Accepted Manuscript

Facultative slave-making ants formica sanguinea label their slaves with own
recognition cues instead of employing the strategy of chemical mimicry

Tomasz Wodarczyk, Lech Szczepaniak

PII: S0022-1910(16)30164-0
DOI: http://dx.doi.org/10.1016/j.jinsphys.2016.10.016
Reference: IP 3565

To appear in: Journal of Insect Physiology

Received Date: 9 June 2016

Revised Date: 24 October 2016
Accepted Date: 24 October 2016

Please cite this article as: Wodarczyk, T., Szczepaniak, L., Facultative slave-making ants formica sanguinea label
their slaves with own recognition cues instead of employing the strategy of chemical mimicry, Journal of Insect
Physiology (2016), doi: http://dx.doi.org/10.1016/j.jinsphys.2016.10.016

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
FACULTATIVE SLAVE-MAKING ANTS Formica sanguinea LABEL THEIR SLAVES WITH OWN
RECOGNITION CUES INSTEAD OF EMPLOYING THE STRATEGY OF CHEMICAL MIMICRY

TOMASZ WODARCZYK1 *, LECH SZCZEPANIAK2

1
Department of Invertebrate Zoology, University of Biaystok, Ciokowskiego 1J, 15-245 Biaystok,
Poland, e-mail: tomwlo@gmail.com, tel. +48 85 738 84 16

2
Department of Chemistry of Environment, University of Biaystok, Ciokowskiego 1K, 15-245
Biaystok, Poland

Abstract: Slave-making ant species use the host workforce to ensure normal colony functioning.
Slaves are robbed as pupae from their natal nest and after eclosion, assume the parasite colony as their
own. A possible factor promoting the successful integration of slaves into a foreign colony is
congruence with the slave-makers in terms of cuticular hydrocarbons, which are known to play the
role of recognition cues in social insects. Such an adaptation is observed in the obligate slave-making
ant species, which are chemically adjusted to their slaves. To date, however, no reports have been
available on facultative slave-making species, which represent an earlier stage of the evolution of
slavery. Such an example is Formica sanguinea, which exploit F. fusca colonies as their main source
of a slave workforce. Our results show that F. sanguinea ants have a distinct cuticular hydrocarbon
profile, which contains compounds not present in free-living F. fusca ants from potential target nests.
Moreover, enslaved F. fusca ants acquire hydrocarbons from their slave-making nestmates to such an
extent that they become chemically differentiated from free-living, conspecific ants. Our study shows
that F. sanguinea ants promote their own recognition cues in their slaves, rather than employing the
strategy of chemical mimicry. Possible reasons why F. sanguinea is not chemically well adjusted to its
main host species are discussed in this paper.

Key words-Ants, slave-making, nestmate recognition, cuticular hydrocarbons

1
1. Introduction

Social parasites rely on a host workforce to ensure colony functioning. Among eusocial Hymenoptera,
they are distributed among different lineages showing convergent evolution in many aspects of
physiology, behavior, morphology and life history (Wilson, 1971). For successful host exploitation, a
social parasite has to break the recognition barrier of the focal colony, since social insects as a rule
reject individuals bearing a foreign chemical label. In ants, hydrocarbons spread over the body surface
are known to be prevalent recognition cues (reviewed in van Zweden and dEttorre, 2010). Nestmates
exchange these compounds during trophallaxis and allogrooming, leading to the formation of a
uniform odor, characteristic of the colony (Soroker et al., 1995; Soroker et al., 1998; Dahbi et al.,
1999; Boulay et al., 2000; Lenoir et al., 2001a and b). In most social insects, the recognition
mechanism is based on phenotype matching, i.e. they compare specific traits of an encountered
individual with the reference odor, which might be encoded in the higher brain centers or at the level
of the peripheral sensory system (Ozaki and Hefetz, 2014). A decision on acceptance or rejection is
made based on the dissimilarity between perceived odor and the reference (see Sherman and Holmes,
1985). The reference odor is memorized during a short period after eclosion from the pupa. The role of
early experience has clearly been demonstrated in the paper wasps, which learn their characteristic
colony hydrocarbon mixture from the nest material. For young Polistes fuscatus females, it is critical
to have contact with nest material for further amicable behavior toward other colony members
occupying that nest (Pfennig et al., 1983). Similarly, the critical period for learning colony odor has
been evidenced for ants (Morel, 1983; Errard and Jaisson, 1991).

One type of social parasitism, whose occurrence is restricted to ants, is slavery. Workers of slave-
making ant species raid host colonies to steal pupae, which are brought to the parasite's nest and are
allowed to complete their development (Wilson, 1971). Workers of the host species which emerge
from the pupae (so-called slaves) memorize the parasite's colony odor and treat parasite workers as
nestmates (D'Ettorre and Heinze, 2001). Dealated females of slave-making species found new colonies
by the usurpation of a host colony and replacing resident workers. However, they face the threat of
being recognized and killed by the host workers. Consequently, they adopt various chemical strategies
to invade and establish themselves in a host colony. Polyergus ant queens use a secretion of the
Dufour's gland to repel or appease resident workers (D'Ettorre et al., 2000, Visicchio et al., 2000).
Moreover, Polyergus rufescens queens have been found to adopt a strategy of chemical
insignificance prior to invading the host colony. This strategy relies on reduced amount of
hydrocarbons on the cuticle coupled with elevated proportion of n-alkanes (DEttorre and Errard,
1998; Lenoir et al., 2001b). A similar strategy is also employed by colony-founding queens of the

2
socially parasitic paper wasp genus Polises (reviewed in Lorenzi 2006). The n-alkanes play a less
important role in the nestmate recognition in social insects than alkenes and branched alkanes (Dani etl
al., 2001; Dani et al., 2005; Martin et. al., 2008c; Krasnec and Breed, 2013), thus their prevalence in
the CHC blend is believed to hamper detection by resident workers. However, the chemical
insignificance strategy imposes a cost of water loss owing to the elevated permeability of the cuticle
with a reduced hydrocarbon layer (Gibbs and Rajpurohit, 2010). This might explain why social
parasites more often employ the strategy of chemical mimicry of their host CHC profile, either by
active biosynthesis or by the acquisition of recognition cues from the environment (Bagnres and
Lorenzi, 2010).

Workers of the European slave-making species Harpagoxenus sublaevis workers have been found to
mimic their host in terms of cuticular hydrocarbons (CHCs). Consequently, the CHC profile of the
parasite differs depending on whether the slaves are Leptothorax acervorum or Leptothorax muscorum
or both (Kaib et al., 1993; Bauer et al., 2010). Local convergence of CHC profiles in sympatric
populations has been postulated to be driven by host-parasite co-evolution for Rossomyrmex minuchae
and its host species Proformica longiseta (Ruano et al., 2011). Notwithstanding, species-specific
differences in the R. minuchae colonies have been revealed with slaves being more similar to
conspecific ants from free-living colonies than their parasites (Errard et al., 2006). Such interspecific
differences have also been observed in the colonies of the obligatory slave-making species Polyergus
rufescens and P. samurai (Bonavita-Cougourdan et al., 1996; Liu et al., 2003), although there is no
evidence on whether these differences are relevant for nestmate recognition. Nevertheless, in both
these species, CHC profiles have been demonstrated to be adjusted to slaves, so that they depend on
which slave species is actually used (Yamaoka, 1990; Bonavita-Cougourdan et al., 1996; Bonavita-
Cougourdan et al., 1997; D'Ettorre et al., 2002). The chemical resemblance of host species has also
been demonstrated in the North American slave-making species Protomognathus americanus. Even
more advanced chemical deception is achieved by Temnothorax pilagens, which perfectly mimics the
odor of its main host species T. ambiguus (Kleeberg and Foitzik, 2016).

The general tendency of slave-making species to be chemically adjusted to their host species raises the
question about the possible adaptive benefits associated with such congruence. One possibility is that
mimicking host odor enhances the chances of successful colony usurpation by newly mated parasite
queens. A change in cuticular chemistry leading to the mimicry of the resident queen's odor has been
postulated to play a crucial role in gaining the acceptance of host workers during colony founding by
Polyergus breviceps queens (Johnson et al., 2001). High inter-colonial variation in the recognition
labels of F. fusca ants is suggested to have evolved in response to the pressure exerted by the species,

3
which found new colonies through temporary social parasitism (Martin et al., 2011). Another
explanation is that odor congruence facilitates the raids of slave-maker workers targeted at host
colonies (see Discussion in Bauer et al., 2010). First, being hardly distinguishable from workers of the
host species might prevent parasite scouts from being killed by ants from the potential target nest. It
has been observed that Temnothorax longispinosus exhibits elevated aggression toward slave-makers
relative to that toward conspecific workers (Alloway, 1990; Scharf et al., 2011). Such behavior might
prevent the host colony from being raided by slave-maker workers guided by a scout's secretions
(Scharf et al., 2011). Second, intraspecific aggression has been found to be lower than interspecific
aggression in many ant species (for references see Stuart, 1991). Thus, possession of a chemical
signature similar to that of the host species might reduce losses among slave-makers during the fight
against ants defending the raided nest. In accordance with this explanation, the slave-making ant
Temnothorax pilagens, which possesses the chemical signature of its host species, faces only a little
aggression from the workers of raided colonies. Ants of this species are able to enslave adult host
workers, which is an unusual phenomenon given the widely believed mechanism of slave integration
into a parasite society (Kleeberg and Foitzik, 2016). Moreover, taking into consideration the similarity
between the chemical profiles of adults and eggs in many social insects (see references in Martin et al.,
2011), chemical mimicry might facilitate the acceptance of a parasite's brood by host workers. Finally,
the chemical discrepancy between heterospecific nestmates might result in conflicts within the parasite
colony (reviewed in Czechowski and Godziska, 2014), as well as recognition errors owing to
variations in the recognition labels among nestmates (Lorenzi, 2003; Wodarczyk 2016).

All of the above-mentioned slave-making species are obligatory social parasites, i.e. slaves are needed
in their colonies to perform all tasks other than reproduction and robbing new pupae. It would be,
however, interesting to compare their chemical strategy to facultative parasites, which represent a
lower stage in the evolution of slavery (Darwin, 1859; Mori et al., 2001). Such a species is Formica
sanguinea, whose colonies almost always contain slaves of the subgenus Serviformica. However, in
opposition to the closely related Polyergus genus, as a rule, slaves in Formica sanguinea colonies
constitute a minority of colony members (Wheeler, 1910; Czechowski, 1996; Savolainen and
Deslippe, 1996). Moreover, slaves are not needed for a F. sanguinea colony to survive, since the slave-
maker workers are able to perform all normal duties of the non-reproductive caste. Thus, studies on
this species could be of special value for understanding the role and evolution of chemical congruency
between a slave-maker and its slaves. In our previous study we found species-specific differences in
CHC profile in an artificially established mixed colony of F. sanguinea and F. rufa, the latter being
hardly ever used as a slave species in the nature (Wodarczyk and Szczepaniak 2014). Here we present
result of a study on ants collected from F. sanguinea field colonies containing workers of the primary

4
host species F. fusca. Ants were studied for their CHCs to reveal whether differences exist between
both species in the mixed colony, and whether enslaved ants differ from free-living conspecifics. We
also conducted species-separation experiments to demonstrate how ants' CHCs are influenced by the
presence of heterospecific individuals.
Based on the raiding behavior of F. sanguinea ants (e.g. Mori et. al., 2000), it is reasonable to assume
that slave nestmates stem from multiple host colonies. Odor-variability between slaves has been
postulated to partly account for their restricted ability to discriminate ants from alien parasitic colonies
(Wodarczyk 2016, Torres and Tsutsui 2016). Thus, we examined whether slaves are characterized by
an increased odor variability as compared to their free-living conspecifics.

2. Methods
2.1. Sampling of ants from the field colonies
To determine the CHC profiles of ants living in natural conditions, 11 F. sanguinea colonies located in
the Knyszyn Forest (5311' N, 2320' E, Poland) were excavated and sampled for slave-making
workers and their slaves during two seasons, in 2013 and 2014. For nine of these colonies, two or
three nests of free-living F. fusca were searched for within a radius of 25 m by tracking the ants
returning to their nests. A total of 21 F. fusca colonies were found in this way and sampled for
workers. Ants were collected from inside the nest and killed by freezing within 8 hours after
collection. For species-separation experiments, eight queenright F. sanguinea stock colonies were
collected from the Knyszyn Forest during the spring of 2014 and maintained in the laboratory. Prior to
collection, four of these colonies were included in the sampling procedure described above.

2.2. Species-separation experiment

From each of the eight queenright stock colonies, three queenless daughter colonies were established:
one mixed and two homospecific (one per species). Attention was paid to maintain the original
proportion of slaves in the mixed daughter colonies (9 to 54%, mean = 34%). The number of ants in
the homospecific colonies was similar to the number of ants of respective species in the mixed colony,
ranging from 170 to 300 for F. sanguinea and from 22 to 200 for F. fusca. Colonies were provided
with eggs removed from the stock colony and the brood was removed at the pupal stage to not allow F.
fusca homospecific colonies to transform into mixed ones. Ants were kept in plastic boxes (25 x 25 x
12 cm) provided with a thin layer of soil, a test tube partly filled with water trapped by a cotton plug,
covered with sheet of aluminum foil and mosses. The inner sides of the nest box walls were coated
with liquid paraffin to prevent ants from escaping. The paraffin was chromatographed to verify the
possibility of sample contamination and no compound present on the ants' cuticles was detected.
Colonies were regularly moisturized and ants were fed with diluted honey and Galleria mellonella

5
larvae. After 8-9 weeks of rearing in such conditions, all of the daughter colonies were sampled for
ants, which were used in chemical analyses.

2.3. CHC profile of F. sanguinea ants from the pure colonies

After a period of species-separation, it could not be ruled out that ants still possessed heterospecific
substances on their body surfaces, which had been acquired prior to separation (see Errard, 1994).
Therefore, we established eight additional colonies in the laboratory to reveal which CHCs could be
incorporated into the recognition signature of F. sanguinea without the influence of slaves. At the
beginning, these colonies consisted of 10 marked F. sanguinea adult workers provided with 200-560 F.
sanguinea worker pupae from the same stock colonies that were used in the separation experiment.
Within 4 weeks, the marked ants were removed. In this way, we obtained colonies where all
individuals were deprived of any contact with F. fusca slaves during their adulthood. These colonies
were kept in similar conditions as the daughter colonies in the species-separation experiment, except
that they were not provided with eggs. From four to nine weeks after their establishment, they were
sampled for the material used in the chemical analyses.

2.4. Chemical analyses

To extract CHCs, samples consisting of 5 dead ants were immersed for 10 minutes in 200 - 400 l of
hexane (depending on the ant body size). The extract was then evaporated and redissolved in 60 l of
hexane. Two samples per species were prepared for each field colony (all F. sanguinea colonies were
mixed species) and one sample per species for each laboratory colony. The hexane solution was
analyzed on the Agilent 6890 Gas Chromatograph fitted with Mass Selective Detector 5973 and
Autosampler 7683, with an electronic pressure control and a split/splitless injector. Separation was
performed on the ZB-5Msi (30 m 0.25 mm I.D.; 0.25 m film thickness) fused silica column.
Helium flow rate through the column was 1 mL/min. The injector worked in a splitless mode and its
temperature was set at 250 C. The EIMS spectra were obtained for ionization energy of 70 eV, at the
source temperature 230 C and quadrupole temperature 150 C. The MSD was set to scan 41600
a.m.u. Chromatograms were registered in a programmed regime of linear temperature increase from
50 C to 340 C at the rate of 3 C/min.

Unequivocal identification in GCMS analyses required both mass spectrum and retention index.
Therefore, a hexane solution of C20C40 n-alkanes was chromatographed under the conditions
described above. Based on retention time values, linear temperature programmed retention indices IT
were calculated for the ant CHC extracts using the equation proposed by van Den Dool and Kratz
(1963). The resulting values of retention indices and mass spectra were compared with those from a
database developed by one of the authors (L.S.), as well as from an on-line library (Stein 2015) and
6
literature data (Carlson and Brenner, 1988; Bonavita-Cougourdan et al., 1996; Katritzky et al., 2000,
Akino et al., 2004; Calderon-Fernandez, 2013; Trabalon et al., 1996). Mass spectra were interpreted by
comparing them with the ones included in the libraries of the GC/MS apparatus software. Moreover, in
the case of branched alkanes, characteristic ions were taken into consideration to determine number
and position of methyl groups.

2.5. Statistical methods

Peak areas in GC-MS chromatograms were standardized by calculating the proportion of the sample
total. For the statistical analysis, we selected only those peaks which were derived from hydrocarbons,
had 23 or more carbon atoms and satisfied at least one of two additional criteria: (1) were present in at
least 30% of the samples, (2) had a relative abundance in at least one sample at 2% or more. The
dataset obtained in this way was then subjected to the principal component analysis (PCA). Our
dataset did not meet the assumption of homogeneity of variance-covariance matrices, thus instead of
the discriminant analysis, we used the nonparametric method of k-nearest neighbor classification (k =
3) with cross-validation, using the variables obtained in the PCA. This method enabled us to verify
whether the assignment of our samples into various groups according to species and type of colony
was justified by the differences in the relative abundance of the analyzed CHCs.

In order to determine the effect of species-separation as well as colony species-purity on the CHC
profile, we compared samples from different categories of laboratory colonies based on their scores on
the first principal component (Fig. 2), which explained 37% of the total variance and differentiated
samples according to species. The mean values of scores were compared among treatments (colony
categories) with the use of repeated measures ANOVA, treating a set of five colonies (one per
treatment) derived from the same stock colony as an object. Then, pair-wise comparisons with the
Bonferroni correction of p-values were made to find significant differences between treatments. Scores
on the first principal component were also compared between samples from field colonies by fitting
the linear mixed model with the combination of species and colony type (free-living or parasitic),
defined as fixed factor, and the colony identity as a random factor to account for the repeated samples
of the same species from the same colony. Owing to the limited role of n-alkanes as recognition cues
(see Introduction), as well as the relatively low transferability between nestmates (Martin et al., 2008b,
van Zweden et al., 2010, Wodarczyk and Szczepaniak, 2014), we repeated the above analyses after
excluding this class of hydrocarbons from our dataset and recalculating the peak proportions in the
samples.

7
We were also interested to learn which peaks could be used to differentiate between the slave-maker
and its host species to highlight the effect of expected CHC exchange between both species in a
parasite colony. Therefore, for each peak, we calculated following discriminant index:

DIi = max[(xfus, i /xsang, i ), (xsang, i /xfus, i )]|PC1 i| if xfus, i and xsang, i 0,

where
xfus, i is the mean proportion of the peak i in F. fusca samples from free-living colonies,
xsang, i is the mean proportion of the peak i in F. sanguinea samples from pure colonies, and
PC1i is the loading of peak i on the first principal component.

We classified the peak to be distinctive for a given species if DI > 1, with the species assignment
dependent on which term fulfilled the condition defined by the max function. A peak which occurred
in one species and was absent in the other (xfus, i = 0 or xsang, i = 0) was by definition classified as
distinctive for this species.

To assess the variability among the CHC profiles of free-living and enslaved F. fusca nestmates, we
calculated the modified distance of Nei (Dronnet et al., 2006) for pairs of samples from the same field
colony. We were interested in disentangling the effect of the diversified origin of F. fusca slaves from
the possible CHC transfer from F. sanguinea ants. Thus, we excluded peaks classified as distinctive for
F. sanguinea from our calculations. We also excluded n-alkanes from our dataset, which otherwise
would generate additional sources of variation, not much relevant for nestmate recognition (see Martin
et al., 2008c, 2010). Based on the same peak set, we compared chemical distance among F. sanguinea
ants and among their slaves to see which of these species produce greater variation in a mixed colony.
Finally, we compared the distance from slaves to slave-makers between field and laboratory mixed
colonies.

Statistical computations were performed with the use of R (R Development Core Team, 2015) and
SPSS 21 software.

3. Results
3.1. Species-related differences in CHC profiles
We identified 74 peaks indicating hydrocarbons and their mixtures (Table S1 and S2; for sample
chromatograms see Fig. S1). In the case of F. sanguinea and some slave samples, one hydrocarbon
peak co-eluted with aldehyde (pentacosanal). Such a distribution of this compound among the samples
suggests that it was a stable component present on the body surface of F. sanguinea. However, given

8
that its role in nestmate recognition is uncertain, we corrected relevant peak areas after the separation
of aldehyde from the hydrocarbons with the use of the multivariate curve resolution alternating least
squares (MCR-ALS) technique included in the ALS packet for R.

Thirty-one peaks (42%) satisfied our conditions to be classified as discriminating between the two
species studied. Nine of them were derived from hydrocarbons, which were more abundant for F.
fusca. These were monomethyl- and dimethyl-branched isomers of tetracosane, pentacosane and
hexacosane (Table S1, Fig. 1). On the other hand, F. sanguinea could be distinguished from its host
species by the presence or abundance of heavier hydrocarbons with more than 26 atoms in the carbon
backbone. Some alkenes as well as some methyl-branched derivatives of tetracosane and pentacosane
have also proved to be distinctive for that species (Table S1, Fig. 1). Compounds distinctive for F.
fusca constituted on average over 12% of the total amount of CHCs in free-living F. fusca ants, while
the mean relative abundance for F. sanguinea from field colonies was less than 2% and less than 1%
for F. sanguinea from laboratory pure colonies (Table 1). Formica sanguinea ants have a higher
proportion of compounds distinctive for their own species in their CHC profile than F. fusca ants,
since in the former species, they constituted on average 11-20% of the total peak area in a sample
(Table 1). Correspondingly, slave-makers had most of the compounds which occurred in the host
species plus many characteristic for themselves. Samples representing different species were fairly
well segregated along the axis of the first principal component, which explained 37% of the total
variance (Fig. 2). A cross-validation of our nearest neighbor model confirmed the differences between
groups, since 104 out of 126 samples (82%) were classified correctly. Mistakes concerned mainly ants
from the mixed colonies which were classified as the other species (Table 2). Seven more samples
were assigned to the correct group when n-alkanes were excluded from the analysis.

3.2. The CHC profile of slaves from the field colonies

Results of the PCA show that slave samples from the field colonies are grouped in between the slave-
makers and free-living F. fusca along the axis of the first principal component. Thus, being a slave has
a clear effect on the F. fusca CHC profile. In particular, enslaved ants had compounds distinctive for
their parasite that contributed on average 11% to the total amount of hydrocarbons, whereas in the case
of free-living ants, they constituted less than 1% (Table 1). Correspondingly, scores on the first
principal component differed significantly between slaves and free-living F. fusca ants (Tukey post-
hoc pair-wise comparison in a linear mixed model, z-value = 7.86, p < 10-10), as well as between slaves
and slave-makers (z-value =11.05, p < 10-10 ) (Fig. 3). The model of the dataset with excluded n-
alkanes qualitatively yielded the same results.

9
The comparison of Nei distances revealed that the variation of CHC profiles is higher among slave
nestmates than among free-living F. fusca (Mann-Whitney test, Z = 4.583, p < 10-5 ; Fig. 5). Moreover,
within-colony odor variation proved to be higher among slaves compared to their slave-maker
nestmates (Wilcoxon signed-rank test, Z = -2.667, p = 0.008). Interestingly, species-related odor
differences were significantly lower in laboratory mixed colonies than in the ones from the field (Z = -
2.147, p = 0.033). This effect can be explained by the smaller nesting space available for ants under
our laboratory conditions, which restricted the spatial segregation of heterospecific ants associated
with the tendency to division of labor between both species (Darwin, 1859; Kharkiv, 1997).

3.3. The effect of the removal of heterospecific nestmates

Being reared in homospecific colonies clearly affected the CHC profiles of F. fusca ex-slaves, which
became similar to conspecific ants from free-living colonies, as indicated by the change in the relative
amount of compounds distinctive for slave-maker and for its own species (Table 1). Correspondingly,
samples from species-separated F. fusca had significantly higher scores on the first discriminant
function compared to ants reared together with slave-makers (Fig. 4). On the other hand, the influence
of heterospecific nestmates on the CHC profile of F. sanguinea ants was smaller, as can be concluded
from the lack of significant differences in scores between F. sanguinea ants from species-separated
and mixed colonies (Fig. 4). The exclusion of n-alkanes from the dataset led to similar results, except
that there was a significant difference between F. fusca from mixed colonies and species-separated F.
sanguinea (p = 0.033). Formica sanguinea ants deprived of contact with slaves showed only a small
difference in the relative amount of hydrocarbons identified to be distinctive for either species,
compared to conspecific ants from mixed colonies (Table 1). The nearest neighbor model model
yielded three out of eight samples representing species-separated F. sanguinea misclassified as F.
sanguinea from mixed colony (Table 2) highlighting a relatively weak effect of the lack of slaves on
the CHC profile of this species.

3.4. Pure CHC blend of F. sanguinea ants

Formica sanguinea ants that had been deprived of any contact with slaves as adults were similar in
terms of the CHC profile to slave-maker ants from other groups (Fig. 1, Table 1). The comparison of
scores on the first principal component failed to detect a significant difference between F. sanguinea
ants from pure, species-separated and mixed colonies (Fig. 4). Moreover, F. sanguinea seemed to be
able to synthesize most of the lighter hydrocarbons distinguishing the F. fusca samples (Fig. 1, Table
S1).

4. Discussion

10
4.1. Chemical discrepancy between the facultative slave-maker and its main host

Despite the possible advantages provided by chemical mimicry (see Introduction), our results show
that the CHC profiles of the facultative slave-making ants F. sanguinea are distinct from those of free-
living F. fusca. The analyses of CHC extracts from ants sampled from originally pure colonies, as well
as from species-segregated groups of workers, strongly suggest that there is a difference between the
inherent pattern of the CHCs production between both species, given the predominant role of de novo
biosynthesis in the formation of the hydrocarbon layer on the insects' cuticles (Blomquist, 2010). In
particular, we detected alkenes and methyl-branched alkanes with more than 29 carbon atoms in the
backbone on the slave-maker's cuticle, whereas these compounds were almost never found in free-
living F. fusca. These results are consistent with the conclusions from the analysis of the similarities of
CHC profiles among the species belonging to the Formica genus (Martin et al., 2008a). The
comparison revealed that F. sanguinea ants are at a lesser chemical distance from members of the
Formica subgenus than from F. fusca, thus highlighting the effect of phylogenetic relations rather than
slave-making on the CHC profile in of this species. It is thus possible that too few generations of the
social parasite have passed yet for more sophisticated forms of chemical deception to have evolved. In
the incipient social parasitic ant species Arcromyrmex insinuator invading the colonies of leaf-cutting
ant A. echinatior, differences between both species have been detected in proportions among CHCs.
This parasitic species adopted a strategy of chemical insignificance rather than mimicry of the host
recognition signature (Lambardi et al., 2007; Nehring et al., 2015).

It is possible that selective pressure on F. sanguinea, whose colonies are able to survive without a
slave workforce, has been too weak to trigger evolutionary changes toward such a degree of mimicry
of the host's odor, which can be observed in obligate parasites (Guillem et al., 2014). The capacity of
ants for odor learning within a short period after eclosion and subsequent template updates enable the
co-existence of heterospecific ants within a mixed colony, even when some species-specific odor
differences in the recognition signatures between nestmates are present (Liu et al., 2003; Errard et al.,
2006; Wodarczyk and Szczepaniak, 2014), although the effect of such differences on colony fitness
remains unknown. On the other hand, F. fusca ants are under pressure from many socially parasitic
species at once, which has led to the high inter-colonial variation of CHC profiles (Martin et al., 2011;
Lorenzi, 2014). This counter-adaptation could make it difficult for F. sanguinea to track host odor
during co-evolution.

4.2. Effect of slave-makers on the slaves' chemical signature

11
We found that enslaved F. fusca ants clearly differ from free-living conspecifics in terms of CHC
profiles. The hydrocarbon blend on their body surface includes compounds from a heavier fraction,
characteristic for the slave-maker. This difference in recognition labels between free-living and
enslaved F. fusca is most probably caused by the acquisition of CHCs from slave-maker ants during
social interactions. It is noteworthy that the outcome of the species-separation experiment failed to
detect a reciprocal effect of F. fusca ants on the social parasite's odor. These conclusions are consistent
with the results of behavioral experiments, which demonstrated that free-living F. fusca ants are
treated by ants from F. sanguinea colonies with higher aggression than enslaved individuals from alien
colonies (Wodarczyk 2016). Behavioral studies were also unable to find evidence that slaves
influence the recognition signature of their social parasite. In contrast, the lack of contact with slave-
makers resulted in F. fusca ex-slaves eliciting augmented aggression from former nestmates from the
F. sanguinea colony (Wodarczyk 2016). Results of the chemical analyses also revealed that the
recognition signature of slaves had higher within-colony variation compared to slave-makers and free-
living conspecific ants. This effect can be explained by the greater genetic diversity among slaves
which usually originate from many, perhaps unrelated colonies (Mori et al., 2000, Torres and Tsutsui
2016).

4.3. The chemical strategy of F. sanguinea ants

Contrary to obligatory slave-making species, Formica sanguinea reduces species-related odor

differences within the colony not by mimicry, but instead by affecting slaves' odor by partly masking it
with its own recognition cues. Masking host odor might be a sufficient strategy at an incipient or
intermediate stage in the evolution of social parasitism. Females of many paper wasp species of the
genus Polistes usurp alien conspecific nests as an alternative way of founding a colony (Cervo and
Dani, 1996). In P. biglumis, this behavior is associated with an increased proportion of methyl-
branched hydrocarbons on the usurper's cuticle, which are used to mark the host nest (Lorenzi et al.,
2011). The scent of the nest is memorized by newly-emerged paper wasps and treated as the colony
recognition signature (Gamboa et al., 1986). Thus, by depositing its own hydrocarbons on the nest
surface, the usurper female increases her chances of being accepted by the host workers that will soon
emerge in the colony headed by an unrelated female (Lorenzi et al., 2011). It has been suggested that a
similar strategy is utilized by the facultative social parasite P. nimphus when usurping colonies of P.
dominulus (Lorenzi et al., 2007).

Our results suggest that by affecting host odor F. sanguinea ants reduce the variation of recognition
signatures among slaves originating from unrelated colonies in this way. The reduction of within-

12
colony odor variation might be crucial in maintaining colony cohesion, e.g. by preventing a decline in
the discriminatory abilities of colony members (Wodarczyk 2016). It is possible that the influence of
F. sanguinea ants on slaves' CHC profiles emerges simply as a consequence of social interactions,
when CHCs are exchanged between individuals (see Introduction). Asymmetric CHC transfer between
heterospecific ants with a greater influence of F. sanguinea on the odor of slaves is likely to be an
effect of the slave-maker to slave ratio within the colony, coupled with the larger body size of F.
sanguinea and consequently, the higher contribution of slave-makers to overall CHC production in the
colony. Moreover, the division of labor between the species, with slaves undertaking mainly intranidal
tasks (Kharkiv, 1997; Mori et al., 2000), might also promote an unequal CHC exchange, since F.
sanguinea ants returning to the nest with liquid food might subsequently distribute it among slaves
during trophallaxis, which is associated with CHC transfer (see Introduction). Further investigations
should show whether F. sanguinea ants have special adaptations to mask or dilute slaves' recognition
cues, such as increased production of CHCs and frequent trophallaxis with slaves.

4.4. Possible limitations on the evolutionary transition to chemical mimicry

The chemical strategy adopted by F. sanguinea ants might be associated with benefits that would not
be available if mimicry was employed. It might be advantageous for F. sanguinea ants to possess a
species-specific chemical signature and not be misidentified as the host species, since F. sanguinea, as
opposed to F. fusca, is an aggressive territorial species occupying the top level in the interspecific
hierarchy (Wilson, 1971; Vepslinen and Pisarski, 1982; Punttila et al., 1996). Moreover, chemical
mimicry could make F. sanguinea colonies exposed to the raids organized by alien slave-maker
colonies. Raids against conspecific colonies have been recorded in the obligate social parasites of P.
rufescens and P. breviceps (Topoff et al., 1984; Mori et al., 1993; Grasso et al., 1994, Bono et al.,
2006). Thus, at the stage of the evolution of slave-making represented by F. sanguinea, selective
pressure on the chemical mimicry of the host odor might be too weak to overcome the conflicting
pressure of possessing a distinct species-specific recognition label. Moreover, the lower slave
proportion in F. sanguinea colonies compared to obligate slave-making species (see Czechowski and
Godziska, 2014) facilitates the strategy of promoting social parasite's own recognition cues in slaves.
Hence, the ecological benefits associated with odor distinctiveness can be expected to be negatively
correlated with the degree of dependence on the slave workforce. Good candidates for comparative
studies enabling to test this hypothesis to be tested are other species from the F. sanguinea group (see
Savolainen and Deslippe, 1996, 2001).

4.5. Degree of the odor discrepancy

13
The outcome of experiments using pure F. sanguinea colonies suggests that this species is able to
produce almost all of the components of the host chemical signature. Thus, our results do not rule out
that some degree of chemical resemblance has indeed evolved in F. sanguinea. It is reasonable to
argue that odor dissimilarity between slave-maker and host should not be too high for the odor
domination strategy to operate successfully. Otherwise, species-related odor differences would not be
reduced to a level guaranteeing colony integrity. Nevertheless, chemical adjustment to the host, even if
it has played some role during the evolution of the CHC profile of F. sanguinea ants, is not as
advanced as in the obligate slave-makers of the Polyergus genus (Yamaoka, 1990; Bonavita-
Courgourdan et al., 1996, 1997; DEttorre et al., 2002).

5. Acknowledgements

We are grateful to anonymous reviewers for their critical remarks which helped us to improve this
paper. This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.

6. References

Akino T, Yamamura K, Wakamura S, and Yamaoka R (2004) Direct behavioral evidence for
hydrocarbons as nestmate recognition cues in Formica japonica (Hymenoptera: Formicidae). Appl
Entomol Zool 39:381387

Alloway TM (1990) Slave-species ant colonies recognize slavemakers as enemies. Anim Behav
39(6):12181220

Bagnres AG and Lorenzi MC (2010) Chemical deception/mimicry using cuticular hydrocarbons. In:
Blomquist, G.J., Bagnres, A.G. (eds) Insect Hydrocarbons: Biology, Biochemistry and Chemical
Ecology, Cambridge University Press, pp. 282-324.

Bauer S, Bhm M, Witte V, Foitzik S (2010) An ant social parasite in-between two chemical disparate
host species. Evol Ecol 24(2):317-332

Blomquist GJ (2010) Biosynthesis of cuticular hydrocarbons. In: Blomquist, G.J., Bagnres, A.G. (eds)
Insect Hydrocarbons: Biology, Biochemistry and Chemical Ecology, Cambridge University Press, pp
3552

Bonavita-Cougourdan A, Rivire G, Provost E, Bagnres A-G, Roux M, Dusticier G, Clment JL

(1996) Selective adaptation of the cuticular hydrocarbon profiles of the slave-making ants Polyergus
rufescens Latr. and their Formica rufibarbis Fab. and F. cunicularia Latr. slaves. Comp Biochem
Physiol 113:313-329

Bonavita-Courgourdan A, Bagnres A-N, Provost E, Dusticier G, Clment JL (1997) Plasticity of the

cuticular hydrocarbon profile of the slave-making ant Polyergus rufescens depending on the social
environment. Comp Biochem and Physiol 116B(3):287-302
14
Bono JM, Gordon ER, Antolin MF, Herbers JM (2006) Raiding activity of an obligate (Polyergus
breviceps) and two facultative (Formica puberula and F. gynocrates) slave-making ants J Insect
Behav. 19:429446

Boulay RM, Hefetz A, Soroker V, Lenoir A (2000) Camponotus fellah colony integration: worker
individuality necessitates frequent hydrocarbons exchanges. Anim Behav 59:1127-113

Brandt M, Heinze J, Schmitt T, Foitzik S (2005) A chemical level in the coevolutionary arms race
between an ant social parasite and its hosts. J Evol Biol 18:576-586

Calderon-Fernandez GM, Juarez MP (2013) The cuticular hydrocarbons of the Triatoma sordida
species subcomplex (Hemiptera: Reduviidae). Mem Inst Oswaldo Cruz 108:778784

Carlson DA, Brenner RJ (1988) Hydrocarbon-Based Discrimination of Three North American

Blattella Cockroach Species (Orthoptera: Blattellidae) Using Gas Chromatography. Ann Entomol Soc
Am 81:711723

Cervo R, Dani FR. 1996. Social parasitism and its evolution in Polistes. In: Turillazzi S, West

Eberhard MJ, eds. Natural history and the evolution of paper wasps. Oxford University Press, 98112.

Czechowski W (1996) Colonies of hybrids and mixed colonies: interspecific nest takeover in wood
ants (Hymenoptera, Formicidae). Memorabilia Zoologica 50:1116

Czechowski W, Godziska EJ (2014) Enslaved ants: not as helpless as they were thought to be.
Insectes Soc 62(1):9-22.

Dahbi A, Hefetz A, Cerd X, Lenoir A (1999) Trophallaxis mediates uniformity of colonial odor in
Cataglyphis iberica ants (Hymenoptera, Formicidae). J Insect Behav 12:559-567

Dani, FR, Jones GR, Corsi S, Beard R, Pradella D, Turillazzi S (2005). Nestmate recognition cues in
the honey bee: differential importance of cuticular alkanes and alkenes. Chemical Senses:30 477-489.

Dani, F. R., Jones, G., Destri, S., Spencer, S., & Turillazzi, S. (2001). Deciphering the recognition
signature within the cuticular chemical profile of paper wasps. Animal Behaviour, 62: 165-171.

Darwin C (1859) The origin of species by means of natural selection, or preservation of favoured races
in the struggle for life. Murray, London

DEttorre P, Errard C. 1998. Chemical disguise during colony founding in the dulotic ant Polyergus
rufescens Latr. (Hymenoptera, Formicidae). Insect Soc. Life 2:7177

DEttorre P, Errard C, Ibarra F, Francke W, Hefetz A (2000) Sneak in or repel your enemy: Dufours
gland repellent as a strategy for successful usurpation in the slave-maker Polyergus rufescens .
Chemoecology 10:135 142

DEttorre P, Heinze J. 2001. Sociobiology of slave-making ants. Acta ethol 3: 67-82.

DEttorre P, Mondy N, Lenoir A, Errard C (2002) Blending in with the crowd: social parasites
integrate into their host colonies using a flexible chemical signature. Proc R Soc Lond B 269:1911

15
1918

Dronnet S, Lohou C, Christides JP, Bagnres AG (2006) Cuticular hydrocarbon composition reflects
genetic relationship among colonies of the introduced termite Reticulitermes santonensis Feytaud. J
Chem Ecol 32:10271042

Errard C (1994) The role of long-term memory in nestmate recognition in ants. Anim Behav 48(2):
263-271

Errard C, Jaisson P (1991) Les premires tapes de la reconnaissance interspcifique chez les fourmis
Manica rubida et Formica selysi (Hymenoptera, Formicidae) leves en colonies mixtes. C R Acad
Sci Paris 313:73-80

Errard C, Ruano F, Richard FJ, Lenoir A, Tinaut A, Hefetz A (2006) Co-evolution-driven cuticular
hydrocarbon variation between the slave-making ant Rossomyrmex minuchae and its host Proformica
longiseta (Hymenoptera: Formicidae). Chemoecology 16:235-240

Franks N, Blum M, Smith R-K, Allies AB (1990) Behavior and chemical disguise of cukoo ant
Leptothorax kutteri in relation to its host Leptothorax acervorum. J Chem Ecol 16(5):1431-1444

Gamboa GJ, Reeve HK, Pfennig DW (1986) The evolution and ontogeny of nestmate recognition in
social wasps. Annu Rev Entomol 31: 431-454

Gibbs AG, Rajpurohit S (2010) Cuticular and water balance. Insect hydrocarbons: biology,
biochemistry, and chemical ecology 6:100121

Grasso DA, Mori A, DEttorre P, Le Moli F (1994) Intraspercific raid and territoriality in Polyergus
rufescens Latr. (Hymenoptera Formicidae). Ethol Ecol Evol, Special Issue 3:8187

Guillem RM, Drijfhout F, Martin SJ (2014) Chemical deception among ant social parasites. Curr Zool
60(1):62-75

Howard RW, Akre RD (1995) Propaganda, crypsis and slavemaking. In: Card R. T., Bell W. J. (eds)
Chemical ecology in insects II. Chapman & Hall, London, pp. 364429

Johnson CA, Vander Meer RK, Lavine B (2001) Changes in the cuticular hydrocarbon profile
of the slave-maker ant queen, Polyergus breviceps Emery, after killing a Formica host queen
(Hymenoptera: Formicidae). J Chem Ecol 27: 17871804.
Kaib M, Heinze J, Ortius D (1993) Cuticular hydrocarbon profiles in the slave-making ant
Harpogoxenus sublaevis and its hosts. Naturwissenschaften 80:281285

Katritzky AR, Chen K, Maran U, Carlson DA (2000) QSPR Correlation and Predictions of GC
Retention Indexes for Methyl-Branched Hydrocarbons Produced by Insects. Anal Chem 72:101109

Kharkiv VA (1997). Competition as a mechanism of distributing functions in colonies of Formica

sanguinea (Hymenoptera, Formicidae). Zool Zh 76:444447

Kleeberg I, Foitzik S (2016) The placid slavemaker: avoiding detection and conflict as an alternative,
peaceful raiding strategy. Behav Ecol Sociobiol 70:2739.

Krasnec MO, Breed MD (2013) Colony-specific cuticular hydrocarbon profile in Formica

16
argentea ants. J Chem Ecol 39(1):59-66

Lambardi, D., Dani, F. R., Turillazzi, S., & Boomsma, J. J. (2007). Chemical mimicry in an incipient
leaf-cutting ant social parasite. Behavioral Ecology and Sociobiology, 61, 843-851.

Le Moli F, Grasso DA, DEttorre P, Mori A (1993) Intraspecific slavery in Polyergus rufescens Latr.
(Hymenoptera, Formicidae): field and laboratory observations. Insectes soc. 40: 433437.

Lenoir A, Cuisset D, Hefetz A. (2001a) Effects of social isolation on hydrocarbon pattern and nestmate
recognition in the ant Aphaenogaster senilis (Hymenoptera, Formicidae). Insectes Soc 48:101-109

Lenoir, A., D'Ettorre, P., Errard, C., Hefetz, A. (2001b). Chemical ecology and social parasitism in
ants. Ann. Rev. Entomol. 46: 573-59.

Liu Z, Bagnres A-N, Yamane S, Wang Q, Kojima J (2003) Cuticular hydrocarbons in workers of the
slave-making ant Polyergus samurai and its slave, Formica japonica (Hymenoptera: Formicidae).
Entomol Sci 6:125-133

Lorenzi MC (2003) Social wasp parasites affect the nestmate recognition abilities of their
hosts (Polistes atrimandibularis and P. biglumis, Hymenoptera,Vespidae). Insectes Soc 50: 82-87.

Lorenzi MC. 2006. The result of an arms race: the chemical strategies of Polistes social parasites.
Annales Zoologici Fennici 43: 550563.

Lorenzi M.C., Caldi M., & Cervo R. 2007. The chemical strategies used by Polistes nymphus social
wasp usurpers (Hymenoptera Vespidae). Biological Journal of the Linnean Society 91: 505-512.

Lorenzi M. C., R. Cervo, A.-G. Bagnres. 2011. Facultative social parasites mark host nests with
branched hydrocarbons. Animal Behaviour, 82: 1149-1157.

Lorenzi MC, Azzani L, Bagnres A-G (2014) Evolutionary consequences of deception: complexity
and informational content of colony signature are favored by social parasitism. Curr Zool 60(1):137-
148

Martin SJ, Helanter H, Drijfhout FP (2008a) Evolution of species-specific cuticular hydrocarbon

patterns in Formica ants. Biol J Linn Soc 95:131140

Martin, S. J., Helanter, H., Falko, P., Drijfhout, F. P. (2008b). Colony-specific hydrocarbons identify
nest mates in two species of Formica ant. J Chem Ecol 34: 1072-1080.

Martin SJ, Carruthers JM, Williams PH, Drijfhout FP (2010) Host Specific Social Parasites (Psithyrus)
Indicate Chemical Recognition System in Bumblebees. J Chem Ecol 36(8):855-63.

Martin SJ, Helanter H, Drijfhout FP (2011) Is parasite pressure a driver of chemical cue diversity in
ants? Proc R Soc B 278:496-503

Martin SJ, Vitikainen E, Helanter H, Drijfhout FP (2008c) Chemical basis of nest-mate

discrimination in the ant Formica exsecta. Proc R Soc B 275(1640):12711278

Morel L (1983) Relation entre comportement agressif et privation sociale prcoce chez les jeunes
immatures de la Fourmi Camponotus vagus Scop. (Hymenoptera: Formicidae). C R Acad Sci Paris
296:449-452

17
Mori A, Grasso DA, Le Moli F (2000) Raiding and foraging behaviour of the blood-red ant, Formica
sanguinea Latr. (Hymenoptera, Formicidae). J Insect Behav 13:421-438

Mori A, Grasso DA, Visicchio R, Le Moli F (2001) Comparison of reproductive strategies and raiding
behaviour in facultative and obligatory slave-making ants: the case of Formica sanguinea and
Polyergus rufescens. Insectes Soc 48:302-314.

Ozaki M, Hefetz A (2014) Neural mechanisms and information processing in recognition systems.
Insects 5:722-741

Pfennig DW, Gamboa GJ, Reeve HK, Shellman-Reeve J, Ferguson ID (1983) The mechanism of
nestmate discrimination in social wasps (Polistes, Hymenoptera: Vespidae). Behav Ecol Sociobiol
13:299-305

Punttila P., Haila Y., Pajunen T., Tukia H. (1996) Ant communities in taiga clearcuts: habitat effects
and species interactions. Ecography 19: 16-28

R Development Core Team (2015) R: a language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org. Accessed 1 February
2015

Ruano F, Devers S, Sanllorente O, Errard C, Tinaut A, Lenoir A (2011) A geographic mosaic of

coevolution in a slave-making hostparasite system. J Evol Biol 24:1071-1079

Savolainen R, Deslippe RJ (1996) Facultative and obligate slavery in formicine ants: frequency of
slavery, and proportion and size of slaves. Biol J Linn Soc 57:4758

Savolainen R, Deslippe RJ (2001) Facultative and obligate slave making in Formica ants
Naturwissenschaften 88:347350

Savolainen R, Vepslinen K (1989) A Competition Hierarchy among Boreal Ants: Impact on resource
partitioning and community structure. Oikos 56(1):3-16

Sherman PW, Holmes WG (1985) Kin recognition: Issues and evidence. Fortschr Zool 31:437-460

Soroker V, Fresneau D, Hefetz A (1998) Formation of colony odor in ponerine ant Pachycondyla
apicalis. J Chem Ecol 24:1077-1090

Soroker VC, Vienne A, Hefetz A (1995) Hydrocarbon dynamics within and between nestmates in
Cataglyphis niger (Hymenoptera, Formicidae). J Chem Ecol 21:365-378

Stein SE (2015) Retention Indices. In: Linstrom PJ and Mallard WG (Eds) NIST Chemistry WebBook,
NIST Standard Reference Database Number 69, National Institute of Standards and Technology,
Gaithersburg MD, 20899, http://webbook.nist.gov. Accessed 15 July 2015

Stuart RJ (1991) Nestmate recognition in leptothoracine ants: testing for effects of queens number,
colony size and species of intruder. Anim Behav 42:277-284

Torres CW, Tsutsui ND (2016) The effect of social parasitism by Polyergus breviceps on the nestmate
recognition system of its host, Formica altipetens. Plose One 11(2): e0147498.
doi:10.1371/journal.pone.0147498

Trabalon M, Bagnres AG, Hartmann N, Vallet AM (1996) Changes in cuticular compounds

18
composition during the gregarious period and after dispersal of the young in Tegenaria atrica
(Araneae, Agelenidae). Insect Biochem Mol Biol 26:7784

Topoff, H., B. LaMon, L. Goodloe and M. Goldstein, 1984. Social and orientation behavior of
Polyergus brevieeps during slave-making raids. Behav. Ecol. Sociobiol. 15: 273- 279.

Vepslinen K, Pisarski B (1982) Assembly of island ant communities. Ann Zool Fenn 19(4):327335

Visicchio R, Sledge MF, Mori A, Grasso DA, Le Moli F, et al. 2000. Dufours gland contents of
queens of the slavemaking ant Polyergus rufescens and its host species Formica cunicularia. Ethol.
Ecol. Evol. 12:6773

van Den Dool H, Kratz PD (1963) A generalization of the retention index system including linear
temperature programmed gas-liquid partition chromatography. J Chromatogr A 11:46371

van Zweden JS, Brask JB, Christensen JH, Boomsma JJ, Linksvayer TA, dEttorre P. 2010 Blending of
heritable recognition cues among ant nestmates creates distinct colony gestalt odours but prevents
within-colony nepotism. J. Evol. Biol. 23, 1498 1508.

van Zweden JS, dEttorre P (2010) Nestmate recognition in social insects and the role of
hydrocarbons. In: Blomquist, G.J., Bagnres, A.G. (eds) Insect Hydrocarbons: Biology, Biochemistry
and Chemical Ecology, Cambridge University Press, pp 222234

Wheeler WM (1910) Ants: their structure, development and behavior. Columbia University Press,
New York, 663 pp

Wilson EO (1971) The Insect Societies. Cambridge, MA: Belknap Press

Wodarczyk T (2016) Discriminatory abilities of facultative slave-making ants and their slaves.
Insectes Sociaux 63: 507-517

Wodarczyk T, Szczepaniak L (2014) Incomplete homogenization of chemical recognition labels

between Formica sanguinea and Formica rufa ants living in a mixed colony. J Insect Sci 14:214

Yamaoka R (1990) Chemical approach to understanding interactions among organisms. Physiol Ecol
Japan 27(special no.):31-52

Fig. 1 Mean relative abundances with standard deviations of ant cuticular hydrocarbons in different
groups of samples. Numeration of the bars corresponds to the peaks listed in Table S1.
Fig. 2 Results of the principal component analysis based on data on the relative abundances of
hydrocarbons extracted from the ants' body surface.

19
Fig. 3 Comparison of scores on the first principal component between groups of samples collected
from field colonies. Each bar covers the range of observed values between the first and third quartile
with the inner line indicating the median. Whiskers depict minimal and maximal values, excluding
outliers, which are marked separately. Significant differences were obtained in the post-hoc test
following the linear mixed regression.
Fig. 4 Comparison of scores on the first principal component between groups of samples collected
from laboratory colonies. Each bar covers the range of observed values between the first and third
quartile with the inner line indicating the median. Whiskers depict minimal and maximal values,
excluding outliers, which are marked separately. Different letters indicate significant differences
obtained in the post-hoc test following repeated measures ANOVA.
Fig. 5 Comparison of distances of Nei between F. fusca nestmates from the free-living or parasite
colony based on relative abundances of CHCs. Each bar covers the range of observed values between
the first and third quartile with the inner line indicating the median. Whiskers depict minimal and
maximal values, excluding outliers, which are marked separately.

20
Table 1. Mean relative amounts standard deviations of hydrocarbons allowing to discriminate
between the species and characteristic for F. sanguinea or F. fusca in different groups of samples
Group of samples Mean relative abundance of
compounds characteristic for
F. fusca F. sanguinea
Free-living F. fusca 12.346.63 0.760.59
colonies
Field

Ensalved F. fusca 4.861.49 11.225.95

F. sanguinea 1.981.31 21.476.80
Species-separated F. fusca 9.962.45 1.800.81
Laboratory

F. fusca, mixed colony 2.471.17 11.304.43

colonies

F. sanguinea, mixed colony 2.461.98 15.234.03

Species-separated F. sanguinea 0.870.42 17.853.35
F. sanguinea, pure colony 0.760.77 20.463.83

21
Table 2. Results of the k-nearest neighbour classification with the cross-validation using different
groups of samples. Numbers of correctly classified samples are given in the diagonal

Group description Predicted True group

group
1 2 3 4 5 6 7 8
Free-living F. fusca 1 42 1 0 0 0 0 0 0
colonies
Field

Ensalved F. fusca 2 0 16 2 0 0 0 0 0
F. sanguinea 3 0 5 19 0 1 0 0 0

Species-separated F. fusca 4 0 0 0 8 0 0 0 0
Laboratory

F. fusca, mixed colony 5 0 0 1 0 6 2 0 0

colonies

F. sanguinea, mixed colony 6 0 0 0 0 0 2 3 0

Species-separated F. sanguinea 7 0 0 0 0 0 3 4 1
F. sanguinea, pure colony 8 0 0 0 0 1 1 1 7

22
23
24
25
26
27
Highlights:
F. sanguinea ants have recognition signature distinct from host ants
CHC profile of enslaved F. fusca differs from that of their free-living conspecifics
F. sanguinea exert an impact on the F. fusca CHC profile
F. sanguinea is able to synthesize or acquire independently most of the host's CHCs

28
29