In the New World tropics, Eciton burchellii and related army ants have complex relationships with the suite of animals (notably, the community of specialist birds) that obtain much or most of their food by capturing arthropods and small vertebrates attempting to escape from ant swarms. Ant-following birds and parasitoid flies regularly attend only the few army ants in the Ecitoninae tribe that develop swarm raids when foraging (other species have less conspicuous foraging behavior and either lack attendants or attract them only sporadically). Colonies of the predominant swarm raider, E. burchellii, can reach enormous sizes and may scour more than a thousand square meters of forest floor each day. Wrege et al. (2005) used exclusion experiments to investigate the impact of ant-following birds on the ants and found that ant-following birds are parasites on E. burchellii, significantly reducing the ants' success rate in capturing prey. An average nomadic colony in central Panama consumes ~22 g of leaf-litter arthropods each day, plus ~24 g of social insect brood, in raids that cover 700 square meters of forest floor (Franks 1982, cited in Wrege et al. 2005). Based on these numbers, Wrege et al. estimated that ant-following birds remove prey biomass (which would otherwise be available to the foraging ants) amounting to around 15% of a migrating colony's daily food requirement. (Wrege et al. 2005 and references therein)
Although large assemblages of army ant-following birds in the Neotropics are typically composed of no more than 20 to 30 individuals (smaller than the larger assemblages often associated with African Dorylus driver ants; Chaves-Campos 2005 and references therein), in the Caribbean foothills of Costa Rica Chaves-Campos (2005) observed a large swarm of E. burchellii attended by at least 20 Ocellated Antbirds (Phaenostictus mcleannani), 10 Bicolored Antbirds (Gymnopithys leucaspis), 10 Spotted Antbirds (Hylophylax naevioides), and 10 Plain-brown Woodcreepers (Dendrocincla fuliginosa), along with several Northern Barred Woodcreepers (Dendrocolaptes sanctithomae), Rufous Motmots (Baryphthengus martii), and White-fronted Nunbirds (Monasa morphoeus). In addition to the overall numbers of birds, the simultaneous presence of 10 or more obligate ant-following birds of the same species at the same swarm was also extraordinary. (Chaves-Campos 2005 and references therein)
Berghoff et al. (2009) studied the mite fauna living on several thousand Eciton burchellii workers from 20 colonies in Panama. Only 5% of the workers had mites (although this percentage is low, extrapolating this result to an average E. burchellii colony of about 500,000 workers, which may or may not not be a valid exercise, implies an average of 20,000 mites associated with each colony). Of those infected, just a single worker had more than one species and the one relatively common parasitic mite (Rettenmeyerius carli) was limited to only two individuals per ant. Only 14 workers had more than one mite and of these just two workers had more than two mites (three and six). Mite diversity was high, with 31 taxa recorded (18 of these were identified to species). Accumulation curves suggest that the authors were able to sample most of the mite taxa present in the population. Most of the symbiotic mites found on E. burchellii workers are probably relatively harmless guests (the two likely exceptions being Planodiscus burchelli and R. carli).
All Eciton species live primarily above ground, which is unusual for army ants, and they all prey on other ants. However, some species are strict ant specialists, whereas others also take non-ant prey. The diet of E. burchellii is unique because although around half their diet consists of ants, the other half consists of large arthropods (katydids, spiders, etc.) that they capture and dismember. No other Eciton species is known to prey on arthropods that are not social insects. Worker morphology varies considerably within and among Eciton species. One important difference among species is the presence or absence of a distinct "submajor" caste. Powell and Franks (2005) argue that the task of transporting novel prey selected for this exaggerated transport caste (which is significantly more exaggerated in E. burchellii than in any other species known to have this caste). Submajors clearly play an important role in the colony, although they constitute only 3% of the workers in an E. burchellii colony. (Powell and Franks 2005 and references therein)
Unlike most other army ants, E. burchellii armies do not run in narrow columns but rather spread out into flat, fan-shaped masses with broad fronts (Holldöbler and Wilson 1994). The spectacular group raids of E. burchellii, which can contain up to 200,000 workers, always remain connected to the nest by a "principal trail" of forager traffic. Swarming E. burchellii exhibit a remarkable behavior, referred to as "plugging behavior", in which some workers use their bodies to plug "potholes" in the natural surfaces over which the principal trail travels, providing a partly living surface for their nestmates to use. Investigations by Powell and Franks (2007) found that this behavior results in a significant improvement in the performance of prey-laden foragers, and that this improvement results in a net benefit for the colony. Although all Eciton species live in colonies with many tens of thousands of workers, raid and emigrate on the surface, and have similar cyclic patterns of nomadism, the swarm density of E. burchellii and the strength of the associated trail traffic are exceptional within the genus, a factor Powell and Franks suggest is key in their interesting discussion of the evolution of this unusual behavior. (Powell and Franks 2007 and references therein)
Lalor and Hughes (2011) studied the alarm response of Eciton burchellii and E. hamatum and discuss their similarities and differences in the context of the behavioral ecology of these two species.
The army ant Eciton burchellii (treated for years as Eciton burchelli, although originally described as Eciton burchellii) is a keystone predator in the leaf litter of many Neotropical forests (keystone species are species whose ecological importance is disproportionately great relative to their biomass). There are many species of army ants (not all related) in both the New World and Old World (mainly, though not only, in the tropics), but the most extensively studied species is E. burchellii, with much of the work on this species undertaken over many decades on Barro Colorado Island in Panama. (O'Donnell et al. 2007 and references therein).
Eciton burchellii live in colonies that can exceed half a million individuals. In the course of a day, a single colony may capture some 30,000 prey items. Their activity also flushes out larger arthropods, some of which are quickly devoured by ant-following birds (in some regions, there are bird species that depend on following army ants for nearly all their food). Eciton burchellii colonies typically exhibit a 35-day activity cycle. For 20 days, a colony resides in a fixed bivouac (a temporary nest, see below) from which raids emerge nearly every other day. During this phase, the colony's single queen may lay as many as 100,000 eggs. At the end of the 20-day period, these eggs hatch into larvae and with the increased demand for food the raiding becomes more frequent and intense. The colony now enters the nomadic phase, during which a new bivouac is formed at the end of each day's raiding. This behaviour lasts for about 15 days, until the larvae pupate, when a new statary (i.e., non-nomadic) phase begins. At the end of each statary phase, the pupae become new callow (i.e., newly emerged) workers. Such 35-day cycles continue throughout the year. Colonies may go extinct when the queen dies or the colony becomes too small. The largest colonies reproduce by rearing a sexual brood (males and a small number of queens) and then splitting. The daughter colonies are then headed either by the existing queen or by one of the new queens reared prior to the fission of the original colony. (Boswell et al. 1998 and references therein)
Eciton burchellii do not build nests like most other ants. Instead, they form temporary above-ground nests known as "bivouacs". Most of the cover in these partly sheltered locations is provided by the bodies of the workers themselves, which link their legs and bodies together with strong hooked claws at the tips of their feet. Together the workers--perhaps a half million, with a mass of about a kilogram-- form a solid cylidrical or ellipsoidal mass around a meter across. Toward the center of the mass are thousands of larvae and the single queen. (Holldöbler and Wilson 1994) For a broad and detailed overview of army ant biology (and ant biology in general) see the extraordinary volume by Holldöbler and Wilson, The Ants (1990), as well as early publications they cite by T.C. Schneirla and later ones by C.W. Rettenmeyer, both of whom were pioneers in understanding the biology of army ants.
Eciton burchelli has a very broad geographic distribution in the Neotropics, being found in lowland forests more or less continuously from southern Mexico to southern Peru and Brazil (Watkins 1976, cited in Brumfield et al. 2007).
Eciton burchellii is among the most conspicuous ants in wet tropical lowland forests from Mexico south to Brazil and Peru or Paraguay (Holldöbler and Wilson 1990,1994).
For studies of intra-colony relatedness and other purposes, Nigel et al. (2004) developed eight highly polymorphic microsatellite markers for Eciton burchellii.
Jaffé et al. (2007) found that genetic effects contribute significantly to the development of different worker castes in Eciton burchellii. However, as in other social Hymenoptera with a demonstrated genetic component for caste determination, the genetic role in E. burchellii is limited. Most males were able to sire daughters of all worker castes and the additive genetic variance accounted for only 15% of the total observed polymorphism, with the remaining 85% presumably being due to environmental and maternal effects. (Jaffé et al. 2007 and references therein)
Meisel (2006) studied the thermal ecology of Eciton burchellii. He found that these ants are extremely sensitive to temperatures above 43°C, that they live in an environment in which they routinely encounter potentially lethal temperatures, and that as a consequence they have developed a range of individual and colony-level responses to both limit such encounters and minimize their risks. These responses constrain the mobility and foraging effectiveness of E. burchellii colonies, particularly in fragmented or disturbed landscapes, and affect their behavior within continuous forest. These ants retreat from warm edges, steer around hot patches in continuous primary forest, and appear in many ways to be living near the upper limit of their temperature tolerance.
Dispersal in Eciton burchellii is strongly sex-biased since males are winged and queens are permanently wingless. Berghoff et al. (2008) used microsatellite and mitochondrial genetic markers to study the population structure of this species on Barro Colorado Island in Panama (microsatellite markers are normally nuclear DNA markers inherited from both parents whereas in most animals mitochondria--and hence mitochondrial DNA--are maternally inherited). Nuclear markers showed little differentiation between subpopulations, whereas mitochondrial differentiation was high, suggesting that although female dispersal was very low, male dispersal may be sufficient to effectively homogenize the studied populations.
Jaffé et al. (2009) used microsatellite genetic markers to study gene flow in a population of E. burchellii in Mexico. Because queens of this species are wingless, long-distance gene flow is virtually all via males (although during their monthly migrations, Eciton colonies are able to cover up to 1.6 km, so queens also contribute, to dispersal, although to a far lesser degree). Jaffé et al. found a high level of heterozygosity, weak population differentiation, and no evidence for inbreeding. They also found that through extreme multiple mating, the queens are able to sample the genes of males from up to ten different colonies, usually located within a radius of around 1 km. Thus, E. burchellii seems to maintain high gene flow via male dispersal.
Vidal-Riggs and Chaves-Campos (2008) describe and assess a method of estimating colony densities of E. burchellii based on raid crossings on trails.
Eciton burchellii colonies normally have a single queen and may have up to half a million polymorphic workers. On average, E. burchellii colonies studied on Barro Colorado Island (BCI) in Panama rear a synchronized cohort of around 50,000 new workers every 35 days. If colonies are sufficiently large at the start of the annual dry season, they may begin to produce a reproductive brood of about 4,000 males and a small number of new queens. When the males and new queens emerge from their pupal cases, the old parental colony undergoes a process of binary fission, splitting into two daughter colonies of approximately equal size and composition of individuals. Approximately one third of the 50 colonies on BCI reproduce each year. Thus colonies may take several years to grow from a daughter colony into a reproductive one. The mother queen may be retained by one of the daughter colonies or, alternatively, both daughter colonies may get a new queen. If a mother queen is retained in a colony, it is likely that she will be rejected at the next fission. Queens and males are very large and robust, but while males are winged and strong fliers, queens are never winged. Males are therefore the dispersing sex, and workers of foreign colonies allow them to enter, to mate with their queen. An important consequence of this unusual mating system is that queens may have the opportunity to mate several times each year of their lives. Males can only mate with the queen of the first colony they attempt to enter, because they immediately lose their wings, preventing further dispersal. (Franks 1985, cited in Denny et al. 2004; Denny et al. 2004 and references therein).
Although colonies of social insects typically have singly mated queens, there are many exceptions to this general rule and, as just noted, Eciton burchellii is a striking one. Denny et al. (2004) used microsatellite markers to investigate mating frequency in E. burchellii and estimated a mating frequency among the highest known for social insects. However, later work involving some of the same authors (Kronauer et al. 2006) showed that mating frequencies, although very high, were somewhat lower than initially believed (in line, in fact, with estimates for other army ant species), with the newer analyses finding mean observed and effective queen mating frequencies of 12.9 each (and, in contrast to earlier data, no evidence that queens are inseminated repeatedly throughout their lives, which would be exceptional for eusocial Hymenoptera; see also Kronauer and Boomsma 2007).
