be found in Seifert (2003).

Cardiocondyla ants are minute to small. Natural nests have small populations and are difficult to discover because of the single and tiny entrance holes, which are usually unmarked by ejections of nesting substrate. When nesting in soil, trials to excavate complete nest populations may have a very frustrating outcome and need a very special skill. As a consequence, Cardiocondyla ants are neglected or overlooked by many field entomologists and are underrepresented in scientific collections. A recent world-wide catalogue recognized 49 taxa as valid species (Bolton 1995). This figure is undoubtedly far from conceiving the real species-richness as the following examples suggest: 371 samples from the Palaearctic, which is much less rich in Cardiocondyla species than the Palaeotropics, contained 31 species of which 14 are described here as new. Within only 67 samples from the tropical rain forests of Indonesia, Malaysia, and Papua New Guinea were 18 species with 8 of them new (Seifert in prep.). Including the Afrotropical, Oriental, and Australasian regions clearly more than 100 species of Cardiocondyla should occur on the globe.

Sociobiologists have paid much attention to Cardiocondyla during the last two decades because of some biological traits, which are rare among ants and provide good models to test several kinship theories. Cardiocondyla species have unusually long-lived ergatoid males performing a constant spermiogenesis throughout their whole imaginal life. These males usually stay within the mother nest, mate intranidally, and try to monopolize all matings by killing other ergatoid males, preferentially when these still are in the pupal stage. As a consequence, such heavily-armed ergatoids do occur within a nest in singularity or in lower numbers (Stuart & al. 1987, Kinomura & Yamauchi 1987, Heinze & Holldobler 1993, Heinze & al. 1993).

The very small space needed for nest construction, the expressed polygyny in several species, a sufficient survival rate after shortage of water, and in particular the fact that, in some species, a dozen of detached workers with brood can establish a fully reproductive new colony containing all castes explains the higher number of cosmopolitan tramp species in Cardiocondyla such as Cardiocondyla mauritanica, Cardiocondyla obscurior, Cardiocondyla wroughtonii, Cardiocondyla emeryi, and Cardiocondyla minutior, which all seem to be abundant around the globe. These species probably have reached many areas of their actual range by passive transport via human trade routes. Others, as the Pacific island-hopping Cardiocondyla nuda, seem to have a more restricted range.

Wing reduction or inability to fly is apparently abundant in sexuals of Cardiocondyla. Within the studied material, macropterous gynes were observed in 13 and brachypterous gynes in 11 species, with five species showing both forms. 14 species with ergatoid males are known. Only five species are reported to produce alate males but this is probably an under-recording because of their temporary occurrence. Isolated occurrence on small habitat patches within large desert systems, the tendency to dominate these habitat patches and to reduce the risk of flight dispersal could have selected for female brachyptery, male aptery, and intranidal mating. The resulting isolation and high inbreeding coefficients could have created an unknown number of rare, locally distributed species. Some of the species described here as new might belong to this category.

Habitats, nesting, and behavior
Knowledge on habitat selection and biology is poor or lacking in most of the known species. Many species of the world fauna are typically found in anthropogenically or naturally disturbed, open and xerothermous habitats along rivers, traffic lines, or wood margins and in sand dunes or other badlands. In particular the tramp species, but not only these, show a preference for habitats with a high degree of urbanization and the question arises which are the natural habitats. These should be, first of all, semi deserts and steppes as well as open habitats on immature soils at rivers, lakes, and sea shore and to a lesser extent forest margins or burned-down woodland patches. In contrast to this open-land group, the original habitats of many tropical species are primary rain forests.

In the majority of species, nests are constructed in the soil and less frequently under stones. Nests usually have a single entrance hole of only 1 - 1.5 mm diameter which is not marked by soil ejections. A very narrow vertical duct leads down to 2 - 15 cm depth before it changes abruptly to a horizontal direction or ends in a simple chamber of 15 - 20 mm diameter and 3 - 4 mm height. Simple nests without a complex vertical structure are found if moist soil horizons are not very deep and if the local climate does not show extreme temperature amplitudes. The depth of the uppermost chamber depends upon climatic conditions and the cohesiveness of soil particles (Creighton & Snelling 1974, Mei 1984; J. Heinze, pers. comm.; own observations). In the desert species Cardiocondyla ulianini, when ground water is very deep, the nest structure can be much more complex with the vertical duct crossing as much as 40 - 50 of such chambers one after the other down to a depth of 1 50 cm. This elaborate vertical structuring ensures direct access to ground water, it enables a free choice of narrow temperature optima during the extreme diurnal temperature changes in the desert, and it provides a protected hibernation during the cold Central Asian winter (Martkovsky & Yakushkin 1974).

Nesting in plant structures above soil surface is obviously rare in Holarctic Cardiocondyla. This behaviour is typical for Cardiocondyla obscurior and Cardiocondyla wroughtonii which occur in open areas, in grassland, at forest margins, in urban areas, or plantations.

Mature natural nests of monogynous and polygynous species usually contain less than 500 workers as found in C. wroughtonii, C. obscurior, Cardiocondyla mauritanica, C. ulianini, Cardiocondyla koshewnikovi, and Cardiocondyla elegans. The real frequency of monogyny within the genus is unknown; it has been observed so far in C. elegans, Cardiocondyla batesii, and C. ulianini. The cosmopolitan tramp species C. wroughtonii, C. obscurior, C. mauritanica, Cardiocondyla emeryi, and C. minutior are polygynous and found new colonies preferentially by nest splitting.

Development at room temperature from oviposition to the eclosion of adults lasted approximately 56 days in ergatoid males of C. mauritanica (Heinze & al. 1993) and 55 days in workers of “C. emeryi” (Creighton & Snelling 1974), with the egg, larval, and pupal stages lasting for 12, 27, and 16 days. Most remarkably, the callow stage in workers was as short as 2 days. Development of gynes of C. ulianini in the Central Asian deserts seems to need 100 days at least. After oviposition in May, gynes are reported to eclose by the end of August and in September, to hibernate in the nest and to leave it in next spring (Marikovsky & Jakushkin 1974). A similar situation seems to exist in C. batesii (J. Heinze, pers. comm.) in which the young gynes are most probably mated intranidally in late summer, hibernate in the nest in an alate condition, and use the moist spring situation for dispersal to have better founding success. Reproductive gynes of C. mauritanica produced 2 - 3 eggs per day in well-fed laboratory colonies and their ovaries contained 3+3 ovarioles (Heinze & al. 1993). No observations or suggestions of worker reproduction are known in Cardiocondyla. Workers of aggressive superior species such as Pheidole dentata, Solenopsis geminata, and Linepithema humile were observed to shrink back when encountering foragers of Cardiocondyla “emeryi” and C. mauritanica (Creighton & Sneilling 1974) which suggests the emission of effective repellents.

Cardiocondyla ants are omnivorous. Zoophagy (zoo necrophagy and killing of small weakly sclerotised arthropods), granivory, and nectarivory are reported (Creighton & Snelling 1974, Marikovsky & Yakushkin 1974, Dlussky 1981). Recruitment to food sources or nest sites can be performed by tandem running as observed in few species (Creighton & Snelling 1974). The latter authors, the present author, and Marikovsky & Jakushkin (1974) noted a well-developed, almost linear light compass orientation of Cardiocondyla mauritanica and Cardiocondyla ulianini foragers on open surfaces.

The ergatoid male syndrome - all ergatoids show a fighter-phenotype
Males were not subject of a thorough morphometric and structural investigation during this revision. The statements presented below are based upon short visual inspection of material seen during this revision and upon critical revision of literature (Baroni Urbani 1973, Marikovsky & Yakushukin 1974, Kugler 1983, Terayama 1999). In one case, species identification is doubtful: Marikovsky & Yakushkin (1974) described a “male-like wingless gyne of Cardiocondyla ulianini” which in fact is an ergatoid male of another related species. The wider petiole and wider spine base distance compared to the ergatoid male of C. ulianini and the locality suggest an allocation to the Central Asian population of C. sahlbergi.

With these objections, alate (A) and/or ergatoid (E) males are known so far in the following Cardiocondyla species: C. batesii (E), C. elegans (E), C. emeryi (A+E), Cardiocondyla kagutsuchi (A+E), C. koshewnikovi (E), C. mauritanica (E), C. minutior (A+E), Cardiocondyla nigra (A+E), Cardiocondyla paranuda (E), papuana (E), cf. sahlbergi (E), Cardiocondyla stambuloffii (E), C. ulianini (E), C. obscurior (A+E), C. wroughtonii (A+E) , and an undescribed Indonesian species (E). This suggests ergatoid males to be present in any species while alate males seem to be lost in 60 % of the species.

The morphology of ergatoid males in Cardiocondyla shows a number of common traits in the manner of a syndrome which indicate similar life-strategies:

(1) All ergatoid males investigated have mandibles very effective for mutilating or seizing male competitors. One evolutionary way is to develop worker-like mandibles with increased sclerotisation and increased size of apical dentition. Worker-like mandibles have been observed in C. elegans, C. ulianini, C cf. sahlbergi, C. mauritanica, C. paranuda, C. kagutsuchi, C. batesii, C. nigra, C. stambuloffii, C. koshewnikovi, C. minutior, and C. emeryi. The other way, the development of long, toothless and saber-shaped mandibles, is so far known in C. wroughtonii, C. obscurior, C. papuana (Reiskind 1965), and an undescribed Indonesian species.

(2) The ergatoid males considerably increase anterior mesosomal width by the development of lateral promesonotal corners and carinae in order to protect this sensitive region against biting attacks. An increase of waist segment width compared to the worker situation is frequent, but not found in all species.

(3) All ergatoid males reduce black pigmentation (leading to a light-yellowish-brown overall colouration), decrease eye size, and reduce the ocelli partially or completely. This syndrome is an indication for entirely intranidal life-cycles.

Hence, morphology suggests the ergatoid males of all species to have a similar mating and competition behaviour as it was directly observed in C. obscurior, C. mauritanica, C. kagutsuchi, and C. minutior. Pathology can also give evidence for ergatoid male fighting: within a sample on ergatoid males of C. nigra from Tunisia (ex coll. G. Mayr, NHM Wien) was one male with two legs partially cut-off. Completed peroxidase reactions in the wound areas indicate that this male survived the amputations.

The fusion of funiculus segments 2 - 8 to a single flattened and elongated segment, reducing the total number of antennal segments from normally 12 - 13 to 6 - 9, is known in 5 species (C. elegans 6 - 9, C. ulianini 6 - 7, C. cf. sahlbergi 6, C. batesii 6 - 7, C. nigra 7 - 8) and is probably without adaptive significance. In other species, the total number of segments in ergatoid males is reduced by only 0 - 2 compared to the known or putative number in alate males.

Gyne polymorphism - an adaptive trait emerging in desert environments
Probably gyne polymorphism is not rare in Cardiocondyla, but undetected in many species because of the small available sample size. It is expressed by strong variation in mesosoma dimensions and weak differences in postocular distance (a result of larger eye size of the flying macrosomatic gynes), whereas other characters are equal. Thus, gyne polymorphism in Cardiocondyla deviates from gyne polymorphism in Leptothorax, Tetramorium, Messor, or Myrmica in which measurements of all body parts differ between micro- and macrogynes. Hence, the use of the terms macro- and microgynes is problematic in Cardiocondyla; instead the terms macro somatic and microsomatic gynes are used here.

Gyne polymorphism was observed in Cardiocondyla ulianini (see also Marikovsky & Yakushkin 1974), Cardiocondyla batesii, Cardiocondyla bicoronata, Cardiocondyla nigra, Cardiocondyla elegans, and Cardiocondyla sahlbergi. All these species are inhabitants of Palaearctic deserts, semi deserts or dry steppes and three of them are known to be monogynous. Only macrosomatic gynes have been collected in the Palaearctic desert species Cardiocondyla fajumensis (which may be a sampling artefact, since probably no nest samples but only queens in dispersal have been collected). Most remarkably, no gyne polymorphism has been observed so far in the cosmopolitan tramp species Cardiocondyla mauritanica, Cardiocondyla minutior, Cardiocondyla emeryi, and Cardiocondyla obscurior.

A distinct bimorphism of macrosomatic and microsomatic forms can be demonstrated for the species C. nigra, C. bicoronata, C. elegans, C. ulianini, and C. sahlbergi by calculating a mesosoma size index IMes = (ML + MW)/(2 CS). 41 microsomatic gynes of these species showed an IMes of 1.005 ± 0.016 [0.959, 1.055] and 28 macrosomatic gynes one of 1.110 ± 0.016 [1.074, 1.158]. The mesosoma size index of C. batesii reveals no distinct bimorphism but shows a strong continuous polymorphism. Microsomatic gynes of the bimorphic species are brachypterous with forewing length amounting only 201 - 267 % of CS (as found in 8 winged specimens) while macrosomatic gynes are macropterous with a forewing length of 318 - 384 % CS (as found in 10 winged specimens). Dissections were not performed to save the rare material but it is obvious that microsomatic gynes have reduced flight muscles and are most certainly not capable of active flight. In C. batesii, in which gynes are continuously polymorphic instead of bimorphic, a highly significiant correlation of mesosoma size with forewing length exists with FWL/CS = 7.349 IMes - 4.402 (r = 0.843, n = 14, P < 0.001).

As emphasised above, gyne polymorphism in Cardiocondyla seems to be typical for continental desert or semi desert species and seems to be absent in cosmopolitan tramp species. Within the ant genera Monomorium, Aphaenogaster, Proformica, and Cataglyphis, wing reduction in gynes is most frequently (if not exclusively) observed in arid environments (Tinaut & Heinze 1992). Collectors of Cardiocondyla in desert or semidesert environments reported the nest sites to be concentrated in rare spots with less extreme living conditions, in particular where ground water horizons are less deep below soil surface (Marikovsky & Jakushkin 1974, Dlussky 1981; own observations). Rare spots with suitable living conditions within large unsuitable areas mean a high mortality risk during or after flight dispersal and strategies to achieve a maximum density on these rare spots should be favoured. Brachypterous gynes minimize the risk of dispersal and try to found colonies near to their mother colony in an area that most probably will offer sufficient resources. Which of the four possible ways of colony foundation they follow, remains to be studied: (1) single gynes found independently, (2) a single gyne leaves its mother nest accompanied by few workers, (3) several gynes found in pleometrosis, or (4) nest-fission leading to polygynous-polycalic societies. Data available for C. elegans, C. batesii, and C. ulianini suggest monogyny and make option (4) less probable.

Overcrowding of rare spots once should increase the cost/benefit ratio of brachypterous gynes and long-range dispersal of macropterous gynes will become advantageous. As a consequence, a mixed strategy offering both brachypterous and macropterous gynes should have the highest long-term adaptive value. The putative rarity of alate males in desert ants means, that both macropterous as well as micropterous gynes usually should be mated intranidally by ergatoid males; the former should use their wings mainly to disperse to distant rare spots and not for mating flight.

The absence of gyne polymorphism in the tramp species C. mauritanica, C. minutior, C. emeryi, and C. obscurior is remarkable. By morphology, all their investigated gynes should be able to fly. Direct observations in C. obscurior and C. mauritanica confirm this view. In a field colony of C. mauritanica in California Creighton & Snelling (1974) observed that single females appeared at nest entrances and took flight; in one observation 8 gynes flew off within 78 minutes in a July morning. The apparent absence of alate males in this species and the presence of ergatoid males in the particular nest strongly suggest these gynes to be inseminated and to disperse for nest foundation. Heinze & al. (1993), however, have doubted single-queen colony foundation in a C. mauritanica population from the Canaries (Isla la Palma); during laboratory observations, they did not note flight activity of the macropterous gynes. Stuart & al. (1987) confirm intranidal insemination by alate males in C. obscurior. Kinomura & Yamauchi (1987) observed young gynes of C. obscurior to copulate intranidally, mainly with ergatoid but also with alate males. A large fraction of these intranidally inseminated queens did not shed their wings and stayed within the nest for one month at least. During two flights taking place at natural light conditions in late June 1986, about 110 gynes flew off; 50 of them were already inseminated and 60 not. Males flew off later than gynes. Only the virgin gynes were mated by alate males on the floor of the cage.

An explanation for the absence of gyne polymorphism in these highly polygynous tramp species, that should be biased to reduce physical flight ability, is most difficult. The simplest explanation seems to be that gynes developed an ethological polymorphism (flyers and non-flyers) but no selection operated to correlate this with a morphological bimorphism. One might also speculate that the native habitats of tramps were not desert habitats and that a selection for gyne brachyptery did not occur.

Castes
Ergatoid males and queens are present in some species.

Nomenclature

 *  CARDIOCONDYLA [Myrmicinae: Formicoxenini]
 * Cardiocondyla Emery, 1869b: 20. Type-species: Cardiocondyla elegans, by monotypy.
 * Cardiocondyla senior synonym of Emeryia: Forel, 1892h: 461; Forel, 1892i: 313.
 * Cardiocondyla senior synonym of Xenometra: Baroni Urbani, 1973: 200; Marikovsky & Yakushin, 1974: 60.
 * Cardiocondyla senior synonym of Dyclona, Loncyda, Prosopidris: Smith, D.R. 1979: 1375; Bolton, 1982: 309.
 * DYCLONA [junior synonym of Cardiocondyla]
 * Dyclona Santschi, 1930b: 70 (footnote) [as subgenus of Cardiocondyla]. Type-species: Monomorium cristatum, by original designation.
 * Dyclona junior synonym of Cardiocondyla: Bolton, 1982: 309.
 * EMERYIA [junior synonym of Cardiocondyla]
 * Emeryia Forel, 1890b: cx. Type-species: Emeryia wroughtonii, by monotypy.
 * Emeryia junior synonym of Cardiocondyla: Forel, 1892h: 461; Forel, 1892i: 313.
 * LONCYDA [junior synonym of Cardiocondyla]
 * Loncyda Santschi, 1930b: 70 [as subgenus of Cardiocondyla]. Type-species: Cardiocondyla (Loncyda) monardi, by monotypy.
 * Loncyda junior synonym of Cardiocondyla: Bolton, 1982: 309.
 * PROSOPIDRIS [junior synonym of Cardiocondyla]
 * Prosopidris Wheeler, W.M. 1935b: 40 [as subgenus of Cardiocondyla]. Type-species: Cardiocondyla (Prosopidris) sima, by original designation.
 * Prosopidris raised to genus: Reiskind, 1965: 80.
 * Prosopidris junior synonym of Cardiocondyla: Bolton, 1982: 309.
 * XENOMETRA [junior synonym of Cardiocondyla]
 * Xenometra Emery, 1917a: 96. Type-species: Xenometra monilicornis (junior synonym of Cardiocondyla emeryi), by monotypy.
 * Xenometra junior synonym of Cardiocondyla: Baroni Urbani, 1973: 199; Marikovsky & Yakushin, 1974: 60.

Eguchi, Bui and Yamane (2011) - The Afrotropical species were revised by Bolton (1982), and the elegans-, bulgarica-, batesii-, nuda-, shuckardi-, stambuloffii-, wroughtonii-, emeryi- and minutior-groups were revised by Seifert (2003). Workers of Vietnamese species have the following features.