Behavioral Observations on African Army Ants of the Genus Aenictus (Hymenoptera: Formicidae)
1976; Wiley; Volume: 8; Issue: 1 Linguagem: Inglês
10.2307/2387819
ISSN1744-7429
Autores Tópico(s)Animal and Plant Science Education
ResumoAfrican army ants of the genus Aenictus are column raiders and are generally hypogaeic. The species thus far observed are specialized predators of other ants, especially of the immature stages. Prey individuals are small, equal or subequal to the predator workers, and are not usually dissected by the foragers for transport to the nest. The foraging behavior of African and Asian Aenictus is compared. Emigration behavior of African Aenictus closely resembles that of Asian species, although the frequency and regularity of emigration and the intra-colony stimuli for nomadic behavior in African species and in hypogaeic species in general are unknown. ANTS OF THE OLD WORLD GENUS Aenictus belong to the subfamily Dorylinae and are among the smallest of the true army ants. The genus is IndoAustralian and African in distribution, and while it is well-developed with 34 species in tropical Asia, New Guinea, and Queensland, it is represented only by approximately 15 species in Africa (Wilson 1964). The Indo-Australian forms were taxonomically revised by Wilson (1964), but the African species are still arrayed among 60 nominate forms and are in need of thorough revision. The Asian species are also better known behaviorly, for Schneirla and Reyes (1966, 1969) produced detailed studies of epigaeic species, and numerous other observations on the Asian forms were recorded by Wroughton (1892), Crawley and Jacobson (1924), Wheeler and Chapman (1925), Wheeler (1930), Chapman (1964), and Wilson (1964). Observations on the behavior of sub-Saharan species are fragmentary and are found only in Brauns (1901), Arnold (1915), Weber (1943), Sudd (1959), and in a few of the original species descriptions (e.g. Weber 1942). Biologically, the African species are virtually unknown. The purpose of this paper is to present a series of new observations on the behavior of African species, and to compare these species with their congeners in Asia and with African army ants in the genus Dorylus. Behavioral data were gathered from a total of 31 colonies in Ghana, Ivory Coast, and Gabon in 1971 and 1974. Eight species are represented among these colonies, but most could not be identified with certainty. Unidentified species are given letter designations in this paper, and each colony is identified with a field number. Voucher specimens with their field numbers are deposited in the Museum of Comparative Zoology, Harvard University. AENICTUS SPECIES STUDIED The 8 species studied include A. bidentatus Donisthorpe (GC-039, IVC-003, GAC-048, GAC-052, GAC-072), A. congolensis Santschi (GAC-076, GAC-087, GAC-091), A. weissi Santschi (GAC-034, GAC-038, GAC-056, GAC-065, GAC-067, GAC069, GAC-075, GAC-084), and species A (GC-027, GC-035, GC-037, GC-047, GC-049), B (GC-071), C (GAC-044, GAC-064, GAC-085), D (GAC-047, GAC-050, GAC-060, GAC-063), and E (GAC-054, GAC-078). It is my opinion that those forms given letter designations are distinct species, and that final determinations await an examination of type-specimens. I am currently revising the genus, and the observations recorded here are an outgrowth of that study. HABITATS: A. bidentatus was observed in moist, semi-deciduous forest at Tafo, Ghana (GC-039), and in rainforest at Parc National du Banco, Ivory Coast (IVC-003), and at the Laboratoire de Primatologie et d'Ecologie Equatoriale, Centre National de la Recherche Scientifique, Makokou, Gabon (GAC048, GAC-052, GAC-072). The distributional data suggest that A. bidentatus is primarily a forest species with an extensive distribution throughout forested West Africa. Species A was observed on four occasions in coastal scrub and grassland at Legon, Ghana, and once in moist, semi-deciduous forest at Jukwa, Ghana (GC-049). Species B was collected only once at Larabanga, in northern Ghana, in Guinea savanna-woodland (GC-071). A. congolensis, A. weissi, and species C, D, and E were all observed at Makokou, Gabon, in rainforest. With the exception of the Guinea savanna-woodland locality, all of the habitats in which the Aenictus species were studied are relatively moist, although subject to periodic dry seasons. BIOTROPICA 8(1): 59-65 1976 59 This content downloaded from 157.55.39.59 on Mon, 17 Oct 2016 04:53:55 UTC All use subject to http://about.jstor.org/terms RESULTS AND DISCUSSION Army ant behavior is characterized by group predation (i.e. group raiding and group retrieval of living prey) and by nomadism or periodic emigrations of colonies to new nesting sites (Wilson 1958). Thus the primary extra-nidal activities of Aenictus workers are foraging and colony emigration, and this investigation focuses on these two activities. Observations were limited to worker ants in columns traveling to or from their nests, discovered exposed on the surface of the soil, beneath the forest litter, or in the soil. Of the 31 colonies of Aenictus observed, as represented by such columns, 13 definitely were foraging (workers were carrying prey), 17 probably were foraging (but workers were not carrying prey), and one was emigrating. FORAGING BEHAVIOR: Foraging columns are seldom more than one-worker wide (fig. 1A), although any one column may vary in width from 1 to 4 workers. Soil-particle walls are not constructed along the column nor do workers assume defensive positions along the column; both behavior patterns are common in African army ants of the genus Dorylus, subgenus Anomma (Raignier and van Boven 1955). However, one column of foraging workers (species A, GC037) returning to the nest was bordered with groups of stationary, irregularly arranged workers, for a distance of 10 cm, before the column entered a tunnel in the soil. Also, some workers of A. bidentatus (GAC-048) were observed removing soil particles from a foraging trail, a pattern that is basic to the construction of earthen works by Anomma workers. Some foraging columns are weakly developed and consist of small groups of 3 or 4 workers running in tandem (fig. 1B) [not in the sense of some nondoryline ants where the following worker maintains close antennal contact with the leading worker (Moglish et al. 1974)]. Such groups are often widely separated from one another but follow precisely the same trail, indicating the presence of a chemical trail substance. Schneirla and Reyes (1966) demonstrated that Aenictus foragers do, in fact, progress along chemical trails. Traffic on any foraging trail may be unior bi-directional with the direction of flow probably depending on the age of the raid. For instance, observations of Dorylus (Anomma) workers reveal that early in a foraging raid flow is unidirectional away from the nest; later, as the workers gather prey, traffic flows in both directions and includes prey-laden workers returning to the nest; late in the raid, traffic becomes unidirectional toward the nest (unpublished data). Periodic, 5-minute counts of workers in a two-directional foraging column of A. bidentatus (GAC-048) produced the following results: at 0900 hours 124 workers moved toward the raiding end of the column and 50 returned toward the nest (some workers carried prey); at 0950 hours, 34 moved toward the raid and 40 returned; at 1135 hours, 108 moved toward the raid and 797 returned; at 1155 hours, 2 moved toward the raid and 15 returned. The last returning worker was seen at 1157 hours. These data suggest a temporal shift in directional flow in foraging columns similar to that observed for Anomma. Workers of colony GAC-048 moving toward the raiding end of the column often did so in small groups of 4 to 6 individuals running in tandem, while returning workers flowed more evenly in much larger groups (but still in single file). African Aenictus are column raiders (sensu Schneirla 1971). A foraging column meanders out from the nest' and terminates in a series of anastomosing branches. Terminal raiding groups of workers develop at the ends of these branches and engage in a general search for prey. Sudd (1959) described similar raiding behavior for a species of Aenictus in Nigeria. The approximate areas occupied by the terminal branches and raiding groups of 5 foraging columns were measured (although these areas are often difficult to define) and were as follows: 2.66 x 0.30 m (sp. C, GAC-044), 1.45 x 1.25 m (sp. D, GAC-050), 2.00 x 1.60 m (sp. E, GAC-054), 1.35 x 0.46 m (A. weissi, GAC-069), and 2.00 x 1.70 m (sp. C, GAC-085). Terminal raiding groups are not highly organized and often consist of diffusely distributed workers engaged in 'general search. Workers involved in general search appear to meander at random, exploring the surface of the soil as well as crevices and openings in the soil. Terminal foraging branches were observed for 17 colonies. In 13 cases these terminal groups were foraging beneath the forest litter and/or in the soil. Terminal groups of species A (GC-035, GC-037) and species E (GAC-054, GAC-078) moved, exposed, on the surface of the soil. Three additional columns (although not terminal groups) of species A, a species that appears to be more or less restricted in distribution to the coastal scrub and grassland region of Ghana, were discovered and all were epigaeic in behavior (including 1 emigrating column. Epigaeic behavior may be specific to species A. In general, African species of Aenictus are negatively phototaxic and hypogaeic in behavior, although Brauns (1901) and Arnold (1915) pointed out that African Aenictus avoid light to a lesser extent than do subterranean forms of Dorylus. Certainly there are no African species of Aenictus dis-
Referência(s)