Cephalotes varians

AntWiki - Where Ant Biologists Share Their Knowledge
Jump to: navigation, search
Cephalotes varians
Scientific classification
Kingdom: Animalia
Phylum: Arthropoda
Class: Insecta
Order: Hymenoptera
Family: Formicidae
Subfamily: Myrmicinae
Tribe: Attini
Genus: Cephalotes
Species: C. varians
Binomial name
Cephalotes varians
(Smith, F., 1876)

Cephalotes varians casent0103755 profile 1.jpg

Cephalotes varians casent0103755 dorsal 1.jpg

Specimen labels

Locally abundant in favorable habitats (e.g., red mangrove in southern Florida). Nests are typically made in hollowed out vegetation that is still standing, including in living plants with compromised cavities. Specialized major workers have bizarrely shaped heads that can serve as effective plugs for the entrances leading into their nests. Observations of this species by Wheeler led to the first note about the potential nest entrance blocking function of the majors of this genus. Creighton and Gregg (1954) "The first person to suggest a phragmotic function for the major of Cryptocerus (=Cephalotes) appears to have been W. M. Wheeler. In 1905 Wheeler stated that the nest entrance of several colonies of Cryptocerus varians, which he had found in hollow twigs in the Bahamas, was always exactly the size and shape of the cephalic disc of the major. Wheeler made no claim that the major blocks the nest entrance with its head, but the implication that it does so was clear and this seems to have been the basis for later views expressed by Wheeler in 1910 and by Emery in 1922 that the major of Cryptocerus is probably phragmotic."


A member of the pallens clade characterised by the following apomorphies: in the worker propodeal sides unarmed or with a narrow membranaceous margin medially denticulate, in the soldier and in the gyne by the combination of head disc with foveae contiguous each other, in the soldier by the outer face of the hind femora superficially carinate, by the head longer than broad, and, in the gyne, by the petiole and postpetiole without or with reduced lateral expansions. (de Andrade and Baroni Urbani 1999)

Identification Keys including this Taxon


Ranges from the tip of southern Florida, including the Florida Keys, to the Bahamas, Cuba and Jamaica. It appears to be a West Indian species that has just managed to colonize the most tropical portion of Florida, and it is the only cephalotine known from the eastern United States (Wilson 1976).

Distribution based on Regional Taxon Lists

Nearctic Region: United States.
Neotropical Region: Bahamas, Cuba (type locality), Greater Antilles.
Cephalotes varians Distribution.png

Check distribution from AntMaps.

Distribution based on specimens

Loading map...

The above specimen data are provided by AntWeb. Please see Cephalotes varians for further details


Wilson (1976) provides a wealth of details regarding the natural history of this ant and, from his findings, some of how its biology may be explained by the ecological setting it has evolved within:

Z. varians has been found nesting in dead branches of a variety of trees, including Bauhinia, the sea grape Coccoloba, and (especially in the Florida Keys) the red mangrove Rhizophora mangle. Colonies have also been found in stems of tall grasses and sedges, including sea oat (Uniola paniculata) and saw grass (Cladium jamaicensis). In southern Florida winged forms have been encountered in the nests during the summer months of June and July. The author's experience in the field in Cuba and Florida indicates that the species is exclusively nocturnal in nature. However, workers come to forage both during the day and night under conditions of constant illumination in the laboratory.

The preferred nest sites in dead twigs and branches are exceptionally dry. The Zacryptocerus colonies live in cavities that are consistently drier than the nest chambers of ground-dwelling ants, including many species that nest in desert soil. It seems probable that at least two peculiarities, the large size of the eggs and the partial consumption of infrabuccal pellets by the workers, are techniques to reduce water loss.

'Mature' colonies, that is, colonies large enough to produce winged queens, normally contain from one to several hundred workers and soldiers. Although I have dissected many nests without encountering any mother queen, large numbers of nests have been opened containing a single mother queen, and none has contained more than one such individual. Consequently Z. varians can be assumed to be normally monogynous. The queen also appears to be essential for the production of new workers and soldiers. While queenright colonies in the laboratory produced large broods of these castes, queenless colonies reared under identical conditions produced only males, or no brood at all. Although exact censuses of colonies freshly collected in the field have not been made, the ratio of adult minor workers to soldiers is approximately 10 : 1, a proportion that is maintained in the laboratory. Colonies are discrete entities. When workers from one colony are introduced into the nests of another in the laboratory, they are attacked and driven out.

The diet of free-living colonies of Zacryptocerus varians is not known, but it can be partly inferred from laboratory studies. In the laboratory the foraging minor workers are totally inept as predators. When workers encounter live insects outside the nest they generally avoid them. The very short mandibles and rigid body form make them physically less capable of pursuing prey even if the behaviour were well developed. Even small aphid nymphs just large enough to fit between the mandibles of the ants were handled ineptly in laboratory tests. The Zacryptocerus pushed at the aphids with their heads, seized and carried them briefly in their mandibles and then, invariably, broke contact and ran away. Not a single aphid was carried back into the nests. It is a remarkable fact that no solid food of any kind was ever seen to be carried into the nests, although the ants were presented with abundant quantities of diverse materials over a period of months. The ants are nevertheless strongly attracted to the tissues and haemolymph of freshly killed insects. This material they scrape, lick, and nibble away, swallowing all that they acquire and later regurgitating the liquid or semiliquid food to their nestmates. All of the insects offered in this manner were accepted: cockroaches, a nymphal chermid, two species of scarabaeid beetles, tachinid and muscid flies, a geometrid moth, and a variety of microlepidopteran moths. On four occasions workers were observed feeding on one of their own larvae. The workers prefer freshly killed insects to decomposed ones. Although fragments of insects in all stages of decomposition were available in the nest vicinity, and occasionally in abundance, foragers were seen to lick them on only two occasions, and then only for a few minutes. Honey and sugar water were avidly accepted by laboratory colonies. Also, natural honeydew was taken when leaves containing aphid colonies were placed near the nests.....In the laboratory, colonies flourish on a diet of fresh insects and honey. In the red mangrove forests of the Florida Keys ....Z. varians is one of the most abundant ants.

In short, Z. varians is a scavenger. The full extent of its diet can be worked out only by field studies, but it is clear that the species will accept a wide range of fresh insect materials. It appears to have surrendered all attempts at predation. This concession is in concert with its passive form of defensive behaviour. The nest sites...are quite dry and it is likely that the large size of the eggs and the consumption and traffic in infrabuccal pellets are adaptations for water conservation. These ants are general scavengers. (Their reliance on this foraging strategy)......, entailing the consumption of large quantities of liquid and semiliquid food, is a concomitant of the virtual failure of the workers to handle solid food collected outside the nest or to feed solid particles directly to the larvae. This specialization explains the high state of oral trophallaxis observed within the Z. varians colony.

The key adaptations of the species are judged to be their exclusively arboreal life and their apparent extreme reliance on scavenging. To make the sequences logically complete, the two adaptations are suggested to have been initiated or at least shaped in part by competition from other species of ants specialized for more conventional modes of existence.


Recruitment has been observed via the use of short lived pheromone trails. Adult transport seems to not be practiced. (Wilson 1976)

Behavioral Ethograms

A remarkable three(!) ethograms have been published for this species.

Wilson (1976) was the first to examine and analyze the behavioral repertoire of this species. An ethogram from his study is reproduced here Cephalotes varians ethogram 1. The abstract of the publication nicely summarizes this work: "An ethogram is presented of Zacryptocerus (=Cephaolotes) varians, a morphologically advanced member of the ‘turtle ants’ comprising the neotropical ant tribe Cephalotini. The species displays unusual and in one or two cases possibly even unique social behaviours, including the consumption and sharing of infrabuccal pellets, the apparent absence of adult transport, a primarily or exclusively mechanical form of colony defence, and a remarkable form of abdominal trophallaxis. All of these specializations seem to be associated with more primary adaptations by the species to arboreal nesting and scavenging. Among the additional results is the demonstration that the major worker is more specialized as a defensive caste than is the case in certain species of Camponotus and Pheidole."

Wilson also states "The major workers are very lethargic on almost all occasions except when defending the nest. Unlike the soldiers of Camponotus (Colobopsis) fraxinicola and Pheidole dentata (Wilson 1974 and unpublished), they are no more effective than minor workers in storing liquid food. This fact was ascertained by depriving a colony of honey for a period of 1 week, weighing a sample of ten minor workers and all of the six major workers, then feeding the colony to satiation for 2 days and weighing again. The average weight gain of the minor workers was 28·8 per cent, while that of the major workers was only 18·3 per cent, essentially the reverse of the situation in C. fraxinicola."

The Zacryptocerus varians major workers appear to function primarily in defence, and they deserve the title of soldiers. Yet they are not wholly defensive automata. As indicated in Table I, the major workers sometimes wash and manipulate larvae and pupae. When the nest is broken open and the brood spilled out in the open air, the major workers join the minor workers in retrieving the immature forms. However, they are less effective in this role and retire more quickly to the interior of the nest. The queens are even less active than the major workers. They are more sluggish than the queens of any ant species I have studied, standing immobile for long periods of time. Both the major workers and the queens appear to live exclusively by liquid food regurgitated to them by the minor workers.

The second ethogram Cephalotes varians ethogram 2 was published by Cole (1980) and examined the behavior of different ant species found living in southern Florida red mangrove (Rhizophora mangle) habitat (dense thickets of red mangrove trees). The study largely focused on C. varians and a Camponotus (Colobopsis) sp., (possibly Colobopsis impressa). The rational for this study was to explore the degree of similarity within a set of ecologically similar yet phylogenetically distinct species. The species detailed in this paper (and three others ants living in this habitat, see Cole's 1983 ethogram) nest in hollow twigs on the red mangrove along the coast in southern Florida (USA). The following is (in slightly modified form) from Cole 1980:

Some ants differ in the size or position of their nest on the mangrove tree, however all must contend with the special problems of arboreal life, twig nesting, and living in an environment of limited fresh water (red mangrove grows in shallow salt or brackish mater). Species of the subgenus Colobopsis and the myrmicine tribe Cephalotini have two of the most bizarre caste systems of any ant. Each has a discrete minor worker caste and major worker caste. The head of the majors in both species is greatly modified for nest defense. In Z. varians the frontal carinae are enormously prolonged backward over the eyes, forming a flange which can be used to plug the nest or to bulldoze intruders out of the nest. In Colobopsis the front of the head is abruptly truncated and circular when viewed from the front. The truncation involves the clypeus, frontal lobes and mandibles of the major ,workers. The modification is not as dramatic as that of the major worker of Z. varians and a single ant is not as effective in plugging the twig nest. Several ants acting in concert can form a barrier to intruders.

The two species presented in this study's ethogram were known to be similar in i) non-behavioral aspects of their biology (e.g. egg size and shield-headed soldiers), which are relevant to their sociobiology. ii) a number of behavioral traits, which are very unusual among ants in general, iii) their entire behavioral repertoire. Results showed the minor workers of both species exhibited a high lever of convergence but the major workers were less similar. C. varians majors are much more specialized than the Colobopsis sp. majors in: a.) being much larger than their respective minors b.) showing a relatively greater reduction in mandibles c.) exhibiting greater morphological modification of the head for use as a nest plug.

In sum Cephalotes majors are a highly specialized caste. Each major can serve as an effective nest plug. The Colobopsis majors, in contrast, work in concert to control entry into their nests and overall are involved in a wider range of activities within the colony. As predicted by Oster and Wilson (1978) it was found that with a higher level of specialization of the majors in C. varians their proportional abundance within a colony (1:10) was lower than what was seen with the Colobopsis majors (1:3).

Cole (1983) examined the behavioral repertoire of the entire mangrove ant community in this habitat. Cephalotes varians ethogram 3. The focus of this study was to examine how worker behavior might influence colony growth and, in turn, how this might affect colonization ability. Some conclusions from this study in regards to C. varians: "colonies produce specialized castes that function in colony defence and little else (Cole 1980): Major workers are not produced during the early stages of colony founding (Wilson 1971). Thus during the period that the colony is at its most vulnerable, there is no defensive caste. By the criterion of caste structure, Z. varians and Colobopsis should have reduced colonizing ability compared to Crematogaster ashmeadi, Xenomyrmex floridanus or Pseudomyrmex elongatus".....also...."The high queen-worker size ratio gives Z. varians an advantage in colonizing ability."


Wilson (1976):

The conventional view of polymorphic ants generally, and of cephalotines in particular, is that the major worker caste responds most vigorously to intrusions of the nest. Zacryptocerus majors have been considered to fill their main function simply by blocking the nest galleries and pushing intruders out of the nest entrances. But this is only a small part of the story. In fact, both minor and major workers of Z. varians proved to be very active. The minor workers have a lower response threshold, forming the 'early warning system' of the colony and disposing of minor intruders. Majors respond less readily, but once activated are individually more effective. The defence responses of the colony as a whole can be conveniently classified into two levels according to intensity:

(1) Low Intensity. The minor workers seize the intruder and drag or carry it out of the nest. The major workers may investigate briefly but do not otherwise participate. This is the form of response shown toward alien Zacryptocerus minor workers (and a worker of the tiny, inoffensive ant Strumigenys louisianae) placed in the nest cavities.

(2) High Intensity. Both castes participate. The minor workers seize the intruder, usually by its appendages, and try to carry or to drag it from the nest. They also join the major workers in pushing the invader with their heads and in pinning it to the nest wall by pressing against it with their backs with legs extended. If these techniques do not work, numerous individuals of both subcastes gather around and form a solid plug, trapping the enemy; then, by squeezing and shoving, they gradually force it to the entrance and out of the nest. The major workers are the more effective in this mode of defence. Using its great saucer-shaped head as though it were the blade of a bulldozer, one individual can block and push as effectively as two or more minor workers. Also, the major workers are more persistent once they have been activated, so that they become more heavily concentrated at the site of disturbance. The high intensity mode of defence was elicited in the laboratory by introducing workers of Tetramorium caespitum, which are approximately the same size as the Zacryptocerus minor workers, and by pushing dissecting needles into the midst of groups of workers.

Curiously, the Zacryptocerus showed no signs of using their stings, which are small but well developed. Nor was there any evidence of the employment of chemical defence, although the Dufour's and poison glands are typical in size for myrmicine ants and apparently normally developed. Intruders ejected from the nests invariably were able to walk away unharmed.

The 'turtle' defence strategy, which so markedly affects the anatomy of the ants as well as other aspects of their behaviour, is an efficient adaptation in two respects. First, the mandibles of a scavenger species, as opposed to a predaceous one, are likely to be blunt and short, rendering them less effective in defence. Those of Zacryptocerus varians have evolved to an extreme in this direction, so that when attacking an intruder the workers are able to do little more than seize a narrow appendage and hang on. As an alternative defence method the species has developed a squat body form with short, robust appendages. When workers are unable to escape more formidable enemies by running and hiding in the many crevices of the plants on which they forage, they can often avoid harm by crouching low and using the flanges and spines arrayed over the top of their bodies as a shield. The antennae are withdrawn into deep scrobes lining the sides of the head. The turtle-like anatomy appears to have evolved in conjunction with at least two other features. The soldier caste has a shield-shaped head which it employs much like the blade of a bulldozer. Lacking long, sharp mandibles, it cannot cut or puncture the bodies of enemies in the manner of the soldier caste of most ant species And because of its rigid body form, it cannot bring its sting or abdominal glands into easy play. The armoured body form also has the consequence of preventing self-grooming of the abdominal tip. As suggested earlier, the strange mode of abdominal trophallaxis practised by Z. varians may be a compensation for this incapacity. The short mandibles and rigid body of the minor worker might also explain why adult transport has never been observed in the species.


In varians the frontal carinae are enormously prolonged backward over the eyes, forming a flange which can be used to plug the nest or to bulldoze intruders out of the nest (Cole 1980).

Wheeler and Holldobler (1985) investigated how Cepholates workers might collect or accumulate debri on their head disks. Such debri can be used as camouflage for individuals that are using their head to block the nest entrance, much like Basicerotini and Stegomyrmicini ant species affix dirt to their bodies. They discovered Cephalotes varians do in fact have encrusted material on their head disks. This varied from slight to extremely built-up material. Unexpectedly these covering materials appeared to be produced by the ants, rather than debri that was somehow affixed to the head disk. The presence of many pores, which are presumably attached to glands, are the likely source of secreted fibrous and filmy substances that were detected.

Wheeler and Holldobler also found "the heads of soldiers were covered with large, blade-like setae arched over large shallow pits. The sparse distribution of the pits gives the head a pock-marked appearance. The most remarkable feature of clean specimens was the high density of pores 1-3 # in diameter. These are almost certainly glandular openings (Fig. 1). The density of pores was about pore/350 #2, and the total number of pores on the disc was estimated to be over 8000 One specimen, which appeared only slightly grimy to the eye, was in fact covered with a dense tangle of material (Figs. 2, 3a) which resembled a mass of fungal filaments (mycelia). Patches of material on the head of a second specimen consisted of an encrusting layer as well as some filaments (Fig. 3b). Examination of the mat material under a light microscope revealed, however, none of the branching or internal structure indicative of a fungal origin. We suggest that the filaments are material extruded through what are apparently glandular openings."


Dimorphic workers - minors and soldiers.






The following information is derived from Barry Bolton's New General Catalogue, a catalogue of the world's ants.

  • varians. Cryptocerus varians Smith, F. 1876d: 606, pl. 11, fig. 6 (w.) CUBA. Wheeler, W.M. 1905b: 102 (s.q.m.); Wheeler, G.C. & Wheeler, J. 1954b: 156 (l.). Combination in Cryptocerus (Cyathocephalus): Emery, 1915i: 192; in Cryptocerus (Cyathomyrmex): Wheeler, W.M. 1937b: 459; in Paracryptocerus (Cyathomyrmex): Smith, M.R. 1951a: 825; in Zacryptocerus: Smith, D.R. 1979: 1403; in Cephalotes: De Andrade & Baroni Urbani, 1999: 457. See also: Wilson, 1976a: 354.

Type Material

Unless otherwise noted the text for the remainder of this section is reported from the publication that includes the original description.

De Andrade and Baroni Urbani (1999) - Among the material of C. varians examined we observed the following variation: all the specimens from the Bahamas and those from Yateras (Cuba) have shorter peduncular spines. There seems to be no evident geographic pattern in the distribution of this character. The Bahamian specimens, in addition, may have the propodeal sides with a pair of teeth of variable size or simply marginate.



De Andrade and Baroni Urbani (1999) - Head slightly broader than long. Frons gently convex. Frontal carinae deeply incavate over the eyes and converging anteriorly. Vertexal angles with a pair of narrow, round or obtuse membranaceous expansions. Membranaceous expansions of the vertexal angles converging and not connected medially. Vertexal margin with a short "neck". Funicular club with two joints. Mandibles with superficial lateral carina.

Mesosoma. Scapular angles not visible in dorsal view. Anterior pronotal border convex. Pronotal sides with a pair of narrow, membranaceous expansions gently converging posteriorly. Promesonotal suture, in dorsal view, superficially impressed in some specimens only. Mesonotal sides with a pair of teeth truncate or pointed. Propodeum with poorly differentiate basal and declivous faces. Propodeal sides medially with a pair of triangular, membranaceous teeth of variable size; some specimens with the propodeal sides simply marginate or with a pair of narrow membranaceous expansions including the median teeth.

Petiole. Anterior face slightly concave. Petiolar sides with a pair of short, truncate or pointed teeth. Postpetiolar node convex dorsally; postpetiolar sides with a pair of spines, pointed or round and directed forwards.

Gaster. Oval and with a pair of anterolateral membranaceous expansions variably developed and not surpassing the stigma posteriorly.

Hind and mid femora gently angulate; their distal part with a dorsal longitudinal carina; the carina less impressed on the mid femora. Mid and hind basitarsi flat, with the proximal part broader than the distal one.

Sculpture. Head dorsum and mesosoma minutely reticulate and with superimposed superficial, dense foveae, shallower on the frontal carinae, sparser on the frons and denser on the mesosoma; some specimens with dense foveae on the whole head dorsum. Ventral part of the head reticulate and superimposed to foveae as on the frontal carinae on its anterior half. Pedicel minutely reticulate and with small, superficial foveae. Pleurae reticulate. First gastral tergite reticulate and with sparser, very superficial foveae, larger and more impressed on the anterior third. Gastral sternites and legs reticulate and with minute, superficial foveae, more impressed and denser on the outer face of the tibiae. Center of the first gastral sternite shining.

Pilosity. Each fovea with an appressed hair. Posterior half of the frontal carinae, vertexal angles, first gastral tergite and sternite, posterior border of the remaining gastral segments, and legs with rare clubbed hairs, slightly thinner on the first gastral sternite. Parts of the gaster and of the legs without foveae with short, appressed, very thin hairs. Gastral sternites with additional, rare, long, pointed hairs.

Colour. Ferruginous to brown with the frontal carinae and the membranaceous expansions of the body slightly lighter.

Measurements (in mm) and indices: TL 3.36-4.97; HL 0.82-1.16; HW 0.88-1.28; EL 0.19-0.26; PW 0.70-1.06; PeW 0.46-0.68; PpW 0.45-0.60; HBaL 0.27-0.36; HBaW 0.10-0.13; CI 107.3-118.5; PI 116.0-125.7; PPeI 138.2-216.7; PPpI 155.5-196.2; HBaI 34.3-38.7.

Soldier - Head disc suboval and covering completely the mandibles in dorsal view. Border of the disc strongly raised and with minutely crenulate margin, the crenulation more impressed on the posterior half. Floor of the disc medially with a small tumulus. Posterior part of the disc medially depressed. Vertexal angles completely separate from the disc, obtuse and with crenulate margin. Eyes slightly convex and hidden by the disc. Mandibles laterally carinate.

Mesosoma. Humeral angles with a pair of obtuse or slightly pointed teeth or with narrow membranaceous expansions. Pronotal sides gently converging posteriorly. Pronotal carina well marked, with crenulate margin and interrupted medially. Pronotal suture in dorsal view more impressed on the sides. Promesonotal suture deeply impressed. Mesonotal sides with a pair of broad, triangular teeth with almost round tip; some specimens with an additional pair of minute denticles. Propodeum with differentiate basal and declivous faces; sides of the basal face with two pairs of small, round swellings followed by a pair of stout teeth curved up and forwards in large specimens. Declivous face of the propodeum narrowing posteriorly.

Pedicel as in the worker but with narrower and shorter spines.

Hind femora dorsally gently angulate and with superficial carina in their distal part. Mid and hind basitarsi flat, with the proximal part broader than the distal one.

Gaster. Oval and with a pair of protruding lobes.

Sculpture. Head dorsum minutely punctate variably shining and with large, dense foveae, smaller on the tumulus and near the posterior border of the disc; foveae of the disc separated by thin interspaces. Sides of the head disc densely and irregularly foveolate, the foveae more superficial on the anterior half. Ventral part of the head punctate, variably shining; this sculpture covered by irregular, longitudinal rugosities on the posterior half and with small, dense, irregular foveae on the anterior half. Mesosoma minutely reticulate-punctate and with dense foveae, slightly irregular or oval on the pronotum, round on the mesonotum, small and superficial on the propodeum. Pedicel with similar sculpture as on the propodeum but with smaller foveae. Pleurae reticulate and with small, superficial foveae on the upper border. Gaster and legs with similar sculpture as in the worker but with slightly larger foveae.

Pilosity. Each fovea with an appressed or subdecumbent hair. Borders of the disc and sides of the head with many suberect, long, clavate hairs, sparser on the anterior border of the disc. Pronotum, first gastral segment, posterior border of the remaining gastral segments with rare, short clubbed hairs, denser on the first gastral sternite. Parts of the gaster and legs without foveae with short, appressed, thin hairs. Gastral sternites with additional rare, long, pointed hairs.

Colour. Dark ferruginous with slightly lighter gaster.

Measurements (in mm) and indices: TL 5.40-7.16; HL 1.44-1.96; HW 1.40-1.76; EL 0.30-0.33; PW 1.32-1.80; PeW 0.58-0.60; PpW 0.57-0.64; HBaL 0.32-0.39; HBaW 0.13-0.15; CI 84.8-97.2; PI 97.8-106.1; PPeI 227.6-305.1; PppI 231.6-300.0; HBaI 37.1-40.6.


De Andrade and Baroni Urbani (1999) - Differing from the soldier as listed below.

Sides of the disc with lower borders and with the margin more crenulate. Posterior part of the disc with deeper median depression.

Mesosoma. Humeral angles with a pair of triangular teeth with pointed or obtuse tip. Pronotal carina lower. Pronotum, mesonotum and scutellum flat in side view. Sides of the basal face of the propodeum anteriorly convex and posteriorly with a pair of teeth with round or pointed tip and slightly diverging backwards.

Wings figured in publication.

Petiole. Anterior face oblique and concave. Petiolar sides unarmed. Postpetiole as in the soldier but with the lateral spines broader, shorter and truncate.

Gaster. Similar to the one of the soldier but much longer.

Legs with less impressed carinae on the outer face of the femora.

Sculpture. Head dorsum with deeper, larger and denser foveae. Foveae on the mesonotum sparser. Pleurae reticulate. Propleurae with thin, faint, rugosities sometimes oriented transversally or longitudinally. Border of the lower metapleurae with thin, longitudinal rugosities. Upper part and center of the lower mesopleurae with small foveae, deeper and denser on the upper mesopleurae.

Pilosity. As in the soldier except for the clubbed hairs which are more abundant on the mesosoma and on the gaster.

Colour. Dark ferruginous.

Measurements (in mm) and indices: TL 7.52-8.36; HL 1.56-1.76; HW 1.44-1.58; EL 0.35-0.38; PW 1.32-1.56; PeW 0.51-0.56; PpW 0.57-0.64; HBaL 0.43-0.51; HBaW 0.15-0.17; CI 89.8-92.3; PI 101.3-109.1; PPeI 258.8-278.6; PPpI 231.6-243.7; HBaI 31.9-37.2.


De Andrade and Baroni Urbani (1999) - Head (eyes included) more than 1/3 broader than long. Vertexal angles converging to the vertexal margin. Vertex dorsally protruding, bearing salient ocelli. Compound eyes broadly convex, at mid head length. Frontal carinae very short and with a raised, convex border.

Frons flat and with a median, superficial sulcus. Clypeus convex posteriorly and incised anteriorly. Mandibles short and superficially carinate on the sides. Scapes thick, twice as long as the first funicular joint; remaining funicular joints thickening from the base to the apex.

Mesosoma. Pronotum in dorsal view with the sides diverging backwards. Mesonotum convex in side view; median Mayrian furrow variably impressed but always present. Scutellum convex, rounded posteriorly. Basal and declivous faces of the propodeum differentiate; basal face gently convex dorsally and slightly declivous; declivous face with converging sides.

Petiole with concave anterior face; posterior face gently convex and sloping posteriorly. Petiolar sides converging backwards, with a minute pair of denticles or unarmed. Postpetiole convex; its sides with a small, round median denticle.

Gaster as broad as the mesosoma.

Genitalia and subgenital plate shown in publication figures.

Wings. As in the gyne.

Sculpture. Head dorsum reticulate and sparsely, irregularly rugose, some specimens only; this structure superimposed to few, superficial, foveae more regular and denser behind the pair ocelli. Ventral part of the head reticulate, irregularly rugose-foveolate. Pronotum, mesonotum and scutellum with superficial, minute reticulation and with foveae as on the posterior part of the head dorsum, the foveae shallower and sparser on the mesonotum and scutellum. Basal face of the propodeum reticulate and with generally irregular, longitudinal rugosities. Declivous face of the propodeum, pedicel and propleurae reticulate. Mesopleurae reticulate; center of the lower and upper mesopleurae superficially foveolate; lower mesopleurae with few additional, longitudinal rugosities. Metapleurae reticulate and with irregular rugosities on the upper part. Gaster superficially reticulate and subopaque to shining. Legs punctate with moderately shining femora.

Pilosity. Body with long, thin, flexuous, pointed, golden hairs, dense on the head, on the mesosoma and on the pedicel, sparser on the gaster and on the femora. Gaster and legs with similar hairs but shorter, appressed on the tergites, decumbent on the sternites and on the femora. Tibiae and tarsi with short, pointed hairs.

Colour. Dark ferruginous-brown with slightly lighter coxae, pedicel and gaster. Legs yellowish with darker tarsomeres.

Measurements (in mm) and indices: TL 4.39-5.20; HL 0.70-0.75; HW 0.84-0.92; EL 0.30-0.41; PW 0.75-0.84; PeW 0.33-0.40; PpW 0.35-0.45; HBaL 0.44-0.47; HBaW 0.07-0.08; CI 120.0-122.7; PI 109.5-112.0; PPeI 210.0-227.3; PPpI 186.7-214.3; HBaI 14.9-18.2.

Type Material

De Andrade and Baroni Urbani (1999) - Type locality: Cuba. Type material 4 syntype workers on the same pin labelled "from Cuba, varians, Sm. Type (Cuba)" in The Natural History Museum, examined.