Jl. Mercier et al., LIMITED AGGRESSIVENESS AMONG AFRICAN ARBOREAL ANTS (HYMENOPTERA, FORMICIDAE) SHARING THE SAME TERRITORIES - THE RESULT OF A CO-EVOLUTIONARYPROCESS, Sociobiology, 32(1), 1998, pp. 139-150
In order to understand how the arboreal ant mosaic functions, we hypot
hesized that dominant, sub-dominant and non-dominant ant species shari
ng the same trees have evolved behaviors permitting them to limit the
cost of interactions during interspecific competition. We chose Oecoph
ylla longinoda, the best known ''dominant'' species and two other Form
icinae accepted on its territories: Polyrhachis laboriosa, a ''sub-dom
inant'' species: and P. weissi, a ''non-dominant'' species. We experim
ented in nature and on colonies reared on shrubs cultivated in planter
s. During dyadic encounters, foragers of the three species generally i
gnored each other. Aggressive interactions were rare and limited to in
timidating postures such as opening the mandibles and raising the gast
er. However, the need for protein leads arboreal ants sharing territor
ies to prey robbing. The expected dominance hierarchy did not hold tru
e at all levels as both O. longinoda and P. weissi rob prey from P. la
boriosa. Also, foragers of P. weissi were able to follow the recruitme
nt trails of P. laboriosa. As a result, predatory behavior is broken d
own into two tasks in the latter species: workers carving the prey tha
t tolerate the presence of interspecific intruders, and other workers
acting as guards. P. laboriosa guards limit the activity of the compet
itors thanks to allomones released during marking rounds, by drumming
the gaster on the substrate, and through dissuasive postures such as o
pening the mandibles, raising the gaster, bending the gaster and an ''
outstretched posture'' described here for the first time. We concluded
that a code containing chemical and behavioral information working at
the interspecific level enables ants sharing territories in the arbor
eal ant mosaic to defuse potentially volatile situations without the l
oss of individuals. These behaviors can be compared to ''true'' ritual
ized displays (i.e. intraspecific), and work here at the interspecific
level, implying a co-evolutionary process as selective pressure.