Birgit Ehmer and Bert Hölldobler.
Foraging behavior of Odontomachus bauri on Barro Colorado Island, Panama.
Psyche 102:215-224, 1995.

Full text (searchable PDF, 660K)
Durable link: http://psyche.entclub.org/102/102-215.html

The following unprocessed text is extracted from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

FORAGING BEHAVIOR OF ODONTOMACHUS BAURI ON BARRO COLORADO ISLAND. PANAMA
Theodor-Boveri-Institut, Lehrstuhl fur Verhaltensphysiologie und Soziobiologie der Universitat
(Zoologie II), Am Hubland, D-97074 Wurzburg, Germany Foraging behavior and partitioning of foraging areas of Odon- tomachus bauri were investigated on Barro Colorado Island in Panama. The activity of the ants did not show any daily pattern; foragers were active day and night. The type of prey captured by 0. bauri supports the idea that in higher Odontomachus and Anochetus species, the high speed of mandible closure serves more for generating power than capturing elusive prey. Polydomous nests may enable 0. bauri colonies to enlarge their foraging areas. The ponerine ant genus Odontomachus is distributed throughout the tropics and is especially abundant in the Neotropics. The most conspicuous feature of this genus is the trap-jaws, which the ants can open to an angle of 180' and close in less than 0.5 ms (Gro- nenberg, Tautz, and Holldobler, 1993; Gronenberg, 1995). Conver- gently evolved trap-jaws exist also in the dacetine ants (Myrmicinae) and in the formicine genus Myrmoteras, most species of which prey on springtails (Brown and Wilson, 1959; Moffett, 1986). Springtails have very fast escape responses (Christ- ian, 1979), so it is likely that the trap-jaw mechanism in dacetines and Myrmoteras evolved to facilitate the capture of these arthro- pods. For Odontomachus species, however, very few field studies exist and surprisingly little is known about the nesting and forag- ing biology of most species. Even for Odontomachus bauri, a com- mon and conspicuous species in the Neotropics, little is known about its foraging behavior and general biology. It is commonly regarded as a generalist predator (Brown, 1976; Levings and Manuscript received 5 May 1996.
215
Pswhf 102:215-224 (19951 hIt@p&yche enlclubwg/102/102-215 html



================================================================================

216 Psyche [vo~. 102
Franks, 1982). This assumption was examined during a study of foraging activity, choice of prey items, foraging range, and nesting habits on Barro Colorado Island in Panama. MATERIALS AND METHODS
Ants were observed on Barro Colorado Island, Panama, during three months from April to June 1995. Ant colonies were collected on the nearby mainland at Gigante. To measure activity a nest was observed 25 minutes per hour for 5 to 6 hours, eventually covering an observation period from 6-23 h. All ants entering or leaving the observed nest were counted. The first 5 minutes of each 25 minute observation period were discarded, as the ants were usually dis- turbed by the observer when approaching the nest. continuous observation of 1 nest for 3 hours showed, that an observation time of 20 minutes is sufficient to give a good representation of the activity of the colony. Foraging success was measured by noting whether or not the returning ants were carrying prey into the nest. The data were collected at the end of the dry season in May. Occa- sional rains had already occurred, but the soil was dry while recording. At the same time, air temperature above and below the leaf litter and humidity (above the leaf litter) was measured . For assessing the foraging area, individual ants were followed out from the nest as far as possible. To facilitate following the ants the leaf litter was removed from the nest entrance area and in 10-15 cm wide strips in regular intervals around the nest. A marker was set where contact with the ant was lost or the ant turned back or captured prey. The distance and direction traveled from the major nest entrance was measured. All observed ants were unmarked, as marking was found to disturb the treated ants considerably and paint markings were not very durable. Nests were studied mainly in one 200m2 area in the east of the island where 23 nest sites were located. Most nests consisted of simple chambers between leaves in accumulations of leaf litter, preferentially next to dead wood.
Activity of Odontomachus bauri
The foraging activity of Odontomachus bauri colonies remained approximately the same during day and night (Fig. 1). The counts



================================================================================

19951 Ehmer & Holldobler 217
varied from 2 to 44 ants entering or leaving the nest during a 20 minute observation period. There was no correlation between time of the day and activity of the ants (Fig. 1). Nor could a correlation be found between activity and air temperature, soil temperature, or humidity. These parameters did not vary much, air temperature ranging from 24OC to 30å¡C the soil temperature from 24.2OC to 27.5OC and humidity from 78% to 97% (Fig. 2). The leaf litter buffers temperature and also probably humidity so that the condi- tions within it are even more constant.
Foraging behavior
Odontomachus bauri is a leaf litter ant and forages individually most of the time underneath the leaf litter. It probes narrow cran- nies for arthropods, and also readily takes dead insects. It was repeatedly observed that individual workers appear to return to locations where they had previously found food. This is apparently due to learning by the foragers, because Oliveira and Holldobler (1989) failed to find evidence of sign posts for directing nestmates in Odontomachus. The foraging success of Odontomachus bauri, as determined by the proportion of workers returning with prey, is at least 28% (n=325). 41% of incoming workers were definitely unsuccessful as they entered the nest with opened mandibles, and the remaining 31% had either no prey or prey which was too small to be discerned.
The prey items taken from incoming workers are listed in Table 1. 0. bauri accepts a wide range of invertebrates, but forages espe- cially frequently on other ants and termites. The termites captured were mainly workers (22 of 25 caught were workers, 3 were sol- diers). The termites all belonged to the family Termitidae: most were Nasutitermes; the rest were Amitermes. The ant prey con- sisted mainly of sexuals. The largest prey items are most likely found dead and then carried to the nest. The successful capture of larger insects was not observed, but the detection and retrieval of dead items was repeatedly witnessed. The size distribution of the prey specimens shows that items between 3 mm and 4 mm are the most common size class (Fig. 3) and the most frequent prey items, like termites and wood lice, are soft bodied. Foraging area and polydomy
Mapping of the paths of foragers revealed that foraging areas of ants belonging to different colonies or to different nests belonging



================================================================================

218 Psyche [vo~. 102
5 6 7 8 9 10 11 12 13 14 IS 16 17 18 19 20 21 22 23 time [h]
Fig. 1. Foraging activity of Odontomachus bauri. Each data point represents an observation period of 20 minutes. Open triangles show the number of ants leaving the nest, filled circles the number of ants entering the nest. A linear function is fit- ted to the data points and represented by a dotted (returning foragers; y=5.55-0.018~; p=0.78) or dashed (leaving foragers; y=5.84-0.001~; p=0.98) line. Data were collected from the observation of six colonies (except 6h,7h n=4; 19h n=5). The bar indicates the lightidark period. Fig. 2. Variation of air (hatched) and soil temperature (solid columns) and humidity (open columns) during the activity measurements. The range from mini- mum to maximum values measured is shown.



================================================================================

Ehmer & Holldobler
Table 1. Identity of prey items brought to the nest by Odon- to~nachus bauri.
Class
Order
number
n=68
Annelida
Crustacea
Myriapoda
Arachnida
unidentified
Oligochaeta
Isopoda
Chilopoda
Aranea
Opiliones
Collembola
Blattodea
Isoptera
Ensifera
Hemiptera
Coleoptera
Hymenoptera
Lepidoptera
Diptera
O 2 4 6 8 10 12 14 16 18 20 22 24 26 28
size category [mm]
Fig. 3. Distribution of size classes of 71 prey items carried to the nest by Odon- tornachus bauri.




================================================================================

220 Psyche [VOI. 102
to the same colony did not overlap. Even nests which were very close together appeared to have distinct foraging areas (Fig. 4A, B). The nests were generally not situated in the center of their for- aging area and nests belonging to a single colony could be one to five meters apart. There was a preference for foragers to walk downslope after leaving the nest entrance, which helped to create asymmetric foraging areas. Neighboring nests appeared to have a strong influence on the shape of foraging areas, with foragers avoiding the foraging areas of neighboring colonies. Odontomachus bauri appears to have a polydomous nest struc- ture. This was initially suggested by the observation of a small number of workers travelling between nests during the measure- ments of colony foraging areas (Fig. 4B). Further evidence for polydomy was provided by observing that feeding a nest to satura- tion with termites led to a large number of ants (more than 101minute for several minutes) transporting termites to a neighbor- ing nest. In addition, ants from neighboring nests seldom fought, whereas encounters between ants from colonies 10 m apart were always aggressive.
No queens were found in 9 colonies excavated at Gigante, but all nests contained larvae and pupae. All of these nests did not extend into the soil, they consisted of gaps and cavities in leaf lit- ter and under wood. Nests were also commonly found in the accu- mulated leaf litter in the tops of Elaeis palms more than 2 m tall. Five of 9 colonies in palms tops contained single queens. It seems that colonies of 0. bauri are monogynous and sometimes polydo- mous in this habitat.
Activity of Odontomachus bauri
On Barro Colorado Island in Panama, the foraging of Odon- tomachus bauri is not limited by temperature and humidity. Their habit of foraging under leaf litter exposes them to little variation in weather conditions. In an arid Australian environment the activity of Odontomachus sp. was shown to depend on temperature by Briese and Macauley (1980), who found a shift from crepuscular activity in spring towards nocturnal activity in summer. Prior stud- ies have demonstrated visual capabilities in 0. bauri related to diurnal foraging: canopy orientation by Oliveira and Holldobler



================================================================================

19951 Ehmer & Holldobler 221
Fig. 4. Foraging areas at two different locations on Barro Colorado Island. Shaded areas indicate the locations of nests. Symbols depict the maximum distance of each worker from the nest. The number of workers followed from each nest is given (n). A The adjacent foraging areas belong to one colony, the third to a differ- ent colony. B All four nests belong to one colony. For a fifth nest no foraging area was established.
(1989) and use of vision in prey capture by Carlin and Gladstein (1989). However, the activity pattern found in the present study suggests that prey can successfully be located, captured, and returned to the nest in the dark.
Transient variations in activity could be caused by varying availability of prey. It is a common phenomenon in ants that the foraging activity rises after prey is brought into the nest. In this primitive kind of recruitment no apparent directional information is transmitted; only the rate at which foragers emerge is influenced. This type of recruitment was demonstrated in 0. bauri by Oliveira and Holldobler (1989), and is known from several additional ant species: Formica spp. (Carroll and Janzen, 1973), Paltothyreus tarsatus (Holldobler, l984), Ectatomma ruidum (Pratt, l989), Odontomachus chelifer (Fowler, 1980), and 0. ruginodis (Carlin and Gladstein, 1989).
Significance of the trap-jaws
In 0. bauri, the trap-jaws appear to be designed to stun or kill prey, usually small, soft-bodied insects. This is suggested by the fact that 0. bauri has short, broad mandibles with blunt instead of



================================================================================

Psyche
[Vol. 102
sharp teeth (Brown, 1976). The function of the mandibles is espe- cially evident when the ants were observed catching termites. The ants snapped at individual termites and either stunned or squashed the insects. Even though the prey was usually stunned by the first strike, often the ant administered repeated blows before carrying the prey to the nest. 0. bauri never stung termite prey, as was regu- larly observed in Ectatomma ruidum. Use of the high speed mandible strike may also be an effective mechanism for avoiding the chemical defenses by arthropods like the termite Nasutitermes. The very high speed of closure of the mandibles, and the fast release of the reflex via range finding trigger hairs (Gronenberg et al., 1993; Gronenberg, 1995), as well as the "strike and recoil" behavior, should be very well suited to handle chemically pro- tected prey (Brown, 1976). Indeed, Traniello (1981) found that 0. bauri was the most efficient ant species tested at successfully attacking Nasutitermes soldiers, which squirt a secretion that entangles and irritates an aggressor and can physically block sen- silla and spiracles (Prestwich, 1979). The presence of Nasutitermes workers among prey items retrieved by 0. bauri workers in this study shows that 0. bauri can exploit a food resource largely unavailable to other ants.
Polydomy
The polydomous colony structure revealed in this study may be an adaptation for maximizing the foraging area, while minimizing predation risk or desiccation while foraging. A polydomous nest allows a more effective use of an area (Holldobler and Lumsden, 1980). This also could explain why the foraging areas of different subnests hardly overlap. How partitioning between subnests is accomplished is not known, one possibility would be route fidelity by foragers as found, for example, in the ponerine Pachycondyla apicalis (Fresneau, 1985).
Another ecological factor supporting polydomy is the instability of nest sites in the leaf litter. Most nest sites did not extend into soil, but were confined to spaces in the leaf litter, so that they are not protected from accidental disruption. According to Holldobler and Wilson (1977, 1990) unstable nest sites favor polydomy and polygyny to ensure colony persistence even if subnests are lost. For 0. troglodytes, which inhabits nests very similar to those of 0.



================================================================================

19951 Ehrner & Holldobler 223
bauri, a polydomous and polygynous organization was shown for colonies exceeding 300 workers (Colombel, 1970). No evidence, however, exists that 0. bauri is polygynous, and according to Frumhoff and Ward (1992) the genus Odontomachus is predomi- nantly monogynous. Polydomy could also facilitate escape emigra- tion during army ant attacks (Droual and Toppoff, 1981), but during one observed Eciton raid the 0. bauri nest was not evacu- ated. So the primary function of polydomy may be to increase the foraging area of 0. bauri colonies.
The Smithsonian Tropical Research Institute generously pro- vided facilities and resources. Many thanks to Wulfila Gronenberg, Stefan Cover and anonymous referees for helpful comments on the manuscript. The Deutsche Forschungsgemeinschaft (Graduiertenkolleg "Arthopodenverhalten" and Gr 93313-2) sup- ported funding.
Briese, D.T., and B.J. Macauley. 1980. Temporal structure of an ant community in semi-arid Australia. Australian J. of Ecol. 5:121-134. Brown, W.L. Jr. 1976. Contributions toward a reclassification of the Formicidae, Part VI. Ponerinae, tribe Ponerini, subtribe Odontomachiti. Section A. Introduc- tion, subtribal characters, genus Odontomachus. Studia Ent. 19 (1-4):67-171. Brown, W.L., and E.O. Wilson. 1959. The evolution of the dacetine ants. The Quar- terly Rev. of Biology 34(4):278-294.
Carlin, N.F., and D.S. Gladstein. 1989. The "bouncer" defense of Odontomachus ruginodis and other odontomachine ants (Hymenoptera: Formicidae). Psyche 96(1-2):l-19.
Carroll, C.R., and D.H. Janzen. 1973. Ecology of foraging by ants. Ann. Rev. of Ecol. and Systematics 4:231-257.
Christian, E. 1979. Der Sprung der Collembolen. Zool. Jb. Physiol. 83:457-490. Colombel, P. 1970. Recherches sur la biologie et lY6thologie d'Odontomachus haematodes L.: Etude des populations dans leur milieu naturel. Ins. SOC. l7(3): 183-198.
Droual, R., and H. Topoff. 1981. The emigration behavior of two species of the genus Pheidole (Hymenoptera: Formicidae). Psyche 88(1-2):135-150. Fowler, H.G. 1980. Populations, prey capture and sharing, and foraging of the Paraguayan ponerine Odontomachus chelifer Latreille. J. of Nat. History 14:79-84.
Fresneau, D. 1985. Individual foraging and path fidelity in a ponerine ant. Insectes Sociaux 32(2): 109-1 16.




================================================================================

224 Psyche [VOI. 102
Frumhoff, P.C., and P.S. Ward. 1992. Individual-level selection, colony-level selec- tion, and the association between polygyny and worker monomorphism in ants. The Am. Nat. 139(3):559-590.
Gronenberg, W. 1995. The fast mandible strike in the trap-jaw ant Odontomachus I.Tempora1 properties and morphological characteristics. J. Comp. Physiol. A 176:391-398.
Gronenberg, W., J. Tautz, and B. Holldobler. 1993. Fast trap-jaws and giant neurons in the ant Odontomachus. Science 262:561-563. Holldobler, B. 1984. Communication during foraging and nest-relocation in the African stink ant Paltothyreus tarsatus Fabr.(Hymenoptera, Formicidae, Poneri- nae). Z. Tierpsychol. 65:40-52
Holldobler, B., and C.J. Lumsden. 1980. Territorial strategies in ants. Science 210:732-739.
Holldobler, B., and E.O. Wilson. 1977. The number of queens: an important trait in ant evolution. Naturwissenschaften 64:8-15. Holldobler, B., and E.O. Wilson. 1990. The Ants. The Belknap Press of Harvard University Press, Cambridge, Mass. 732 p. Levings, S.C., and N.R. Franks. 1982. Patterns of nest dispersion in a tropical ground ant community. Ecology 63(2):338-344. Moffett, M.W. 1986. Trap-jaw predation and other observations on two species of Myrmoteras (Hymenoptera: Formicidae). Ins. Soc. 33(l) : 85-99. Oliveira, P.S., and B. Holldobler. 1989. Orientation and communication in the Neotropical ant Odontomachus bauri Emery (Hymenoptera, Formicidae, Poneri- nae). Ethology 83: 154-1 66.
Pratt, S.C. 1989. Recruitment and other communication behavior in the ponerine ant Ectatomma ruidum. Ethology 8 1 :3 13-33 1. Prestwich, G.D. 1979. Chemical defense by termite soldiers. J. of Chem. Ecol. 5(3):459-480.
Traniello, J.F.A. 1981. Enemy deterrence in the recruitment strategy of a termite: soldier-organized foraging in Nasutitermes costalis. Proc. Natl. Acad. Sci. USA 78(3): 1976-1979.




================================================================================