Afg. Bourke et J. Heinze, THE ECOLOGY OF COMMUNAL BREEDING - THE CASE OF MULTIPLE-QUEEN LEPTOTHORACINE ANTS, Philosophical transactions-Royal Society of London. Biological sciences, 345(1314), 1994, pp. 359-372
Multiple-queen societies of ants are key subjects in the study of comm
unal breeding. Societies of leptothoracine ants may be obligately mono
gynous (contain a single reproductive queen), functionally monogynous
(only one of several mated queens lays eggs), or facultatively polygyn
ous (some colonies contain more than one egg-laying, mated queen). Thi
s paper presents a framework for understanding these diverse social sy
stems as a function of leptothoracine ecology. The framework is derive
d from a synthesis of empirical information - in particular, a link be
tween the social system and the degree of habitat patchiness - with th
ree bodies of theory. These are ecological constraints theory, ESS (ev
olutionarily stable strategy) models of dispersal, and kin selection m
odels predicting the stable reproductive skew (allocation of reproduct
ion). In contrast to several previous ecological hypotheses, multiple-
queening in leptothoracines almost certainly results from high costs t
o single queens of dispersal and colony foundation (high ecological co
nstraints), which select for queens to seek adoption in their natal co
lony. Factors raising these costs include nest-site limitation, cold c
limate, and habitat patchiness. ESS models suggest that high dispersal
costs lead to a larger stable fraction of non-dispersers and hence to
higher relatedness between queens. Skew models predict that high ecol
ogical constraints and high relatedness promote high skew (one or a fe
w individuals dominate reproduction) and high within-colony aggression
. Therefore, (i) extensive habitats with moderate costs of solitary co
lony foundation should promote multiple-queening with high dispersal l
evels, moderate queen relatedness, low reproductive skew, and low quee
n aggression. By contrast, (ii) patchy habitats should induce multiple
-queening with less dispersal, higher queen relatedness, higher skew,
and higher aggression. In addition, (iii) habitats with small or widel
y spaced nest-sites, or with low costs of founding colonies alone, sho
uld lead to universal dispersal without multiple-queening. These assoc
iations of traits occur in facultatively polygynous, functionally mono
gynous and obligately monogynous leptothoracines respectively. Therefo
re, the framework in this paper explains a substantial amount of the s
ocial and ecological diversity of leptothoracine ants.