Abandoning aggression but maintaining self-nonself discrimination as a first stage in ant supercolony formation.
ABSTRACT An ant supercolony is a very large entity with very many queens. Although normal colonies of small extent and few queens remain distinct, a supercolony is integrated harmoniously over a large area [1, 2]. The lack of aggression is advantageous: Aggression is costly, involving direct and indirect losses and recognition errors [3, 4]. Indeed, supercolonial ants are among the ecologically most successful organisms [5-7]. But how supercolonies arise remains mysterious [1, 2, 8]. Suggestions include that reduced within-colony relatedness or reduced self-nonself discrimination would foster supercolony formation [1, 2, 5, 7, 9-12]. However, one risks confusing correlation and causality in deducing the evolution from distinct colonies to supercolonies when observing established supercolonies. It might help to follow up observations of another lack of aggression, that between single-queened colonies in some ant species. We show that the single-queened Lasius austriacus lacks aggression between colonies and occasionally integrates workers across colonies but maintains high within-colony relatedness and self-nonself discrimination. Provided that the ecological framework permits, reduced aggression might prove adaptive for any ant colony irrespective of within-colony relatedness. Abandoning aggression while maintaining discrimination might be a first stage in supercolony formation. This adds to the emphasis of ecology as central to the evolution of cooperation in general .
- SourceAvailable from: uci.edu[show abstract] [hide abstract]
ABSTRACT: Despite the innumerable ecological problems and large economic costs associated with biological invasions, the proximate causes of invasion success are often poorly understood. Here, evidence is provided that reduced intraspecific aggression and the concomitant abandonment of territorial behavior unique to introduced populations of the Argentine ant contribute to the elevated population densities directly responsible for its widespread success as an invader. In the laboratory, nonaggressive pairs of colonies experienced lower mortality and greater foraging activity relative to aggressive pairs. These differences translated into higher rates of resource retrieval, greater brood production, and larger worker populations.Science 11/1998; 282(5390):949-52. · 31.20 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Kinship among group members has long been recognized as a main factor promoting the evolution of sociality and reproductive altruism, yet some ants have an extraordinary social organization, called unicoloniality, whereby individuals mix freely among physically separated nests. This type of social organization is not only a key attribute responsible for the ecological dominance of these ants, but also an evolutionary paradox because relatedness between nestmates is effectively zero. Recently, it has been proposed that, in the Argentine ant, unicoloniality is a derived trait that evolved after its introduction into new habitats. Here we test this basic assumption by conducting a detailed genetic analysis of four native and six introduced populations with five to 15 microsatellite loci and one mitochondrial gene. In contrast to the assumption that native populations consist of family-based colonies with related individuals who are aggressive toward members of other colonies, we found that native populations also form supercolonies, and are effectively unicolonial. Moreover, just as in introduced populations, the relatedness between nestmates is not distinguishable from zero in these native range supercolonies. Genetic differentiation between native supercolonies was very high for both nuclear and mitochondrial markers, indicating extremely limited gene flow between supercolonies. The only important difference between the native and introduced populations was that supercolonies were several orders of magnitude smaller in the native range (25-500 m). This size difference has important consequences for our understanding of the evolution and stability of unicolonial structures because the relatively small size of supercolonies in the native range implies that competition can occur between supercolonies, which can act as a break on the spread of selfish mutants by eliminating supercolonies harboring them.Evolution 05/2006; 60(4):782-91. · 4.86 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Ant colonies may have a single or several reproductive queens (monogyny and polygyny, respectively). In polygynous colonies, colony reproduction may occur by budding, forming multinest, polydomous colonies. In most cases, budding leads to strong genetic structuring within populations, and positive relatedness among nestmates. However, in a few cases, polydomous populations may be unicolonial, with no structuring and intra-nest relatedness approaching zero. We investigated the spatial organisation and genetic structure of a polygynous, polydomous population of Formica truncorum in Finland. F. truncorum shifts nest sites between hibernation and the reproductive season, which raises the following question: are colonies maintained as genetic entities throughout the seasons, or is the population unicolonial throughout the season? Using nest-specific marking and five microsatellite loci, we found a high degree of mixing between individuals of the population, and no evidence for a biologically significant genetic structuring. The nestmate relatedness was also indistinguishable from zero. Taken together, the results show that the population is unicolonial. In addition, we found that the population has undergone a recent bottleneck, suggesting that the entire population may have been founded by a very limited number of females. The precise causes for unicoloniality in this species remain open, but we discuss the potential influence of intra-specific competition, disintegration of recognition cues and the particular hibernation habits of this species.Behavioral Ecology and Sociobiology 01/2005; 57(4):339-349. · 2.75 Impact Factor
Current Biology 17, 1903–1907, November 6, 2007 ª2007 Elsevier Ltd All rights reservedDOI 10.1016/j.cub.2007.09.061
Abandoning Aggression but Maintaining
Self-Nonself Discrimination as a First
Stage in Ant Supercolony Formation
Florian M. Steiner,1,2,4,9Birgit C. Schlick-Steiner,1,2,4,9
Karl Moder,3Christian Stauffer,4Wolfgang Arthofer,4
Alfred Buschinger,5Xavier Espadaler,6
Erhard Christian,2Katrin Einfinger,7
Eberhard Lorbeer,7Christa Schafellner,4
Manfred Ayasse,8and Ross H. Crozier1,*
1School of Marine and Tropical Biology
James Cook University
2Institute of Zoology
Department of Integrative Biology and Biodiversity
3Institute of Mathematics and Applied Statistics
Department of Spatial-, Landscape-, and
Boku, University of Natural Resources and Applied Life
4Institute of Forest Entomology, Forest Pathology, and
Department of Forest and Soil Sciences
Boku, University of Natural Resources and Applied Life
6Centre de Recerca Ecolo `gica i Aplicacions Forestals
Universitat Auto `noma de Barcelona
7Institute of Organic Chemistry
University of Vienna
8Department of Experimental Ecology
University of Ulm
An ant supercolony is a very large entity with very
many queens. Although normal colonies of small ex-
tent and few queens remain distinct, a supercolony
is integrated harmoniously over a large area [1, 2].
The lack of aggression is advantageous: Aggression
is costly, involving direct and indirect losses and rec-
ognition errors [3, 4]. Indeed, supercolonial ants are
among the ecologically most successful organisms
[5–7]. But how supercolonies arise remains mysteri-
ous [1, 2, 8]. Suggestions include that reduced
within-colony relatedness or reduced self-nonself dis-
crimination would foster supercolony formation [1, 2,
5, 7, 9–12]. However, one risks confusing correlation
and causality in deducing the evolution from distinct
colonies to supercolonies when observing estab-
lished supercolonies. It might help to follow up obser-
vations of another lack of aggression, that between
single-queened colonies in some ant species. We
show that the single-queened Lasius austriacus lacks
aggression between colonies and occasionally inte-
grates workers across colonies but maintains high
within-colony relatedness and self-nonself discrimi-
nation. Provided that the ecological framework per-
mits, reduced aggression might prove adaptive for
any ant colony irrespective of within-colony related-
ness. Abandoning aggression while maintaining dis-
crimination might be a first stage in supercolony for-
mation. This adds to the emphasis of ecology as
central to the evolution of cooperation in general .
Results and Discussion
Whereas fierce fights occur between single-queened
ant colonies as a rule [1, 14, 15], we compiled scattered
literature observations of the lack of aggression be-
tween workers of different single-queened colonies for
21 ant species (see the Supplemental Data available on-
line). In ants, cooperation occurs within self, the colony,
but the boundary against nonself, other colonies, is well
guarded because the transmission of self genes is max-
imized when the dilution of self is prevented . Hence,
across the self-nonself boundary aggression normally
We dissected one of the hitherto underplayed cases,
the underground-living ant Lasius austriacus. From
queen morphology and colony excavations , colo-
nies were inferred to be single queened, and within-col-
ony relatedness was thus inferred to be high. We ex-
plored the self-nonself boundary for 16 colonies from
four populations in terms of genetic relatedness, signal
chemistry, self-nonself discrimination, and interactions
within self and across the self-nonself boundary. Dis-
tances between colonies were at least 10 m (neighbor-
ing colonies) and covered five orders of magnitude
Data on nine highly variable microsatellites (12.2 6 4.4
alleles) revealed high relatedness within colonies and
low relatedness between colonies, with significant dif-
ferences between self and nonself over all distances
(Figure 1). Genotype inspection showed that the major-
ity of workers within colonies (71 of 80; the remaining
nine, from seven colonies, are discussed below) had al-
ing the suggestion of single-queened colonies .
The self-nonself boundary was reflected in gas chro-
matograms of body surface hydrocarbons, which, in
9These authors contributed equally to this work.
ants, mediate the discrimination of self against nonself
. In an analysis of ranked similarities randomization,
the grouping of hydrocarbons was stronger according
to colony identity (global Rcol= 0.228, p = 0.001) than
to individual identity within colonies (global Rind =
0.080, p = 0.002) or to population identity (global Rpop
= 0.087, p = 0.004). After the comparison of similarity
index values, significant differences between self and
nonself emerged at all distances (Figure 1).
We alsofound that L.austriacus is capable of discrim-
inating self against nonself. Video analysis of one-on-
tennation bouts in self-self than in self-nonself pairings
(Figure 1). In a second experiment, bioassays indicated
the existence of chemically cued discrimination be-
cause workers lingered significantly more frequently
(p < 0.0001) on self than on nonself body surface hydro-
carbon extracts when offered the choice with the sol-
vent as control.
In one-on-one worker confrontations, we invariably
observed aggression between L. austriacus and four
other ant species (some active underground, others
aboveground), which represented all community domi-
nance levels. Combats were frequently begun by L. aus-
triacus (from 51 6 29% to 66 6 20% of fights). Encoun-
ters between two L. austriacus workers, however, never
yielded aggression, irrespective of the distance be-
tween the home colonies. To test for context depen-
dency, we repeated all of the latter tests in underground
conditions, and we again recorded no aggression.
Moreover, we observed integrative behavior across
the self-nonself boundary, but this behavior is normally
bonds for cooperation. In the laboratory, group forma-
tion occurred between any colony fragments put to-
gether (eight of eight tests, each with fragments of four
colonies), as did food exchange within and between col-
als). Integrative behavior across the self-nonself bound-
ary, if not restricted to laboratory conditions, should in
natureresult intheintegration ofworkers fromother col-
onies. In scrutinizing the field data, we compared micro-
satelliteallelesofthenine workers,mismatching thesin-
gle-queen pattern to the corresponding alleles in other
colonies. Two workers of one colony shared a mother
and father with workers from a colony ten meters
away. Considering the polymorphism of the nine micro-
p = 4.29 3 10217. Because we have sampled only one of
the neighboring colonies of each colony, it is likely that
the seven other aberrant workers also stemmed from
neighboring colonies. We infer that the integration of
nonself workers occurs under natural conditions.
Our data show that Lasius austriacus perceives self-
nonself cues but behaves nonaggressively and even in-
tegratively at the self-nonself boundary. Our sampling
scheme and the lack of aggression regardless of geo-
graphic distance excludes the fact that the nonaggres-
sive behavior is due to reduced aggression between
neighbors, compared to strangers. The driving forces
for peacefully bridging the boundary could be part of
the ecological framework and namely involve the nature
of the food source. Lasius austriacus is strictly subterra-
nean, tending mealybugs inside nests for honeydew
out the need for long-distance foraging , this might
reduce both the interaction between colonies and the
tension at the self-nonself boundary . The potential
benefits of aggression might be low relative to its costs
in such cases of reduced between-colony competition
[18–20], as was suggested in essence for situations of
increased competition  in which the same ratio of
costs and benefits of aggression might apply, although
Figure 1. The Self-Nonself Boundary for 16 Colonies of the Ant
Self is indicated by green, and nonself is indicated by red; means
and standard errors are depicted. ‘‘*’’ indicates a significant differ-
ence (a = 0.05) between self and nonself, as revealed by Bonfer-
roni-Holm adjusted Student’s t test with the Satterthwaite approxi-
mation. The genetic relatedness panel shows the relatednesses
within and between colonies, based on nine microsatellites. The
body surface hydrocarbons panel shows Bray-Curtis similarity in-
dex values for 26 body surface hydrocarbons measured by gas
chromatography. The discrimination panel shows the antennation-
bout lengths in one-on-one worker encounters. The aggression
panel shows the presence (‘‘yes’’) or absence (‘‘no’’) of aggression
in one-on-one worker confrontations.
Current Biology Vol 17 No 21
the absolute values would be expected to be much
Social strategies that reduce intragroup relatedness
through the adoption of foreigners might appear to be
nonadaptive. But the incorporation of additional nonre-
productive workers can augment the workforce. Also,
the loss of workers to another colony might be out-
weighed through the avoidance of erroneous self rejec-
tion [4, 22]. The lack of aggression against colonies of
the same species might save resources, which can be
invested in colony growth and reproduction, thus in-
creasing colony fitness, as argued for supercolonies
. From the vantage point of the individual, integration
into a nonself colony seems even less adaptive because
then the worker’s genes are dissimilar to those transmit-
ted. However, the reciprocal exchange of workers can
render this a stable strategy. This finding adds L. aus-
triacus to the wide range of social organisations of
Lasius, which includes supercoloniality in two species
(reviewed by [1, 2]), and confirms the genus as an ideal
system for exploring social evolution.
The consistency of nonaggressive behavior despite
the recognition of the boundary in Lasius austriacus
makes it improbable that recognition errors  trigger
nonaggressive, integrative behavior. The same might
apply to the lack of intraspecific aggression in another
20 single-queened, phylogenetically diverse ant species
with various ecologies, including activity centers from
underground to trees and food sources from scattered
to centralized (Supplemental Data). In some of these
species, integration as well as aggression occurs. All
of this suggests that the relevant benefit-cost ratio
rather than a species-specific trait turns the scales to
aggressive or integrative behavior at the self-nonself
boundary. This awareness could pave the way to an
unexpected explanation of supercolonies.
Supercolonies have evolved multiply and via various
routes [1, 2, 6–12, 22–24], and finding a general principle
for supercolony formation proves difficult . Possibly,
lumping the result of the independent pathways under
the term supercolonies—ecologically most significant,
as exemplified by their infamous invasion success [5,
7, 8, 12]—might give us little chance to disentangle
causal and correlated traits.
The social organization of integration despite discrimi-
supercolonies are defined [1, 2]: worker exchange. Con-
cerning the collateral traits of supercolonies—equally
low relatedness within and between nests, lack of dis-
crimination, and lack of aggression—only the latter is
shared. This indicates that the lack of aggression in
supercolonies might primarily have evolved through nei-
ther reduced relatedness nor reduced discrimination.
In principle, reduced relatedness within self as found
evolution [1, 2, 9], in that for such nests, the integration
of nonself individuals does not lower relatedness and
dilute nest identity by much, reducing the initial barrier
to worker exchange. However, native and invasive
cases in which some species have both single-queened
and supercolonial populations (e.g., Formica, Myrmica,
Solenopsis [1, 10, 25]) argue against relatedness reduc-
tion’s being crucial to supercolony evolution. The exis-
ants further weakens the view that supercolonial spe-
cies evolved from multiple-queened ones [1, 24].
Reduced self-nonself discrimination could also initi-
ate the cessation of aggression [2, 5, 7, 8, 12]. Although
it was originally proposed for bottleneck situations after
anthropogenic introductions , it was later suggested
that the breakdown of discrimination could also be trig-
gered by natural disturbances  or by the evolution of
a green-beard mechanism whereby recognition and
peaceful acceptance of the bearer of a specific allele ir-
respective of genetic relatedness is understood [1, 11].
In at least one species, though, individuals within super-
colonies can discriminate between nestmates and non-
tion might not be initial in their formation.
We argue that the abandonment of aggression itself
could be the first stage on the route to harmony. Under
some circumstances, this might result in social organi-
zations like that of L. austriacus. Under other circum-
stances, this might be followed by supercolony evolu-
tion, resulting in a wide range of native and invasive
supercolonies because of the wide range of phyloge-
netic constraints and ecological settings. Although until
now supercolonies were viewed as a byproduct of other
processes, we propose that the adaptive value of avoid-
ing the costs of aggression is so strong that under ap-
propriate ecological conditions, it will promote the elim-
ination of territorial aggression. The simplicity of this
scenario and the case of L. austriacus that illustrates
the principle make the argument intuitively plausible, al-
though it is not possible to test our pathway at present
because we do not know ofa single case of supercolony
evolution in progress. It remains open whether the har-
mony of Lasius austriacus has limits and reverts to ag-
gression under some conditions, but the issue is not rel-
evant to our scenario: In both native and invasive ants,
both unlimited [26, 27] and limited integration occurs,
the latter resulting in aggression between supercolonies
of the same species [8, 9]. We recognize that our find-
ings could tend to the modification of the view [7, 8,
28] that the pathways to supercolony necessarily differ
between native and invasive species, thus adding to re-
cent, independent evidence along the same lines .
Our scenario, however, does not so much challenge or
replace previous insights on native and invasive super-
colonies as it does complement them by offering
a broader view. If our scenario proves generally impor-
biology, in that many more species than are currently
thought to be potentially invasive might in fact have
the potential to become so. Our scenario is also timely
in a broader context, considering recent findings from
a diverse range of organisms, including wasps, spiders,
and bacteria [29–31], on how competition is overcome:
The circumstances are very diverse, as are the mediat-
ing mechanisms, but ecology might be central to the
evolution of cooperation in a competitive world .
Excavations revealed that neighboring colonies of Lasius austriacus
are about 10 m apart. We collected 300 workers from each of 16
colonies, four each from four populations (with two dyads of
No Aggression but Self-Nonself Discrimination
neighboring colonies 100 m apart) in eastern Austria, resulting in in-
tercolonial distances of approximately 10, 100, 1,000, 10,000, and
100,000 m. We also collected 100 workers each from two colonies
of four other formicine species, three epigaeic, having various dom-
inance hierarchy positions (Formica pratensis, top; Lasius alienus,
medium; Plagiolepis vindobonensis, bottom; ), and one hypo-
gaeic (Lasius flavus; ). We kept all ants in the laboratory for 3
weeks, following .
Following , we extracted the DNA of five workers per L. austria-
cus colony and performed microsatellite analysis for La33b, La33c,
La34c, La34e, La35g, La35h, La36a, and La36d. We also used
microsatellite primers ‘‘La35eF,’’ 50-CGTTTGCCTCCTTGTTTTG-30,
EF599953). By using KinGroup v2_01212b , we calculated the
pairwise relatedness for all 3,160 possible pairs of workers, within
and across colonies, implementing the estimator of . The allele
inspections by eye, we regarded workers as sharing a mother and
father when for all nine loci the number of alleles was % three and
one allele was shared by all workers. Following logic similar to
, we calculated the probability that two unrelated workers might
be sufficiently similar genetically to appear to be full sisters by as-
suming equal allele frequencies for the nine microsatellites typed,
which have 6, 8, 9, 9, 12, 16, 16, 17, and 17 alleles. The probability
of two unrelated individuals appearing to be full sisters is (1/6 3 2/
6) 3 (1/8 3 2/8) 3 (1/9 3 2/9) 3 (1/9 3 2/9) 3 (1/12 3 2/12) 3 (1/16
We performed one-on-one worker encounters following  but used
glass vials of 2 cm diameter, with Fluon-coated walls, combining
L. austriacus with the above four ant species, with 20 trials each.
We observed six behaviors: ignoring, antennation, avoidance, gas-
terflexion,biting, and fighting; wescored the former threeasnonag-
gressive and the latter three as aggressive. Over the 5 min observa-
tion periods, we took down the numbers and inducers of aggressive
Analogous one-on-one worker encounters were performed with
L. austriacus, with 32 trials for self (same colony) and 16 trials for
each of the five distances of nonself (other conspecific colonies), re-
sulting in 112 encounters. We recorded antennation-bout lengths
with video documentation. We repeated the encounters, mimicking
underground conditions, in 2 3 1 3 0.5 cm red film-covered cavities
of soil from native colonies, testing the soil from both colonies each,
totaling 112 encounters. Following , in eight tests across all geo-
graphic distances between colonies, we combined colony frag-
ments (50 workers) from four colonies marked with honeybee paints
for 24 hr and took down any group formation. To analyze food ex-
change, we kept marked workers for 24 hr, in one treatment without
food and in another treatment with honeywater. In eight trials per
distance, totaling to 40 trials, we placed one starving, one fed con-
colonial, and one fed allocolonial worker (different honeywater
dyes) together and subsequently detected any food exchange by
analyzing the gaster of killed ants.
We performed blind bioassays by using body surface hydrocar-
bon extracts. We successively extracted five workers in 100 ml hex-
ane, 100 ml ethyl acetate, and 100 ml ethanol for 90 s each and then
pooled the extracts. We covered the bottom of 60 glass vials of
2 cm diameter each with 1 ml paraffin oil as a keeper substance.
We applied 1 ml solvent to one 120 degree sector of each bottom,
as a control, to the second sector 1 ml self extracts, and to the third
1 ml nonself extracts. We performed two trials per vial, involving one
worker from each of the two extracted colonies, under dark condi-
tions. Every 15 min, we took down the worker’s position (control,
self, nonself), during a 10 s period of light, with 16 repeats (1,920 ob-
servations). After each observation, we turned the vial 120 degrees
and forced the worker to move to another sector.
We extracted five workers per colony separately for 90 s with 25 ml
of a mixture of 50% hexane, 45% ethyl acetate, and 5% ethanol
including 0.00025% heptacosane. Gas chromatography (GC) was
performed on a GC8000 (Fisons Instruments, Italy; flame ionization
detector [FID]; DB-5 column diameter 0.25 mm, length 30 m, film
0.25 mm; carrier: hydrogen; 1 ml split-splitless injection; 60?C for
2min,15?C min21to200?C and4?Cmin21to320?Cfor5min).Profile
synchronization was achieved with retention times. Gas chromatog-
raphy-mass spectrometry (MS) analyses were run on the GC8000
helium; 60?C for 3 min, 15?C min21to 200?C and 3?C min21to 320?C
for 5 min; electron impact [EI] mode at 70eV; full-scan mode 50–655
amu; scan time: 1.8 s). With MS data, substances were determined
as hydrocarbons. MS data were also used so that the GC synchro-
nization could be checked. Peak intensities were calculated by the
integration of peak areas with Chrom-Card (Fisons Instruments).
We followed  in calculating the similarity in the relative hydrocar-
bon intensities between each of the 3,160 possible pairs of workers
with the Bray-Curtis similarity index implemented in PRIMERv5 (Ply-
mouth, UK). We applied Student’s t test by using the Satterthwaite
approximation implemented in SAS9 (SAS Institute [Cary, NC]) and
the Bonferroni-Holm correction  to compare intracolonial data
with data of each of the five geographic distances, for relatedness,
antennation-bout length, and Bray-Curtis values of hydrocarbons
and to compare the summed values of the 16 observations for the
self and the nonself sector in the bioassays. We used the analysis
of ranked similarities randomization in PRIMERv5 to test the proba-
bility that the pairwise similarities of hydrocarbons within and be-
tween populations and colonies, as well as within and between indi-
vidual identities, were the same. For this we randomly allocated
numeric identities, 1–5, to the individuals of each colony.
One table is available at http://www.current-biology.com/cgi/
We thank S. Krumbo ¨ck, G. Motlik, and A.Stradnerfor help in the lab-
oratory, W. Francke for methodological advice, A. Nikiforov for me-
diating contacts, J.J. Boomsma, S. Cremer, L. Keller, and J.S. Pe-
dersen for critically reading an earlier version of the manuscript,
and three anonymous referees for constructive and stimulating crit-
icism. This research was supported by a FWF P-17219-B06 grant.
Received: August 2, 2007
Revised: September 20, 2007
Accepted: September 20, 2007
Published online: October 25, 2007
1. Crozier, R.H., and Pamilo, P. (1996). Evolution of Social Insect
Colonies (New York: Oxford University Press).
2. Bourke, A.F.G., and Franks, N.R. (1995). Social Evolution in Ants
(Princeton, New Jersey: Princeton University Press).
3. Davies, N.B., and Houston, A.I. (1984). Territory economics. In
Behavioral Ecology: An Evolutionary Approach, J.R. Krebs and
N.B. Davies, eds. (Oxford: Blackwell Scientific), pp. 148–169.
4. Reeve, H.K. (1989). The evolution of conspecific acceptance
thresholds. Am. Nat. 133, 407–435.
5. Holway, D.A., Suarez, A.V., and Case, T.J. (1998). Loss of intra-
specific aggression in the success of a widespread invasive so-
cial insect. Science 282, 949–952.
6. Pedersen, J.S., Krieger, M.J.B., Vogel, V., Giraud, T., and Keller,
L. (2006). Native supercolonies of unrelated individuals in the in-
vasive Argentine ant. Evolution Int. J. Org. Evolution 60, 782–
7. Tsutsui, N.D., Suarez, A.V., Holway, D.A., and Case, T.J. (2000).
Reduced genetic variation and the success of an invasive spe-
cies. Proc. Natl. Acad. Sci. USA 97, 5948–5953.
8. Giraud, T., Pedersen, J.S., and Keller, L. (2002). Evolution of
supercolonies: The argentine ants of southern Europe. Proc.
Natl. Acad. Sci. USA 99, 6075–6079.
Current Biology Vol 17 No 21
9. Holzer, B., Chapuisat, M., Kremer, N., Finet, C., and Keller, L.
(2006). Unicoloniality, recognition and genetic differentiation in
a native Formica ant. J. Evol. Biol. 19, 2031–2039.
10. Elias, M., Rosengren, R., and Sundstro ¨m, L. (2005). Seasonal
polydomy and unicoloniality in a polygynous population of the
red wood ant Formica truncorum. Behav. Ecol. Sociobiol. 57,
11. Krieger, M.J.B. (2005). To b or not to b: A pheromone-binding
protein regulates colony social organization in fire ants. Bioes-
says 27, 91–99.
12. Errard, C., Delabie, J., Jourdan, H., and Hefetz, A. (2005). Inter-
continental chemical variation in the invasive ant Wasmannia
auropunctata (Roger) (Hymenoptera Formicidae): A key to the
invasive success of a tramp species. Naturwissenschaften 92,
13. Foster,K.R., andXavier,J.B.(2007). Cooperation:Bridgingecol-
ogy and sociobiology. Curr. Biol. 17, R319–R321.
14. Vander Meer, R.K., and Morel, L. (1998). Nestmate recognition in
ants. In Pheromone Communication in Social Insects: Ants,
Wasps, Bees and Termites, M.D. Breed, K.E. Espelie, R.K. Van-
der Meer, and M.L. Winston, eds. (Boulder, Colorado: Westview
Press), pp. 79–103.
15. Pirk, C.W.W., Neumann, P.,Moritz,R.F.A.,and Pamilo, P.(2001).
Intranest relatedness and nestmate recognition in the meadow
ant Formica pratensis (R.). Behav. Ecol. Sociobiol. 49, 366–374.
16. Steiner, F.M., Schlick-Steiner, B.C., Scho ¨dl, S., Espadaler, X.,
Seifert, B., Christian, E., and Stauffer, C. (2004). Phylogeny and
bionomics of Lasius austriacus (Hymenoptera, Formicidae). In-
sectes Sociaux 51, 24–29.
17. Pontin, A.J. (1978). The numbers and distribution of subterra-
nean aphids and their exploitation by the ant Lasius flavus
(Fabr.). Ecol. Entomol. 3, 203–207.
18. Katzerke, A., Neumann, P., Pirk, C.W.W., Bliss, P., and Moritz,
R.F.A. (2006). Seasonal nestmate recognition in the ant Formica
exsecta. Behav. Ecol. Sociobiol. 61, 143–150.
19. Mabelis, A.A. (1979). Wood ant wars. Neth. J. Zool. 29, 451–620.
20. Downs, S.G., and Ratnieks, F.L.W. (2000). Adaptive shifts in
honey bee (Apis mellifera L.) guarding behavior support predic-
tions of the acceptance threshold model. Behav. Ecol. 11, 326–
21. Ross, K.G., and Keller, L. (1995). Ecology and evolution of social
organization: Insights from fire ants and other highly eusocial in-
sects. Annu. Rev. Ecol. Syst. 26, 631–656.
22. Chapuisat, M., Bernasconi, C., Hoehn, S., and Reuter, M. (2005).
Nestmate recognition in the unicolonial ant Formica paralugub-
ris. Behav. Ecol. 16, 15–19.
23. Ho ¨lldobler, B., and Wilson, E.O. (1990). The Ants (Cambridge,
Massachusetts: Harvard University Press).
24. Wilson, E.O. (1971). The Insect Societies (Cambridge, Massa-
chusetts: Harvard University Press).
25. Seifert, B. (2007). Die Ameisen Mittel- und Nordeuropas (Tauer,
Germany: lutra Verlags- und Vertriebsgesellschaft).
26. Schlick-Steiner, B.C., Steiner, F.M., Stauffer, C., and Bu-
schinger, A. (2005). Life history traits of a European Messor har-
vester ant. Insectes Sociaux 52, 360–365.
27. Tschinkel, W.R. (2006). The Fire Ants (Cambridge, Massachu-
setts: Harvard University Press).
28. Tsutsui, N.D., and Suarez, A.V. (2003). The colony structure and
population biology of invasive ants. Conserv. Biol. 17, 48–58.
29. Sumner, S., Lucas, E., Barker, J., and Isaac, N. (2007). Radio-
tagging technology reveals extreme nest-drifting behavior in
a eusocial insect. Curr. Biol. 17, 140–145.
30. Evans, T.A. (1999). Kin recognition in a social spider. Proc. R.
Soc. Lond. B. Biol. Sci. 266, 287–292.
31. Brockhurst, M.A., Buckling, A., and Gardner, A. (2007). Cooper-
ation peaks at intermediate disturbance. Curr. Biol. 17, 761–765.
32. Steiner, F.M., Arthofer, W., Schlick-Steiner, B.C., Crozier, R.H.,
and Stauffer, C. (2007).Eleven microsatellite loci in the sociobio-
logically enigmatic ant Lasius austriacus (Hymenoptera: Formi-
cidae). Mol. Ecol. Notes 7, 498–500.
33. Konovalov, D.A., Manning, C., and Henshaw, M.T. (2004). KIN-
GROUP: A program for pedigree relationship reconstruction
and kin group assignments using genetic markers. Mol. Ecol.
Notes 4, 779–782.
34. Milligan, B.G. (2003). Maximum-likelihood estimation of related-
ness. Genetics 163, 1153–1167.
ponera sp. 12 ants inferred from microsatellite analysis. Mol.
Ecol. 10, 167–173.
36. Astruc, C., Malosse, C., and Errard, C. (2001). Lack of intraspe-
cific aggression in the ant Tetramorium bicarinatum: A chemical
hypothesis. J. Chem. Ecol. 27, 1229–1248.
37. Schlick-Steiner, B.C., Steiner, F.M., Ho ¨ttinger, H., Nikiforov, A.,
Mistrik, R., Schafellner, C., Baier, P., and Christian, E. (2004). A
butterfly’s chemical key to various ant forts: Intersection-odour
or aggregate-odour multi-host mimicry? Naturwissenschaften
38. Holm, S. (1979).A simple sequentially rejective multiple test pro-
cedure. Scand. J. Stat. 6, 65–70.
TheLa35eFandLa35eR microsatelliteprimersequences reportedin
this paper have been deposited in the GenBank with the accession
No Aggression but Self-Nonself Discrimination
Abandoning Aggression but Maintaining
Self-Nonself Discrimination as a First
Stage in Ant Supercolony Formation
Florian M. Steiner, Birgit C. Schlick-Steiner,
Karl Moder, Christian Stauffer, Wolfgang Arthofer,
Alfred Buschinger, Xavier Espadaler,
Erhard Christian, Katrin Einfinger, Eberhard Lorbeer,
Christa Schafellner, Manfred Ayasse,
and Ross H. Crozier
S1. de Souza, D.J., Castro della Lucia, T.M., and de Almeida Bar-
mex subterraneus molestans from monogynous and polygy-
nous colonies. Brazilian Archives of Biology and Technology
S2. Lenoir, A., Isingrini, M., and Nowbahari, N. (1987). Colony rec-
ognition in the ant Cataglyphis cursor (Hymenoptera: Formici-
dae). In Chemistry and Biology of Social Insects, J. Eder and H.
Rembold, eds. (Mu ¨nchen, Germany: Peperny), pp. 476–477.
S3. Pearcy, M., Aron, S., Doums, C., and Keller, L. (2004). Condi-
tional use of sex and parthenogenesis for worker and queen
production in ants. Science 306, 1780–1783.
S4. Cerda ´, X., Retana, J., and Cros, S. (1998). Critical thermal limits
in Mediterranean ant species: Trade-off between mortality risk
and foraging performance. Funct. Ecol. 12, 45–55.
S5. Debout, G., Provost, E., Renucci, M., Tirard, A., Schatz, B., and
McKey, D. (2003). Colony structure in a plant-ant: Behavioural,
chemical and genetic study of polydomy in Cataulacus mckeyi
(Myrmicinae). Oecologia 137, 195–204.
S6. Maschwitz, U., Dorow, W.H.O., Buschinger, A., and Kalytta, G.
(2000).Socialparasitism involvingantsofdifferent subfamilies:
Polyrhachis lama (Formicinae) an obligatory inquiline of Diac-
amma sp. (Ponerinae) in Java. Insectes Sociaux 47, 27–35.
S7. Berghoff, S.M., Gadau, J., Winter, T., Linsenmair, K.E., and
Maschwitz, U. (2003). Sociobiology of hypogaeic army ants:
Characterization of two sympatric Dorylus species in Borneo
and their colony conflicts. Insectes Sociaux 50, 139–147.
S8. Wenseleers,T., Ito,F.,van Borm,S.,Huybrechts,R.,Volckaert,
F., and Billen, J. (1998). Widespread occurrence of the micro-
organism Wolbachia in ants. Proc. R. Soc. Lond. B. Biol. Sci.
S9. Berghoff, S.M., Weissflog, A., Linsenmair, K.E., Hashim, R.,
and Maschwitz, U. (2002). Foraging of a hypogaeic army ant:
a long neglected majority. Insectes Sociaux 49, 133–141.
S10. Schneirla, T.C., and Brown, R.Z. (1950). Army-ant life and be-
havior under dry-season conditions, 4: Further investigation
of cyclic processes in behavioral and reproductive functions.
Bull. Am. Mus. Nat. Hist. 95, 263–353.
S11. Schneirla, T.C. (1934). Raiding and other outstanding phenom-
ena in the behavior of army ants. Proc. Natl. Acad. Sci. USA 20,
S12. Goodloe, L.P., and Topoff, H. (1987). Pupa acceptance by
slaves of the social-parasitic ant Polyergus (Hymenoptera:
Formicidae). Psyche (Cambridge) 94, 293–302.
S13. Fourcassie ´, V., and Traniello, J.F.A. (1994). Food searching
behvaviour in the ant Formica schaufussi (Hymenoptera, For-
micidae): Response of naive foragers to protein and carbohy-
drate food. Anim. Behav. 48, 69–79.
S14. Post, D.C., and Jeanne, R.L. (1982). Rate of exploitation of ar-
boreal baits by ants in an old-field habitat in Wisconsin. Am.
Midl. Nat. 108, 88–95.
S15. Steiner, F.M., Schlick-Steiner, B.C., Scho ¨dl, S., Espadaler, X.,
Seifert, B., Christian, E., and Stauffer, C. (2004). Phylogeny
and bionomics of Lasius austriacus (Hymenoptera, Formici-
dae). Insectes Sociaux. 51, 24–29.
S16. Seifert, B. (2007). Die Ameisen Mittel- und Nordeuropas (Tauer,
Germany: lutra Verlags- und Vertriebsgesellschaft).
S17. Espadaler, X., Bernal, V., and Rojo, M. (2006). Lasius brunneus
(Hymenoptera, Formicidae) una plaga del corcho en el NE de
Espan ˜a: I. Biologia y pruebas de control. Boletı ´n de sanidad
vegetal. Plagas 32, 411–424.
S18. Pontin, A.J. (1961). Population stabilization and competition
between the ants Lasius flavus (F.) and L. niger (L.). J. Anim.
Ecol. 30, 47–54.
S19. Debout, G., Schatz, B., Elias, M., and McKey, D. (2007). Poly-
domy in ants: What we know, what we think we know, and
what remains to be done. Biological Journal of the Linnean So-
ciety. 90, 319–348.
S20. Wilson, E.O. (1971). The Insect Societies (Cambridge, Massa-
chusetts: Harvard University Press).
S21. Fjerdingstad, E.J., Gertsch, P.J., and Keller, L. (2002). Why do
some social insect queens mate with several males? Testing
the sex-ratio manipulation hypothesis in Lasius niger. Evolu-
tion Int. J. Org. Evolution 56, 553–562.
S22. Jaffe, K., and Marcuse, M. (1983). Nestmate recognition and
territorial behaviour in the ant Odontomachus bauri Emery
(Formicidae: Ponerinae). Insectes Sociaux 30, 466–481.
S23. Ehmer, B., and Ho ¨lldobler, B. (1995). Foraging behavior of
Odontomachus bauri on Barro Colorado Island, Panama. Psy-
che (Cambridge) 102, 215–224.
S24. Armbrecht, I., Jimenez, E., Alvarez, G., Ulloa-Chacon, P., and
Armbrecht, H. (2001). An ant mosaic in the Colombian rain for-
est of Choco (Hymenoptera: Formicidae). Sociobiology 37,
S25. Harrison, J.S., and Gentry, J.B. (1981). Foraging pattern, col-
ony distributin, and foraging range of the Florida harvester
ant, Pogonomyrmex badius. Ecology 62, 1467–1473.
S26. Rheindt, F.E., Gadau, J., Strehl, C.-P., and Ho ¨lldobler, B.
(2004). Extremely high mating frequency in the Florida har-
vester ant (Pogonomyrmex badius). Behav. Ecol. Sociobiol.
S27. MacMahon, J.A., Mull, J.F., and Crist, T.O. (2000). Harvester
ants (Pogonomyrmex spp.): Their community and ecosystem
influences. Annu. Rev. Ecol. Syst. 31, 265–291.
S28. Gordon, D. (1989). Ants distinguish neighbors from strangers.
Oecologia 81, 198–200.
S29. Whitford, W.G., Johnson, P., and Ramirez, J. (1976). Compara-
tive ecology of the harvester ants Pogonomyrmex barbatus (F.
Smith) and Pogonomyrmex rugosus (Emery). Insectes Sociaux
S30. Volny, V.P., and Gordon, D.M. (2002). Characterization of poly-
morphic microsatellite loci in the red harvester ant, Pogono-
myrmex barbatus. Mol. Ecol. Notes 2, 302–303.
S31. Ho ¨lldobler, B. (1976). Recruitment behavior, home range orien-
tation and territoriality in harvester ants, Pogonomyrmex.
Behav. Ecol. Sociobiol. 1, 3–44.
S32. Whitford, W.G.(1976).Foraging behaviorofChihuahuan desert
harvester ants. Am. Midl. Nat. 95, 455–458.
S33. Gadau, J., Strehl, C.-P., Oettler, J., and Ho ¨lldobler, B. (2003).
Determinants of intracolonial relatedness in Pogonomyrmex
rugosus (Hymenoptera; Formicidae): Mating frequency and
brood raids. Mol. Ecol. 12, 1931–1938.
S34. Goodloe, L., Sanwald, R., and Topoff, H. (1987). Host specific-
ity in raiding behavior of the slave-making ant Polyergus luci-
dus. Psyche (Cambridge) 94, 39–44.
S35. Marlin, J.C. (1971). The mating, nesting and ant enemies of
Polyergus lucidus Mayr (Hymenoptera: Formicidae). Am.
Midl. Nat. 86, 181–189.
S36. Maschwitz, U., Go, C., Kaufmann, E., and Buschinger, A.
(2004). A unique strategy of host colony exploitation in
a parasitic ant: workers of Polyrhachis lama rear their brood in
neighbouring host nests. Naturwissenschaften 91, 40–43.
S37. Maschwitz, U., Go, C., Dorow, W.H.O., Buschinger, A., and
Kohout, R.J. (2003). Polyrhachis loweryi (Formicinae): A guest
ant parasitizing Rhytidoponera sp. (Ponerinae) in Queensland,
Australia. Insectes Sociaux 50, 69–76.
S38. Provost, E. (1991). Non-nestmate kin recognition in the ant
Leptothorax lichtensteini, evidence that genetic factors regu-
late colony recognition. Behav. Genet. 21, 151–167.
S39. Provost, E. (1979). Etude de la fermeture de la societe de four-
mis chez diverses especes de Leptothorax et chez Campono-
tus lateralis (Hymenopteres, Formicidae). Comptes Rendus
Hebdomadaires des Seances de l’Academie des Sciences
Serie D Sciences Naturelles 288, 429–432.
S40. Heinze, J., Foitzik, S., Hippert, A., and Ho ¨lldobler, B. (1996).
Apparent dear-enemy phenomenon and environment-based
recognition cues in the ant Leptothorax nylanderi. Ethology
S41. Buschinger, A. (1968). Mono- und Polygynie bei Arten der Gat-
tung Leptothorax Mayr (Hymenoptera Formicidae). Insectes
Sociaux 15, 217–225.
S42. Bolton, B. (2003). Synopsis and Classification of Formicidae
(Gainesville, Florida: American Entomological Institute).
Table S1. Reports of Lack of Intraspecific Aggression of Workers of Single-Queened Ant Colonies
subterraneus [S1]; [S1]
Cataglyphis cursor [S2];
Cataulacus mckeyi [S5];
Diacamma sp [S6]; [S6]
Dorylus laevigatus [S7];
Eciton hamatum [S10];
Formica schaufussi [S12];
Lasius austriacus [S15];
Lasius brunneus [X.E.,
unpublished data]; [S16]
Lasius flavus [S18]; [S19]
Lasius niger [S20]; [S21]
[S28, S29]; [S30]
[S29, S32]; [S33]
Polyergus lucidus [S34];
Polyrhachis lama [S6,
Polyrhachis loweryi [S37];
[S38, S39]; [S38]
Myr tro [S1]gro [S1]fun [S1]mdo [S1]n/a yes [S1]
Fortem [S3]gro [S2] sca, tne [S4]mdo [S2]sub [S4] yes [S2]
Myr tro [S5]tre [S5] tne [S5]pdo [S5]enc [S5] yes [S5]
Ecisub, tro [S11] gro [S11]pre [S11] biv [S11]tnt [S11]no
Fortem [S13] gro [S13]sca, tne [S13]
mdo [S12] ter [S14] yes [S12]
Fortem [S15]bel [S15]trn [S15]mdo [S15] n/a no
For tem [S16] tre [S16] trn, sca [S16, S17]mdo [S16] sub [S16]yes [S16]
trn, pred [S16]
Myrtem [S27]gro [S27] gra [S27] mdo [S25]ter [S25] yes [S25]
Myr tem [S27]
gro [S27] gra [S27]mdo [S28] ter [S28]yes [S31]
Myrtem [S27]gro [S27]gra [S27] mdo [S29]ter [S29]yes [S31]
Fortem [S34]gro [S34] ant [S34]mdo [S35] enc [S35]no
Forsub, tro [S6]bel [S6]ant [S6] soc [S6]n/ano
For tro [S37]bel [S37]ant [S37]soc [S37]n/ayes [S37]
Myrtem [S38] gro [S38]omn [S38] mdo [S38]sub [S16]yes [S38]
Myrtem [S16] gro [S16] omn [S16]pdo [S19]sub [S16] yes [S40]
Myr tem [S16]gro [S16]omn [S16] mdo [S16]sub [S16]no
The completeness of the list cannot be claimed because notes on the lack of intraspecific aggression of workers of single-queened ant colonies
are often found in marginal notes of publications, rendering a systematic literature search difficult. We have included notes on species that were
found to entertain single-queened colonies, i.e., colonies with a single egg-laying individual, in the mentioned reference but are reported else-
a‘‘Nonaggression’’ indicates references for a lack of intraspecific aggresion; ‘‘Single Queen’’ indicates references for a single egg-laying individ-
ual per colony.
bThe following abbreviations are used: Dorylinae (Dor), Ecitoninae (Eci), Formicinae (For), Myrmicinae (Myr), and Ponerinae (Pon). Information is
taken from [S42].
cThe following abbreviations are used: subtropical (sub), temperate (tem), and tropical (tro).
dThe following abbreviations are used: activities centered on below ground surface (bel), activities centered on ground surface (gro), and activ-
ities centered on trees and shrubs (tre).
eThe following abbreviations are used: being fed by another ant species (ant), fungivorous (fun), granivorous (gra), omnivorous (omn), predator
(pre), food robbery from other ant nests (rob), scavenger (sca), trophobiosis, partner outside nest, or nectarivory (tne), and trophobiosis, partner
in nest (trn).
of other ant species (soc).
gThe following abbreviations are used: encounter (enc), submissive (sub), territorial (ter), and top, but nonterritorial (tnt).
h‘‘no’’ indicates that there is no reference known for any intraspecific aggression, and ‘‘yes’’ indicates that intraspecific aggression is known,
in parallel to the mentioned lack of intraspecific aggression, indicating that aggression might arise under certain circumstances or in certain