Content uploaded by Chantal Poteaux
Author content
All content in this area was uploaded by Chantal Poteaux on Jan 07, 2018
Content may be subject to copyright.
Socio-genetic structure and mating system of a wild
boar population
C. Poteaux
1
, E. Baubet
2
, G. Kaminski
1,2
, S. Brandt
2
, F. S. Dobson
3
& C. Baudoin
1
1 CNRS UMR 7153, Laboratoire d’Ethologie, Expe
´rimentale et Compare
´e, Universite
´Paris 13, Villetaneuse, France
2 Office National de la Chasse et de la Faune Sauvage, CNERA – CS, station d’e
´tude sanglier, Cha
ˆteauvillain, France
3 Centre d’Ecologie Fonctionnelle et Evolutive, CNRS, Montpellier Cedex 5, France; & Department of Biological Sciences, Auburn University,
Auburn, AL, USA
Keywords
matrilines; polygyny; social organization;
genetic structure; wild boar.
Correspondence
Poteaux Chantal, CNRS UMR 7153,
Laboratoire d’Ethologie Expe
´rimentale et
Compare
´e, Universite
´Paris 13, 99 Avenue
J. -B. Cle
´ment, 93430 Villetaneuse, France.
Tel: +33 0 1 49 40 32 29; Fax: +33 0 1 49
40 39 75
Email: poteaux@leec.univ-paris13.fr
Editor: Jean-Nicolas Volff
Received 1 October 2008; revised 14
December 2008; accepted 28 December 2008
doi:10.1111/j.1469-7998.2009.00553.x
Abstract
Wild boars Sus scrofa have a social organization based on female groups that can
include several generations of adults and offspring, and are thus likely matrilineal.
However, little is known about the degree of relatedness between animals living in
such groups or occupying the same core area of spatial activity. Also, polygynous
male mating combined with matrilineal female groups can have strong influences
on the genetic structure of populations. We used microsatellite genotyping
combined with behavioral data to investigate the fine-scale population genetic
structure and the mating system of wild boars in a multi-year study at Chaˆ teau-
villain-Arc-en-Barrois (France). According to spatial genetic autocorrelation,
females in spatial proximity were significantly inter-related. However, we found
that numerous males contributed to the next generation, even within the same
social group. Based on our genetic data and behavioral observations, wild boars in
this population appear to have a low level of polygyny associated with matrilineal
female groups, and infrequent multiple paternity. Mortality due to hunting may
facilitate the breakup of what historically has been a more predominantly
polygynous mating system, and likely accelerates the turnover of adults within
the matrilineal groups.
Introduction
The evolution of mating systems is an important topic
in behavioral ecology (e.g. Emlen & Oring, 1977; Clutton-
Brock, 1989). In mammalian species, polygynous or poly-
gynandrous mating systems are common (Greenwood,
1980; Dobson, 1982; Komers & Brotherton, 1997). These
mating systems are often characterized by matrilineal asso-
ciations of philopatric females, and they also can have
substantial effects on the genetic properties of populations
(e.g. Sugg et al., 1996). In particular, studies of paternity
suggest that multiple mating by both sexes may be common
among mammals that produce more than one offspring at a
time (e.g. Lacey, Wieczorek & Tucker, 1997; Hoogland,
1998; Lane et al., 2007), and the different contribution of
fathers to the female offspring might have substantial
influences on sexual selection and on the genetic structure
of demes (reviewed by Parker & Waite, 1997; Dobson &
Zinner, 2003). Genetic influences of multiple mating likely
depend on both the number and pattern of patrilines, as well
as the number of offspring produced at each reproduction
and their pattern of survival.
Swine species reflect the mammalian variety of mating
systems, ranging from monogamous pairs to large female
groups with solitary males (Martys, 1991). Wild boars Sus
scrofa form female family groups, from 6 to 30 individuals,
as documented by both direct observations and mark–
recapture studies in semi-natural areas (Mauget et al., 1984;
Boitani et al., 1994). However, very little is known about the
degree of relatedness between animals living in such groups
or occupying the same core range area. Female family groups
might be based on kinship, in which case the conditions for
cooperation through kin selection and the genetic basis
for structured demes might be present (Chesser, 1991a).
Patterns of paternity are also unknown, and these might
also influence kinship and genetic structure (Chesser, 1991b;
Sugg & Chesser, 1994). The purpose of our study was to
examine relatedness structure in social groups of wild boars,
to discern their potential for kin-selected cooperation and to
define the genetic structure caused by their mating patterns
and social groups.
Socio-genetic structure depends thus on both social and
mating relationships among groups, and on inter-group
spatial organization. First, in wild boars, drastic changes in
group structure depend on dispersal pattern, which is
essentially male sex biased as in other mammals (Green-
wood, 1980; Dobson, 1982), and on fusion–fission events
among female groups. Another factor influencing group
Journal of Zoology
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London 1
Journal of Zoology. Print ISSN 0952-8369
structure is hunting impact (Calenge et al., 2002; Sodeikat &
Pohlmeyer, 2003). In France, wild boars remain an impor-
tant hunted species and are managed as a game animal.
Because it reaches high densities throughout its range, wild
boars are also considered as a pest, due to their impact on
both agricultural and natural ecosystems. Consequently,
wild boar populations are controlled to limit such depreda-
tions. Secondly, even though wild boars are able to travel up
to 10 km per night (radio telemetry studies, Goulding, 1998),
social groups tend to be faithful to a resting place (Boitani
et al., 1994). Moreover, daughters tend to stay with their
mother rather than to constitute a new group (Kaminski
et al., 2005). We expected a high relatedness among group
members and local differentiation between groups of differ-
ent matrilineal origin.
In order to describe the mating system in the studied
population, we estimated the number of males contributing
to the next generation. The mating system in wild boars is
hypothesized to be polygynous based on (1) the spatial
distribution of individuals (Brandt, Vassant & Jullien,
1998); (2) the increase in male aggressiveness during the
rutting season (Beuerle, 1975); (3) male-biased sexual size
dimorphism. According to the polygyny assumption, we
predict that a male would monopolize all females in a group.
As females often belong to different generations, all off-
spring are expected to be full or half-sibs, increasing the
relatedness level between group members. In addition, wild
boars exhibit high reproductive capacity (5.5 0.1 pig-
lets per female in the studied population, Servanty et al.,
2007) compared with other ungulates of similar body size
(Carranza, 1996), which facilitates opportunities for multi-
ple paternity. The hypothesis of polygyny has been chal-
lenged in favor of polygynandry in other mammal species
(snowshoe hare: Burton, 2002; pronghorn antelope: Car-
ling, Wiseman & Byersa, 2003).
Wild boars are a good biological model for the study of a
polygynous social system in a large mammalian species.
However, these nocturnal animals are difficult to observe in
the field, not easy to follow in brush or forest, and stay in
dense cover during daytime (Boitani et al., 1994). Studies of
genealogical relationships in female groups have been based
on small samples of tagged animals, have not applied genetic
analyses (Mauget et al., 1984; Boitani et al., 1994), and were
performed in enclosures or semi-opened populations (Del-
croix, Mauget & Signoret, 1990). In the present study, we
used behavioral and genetic approaches to investigate the
group structure and reproductive system of an uncon-
strained and hunted wild boar population. Demography of
this population, located in an eastern French forest, is well
known because of observation and monitoring for over
25 years (Gaillard et al., 1992; Calenge et al., 2002; Kamins-
ki et al., 2005; Servanty et al., 2007). These demographical
observations were combined with genetic data. We could
thus investigate the fine-scale genetic structure of the popu-
lation and discuss about the potential impact of hunting on
genetic structure. We also estimated how many males con-
tributed to the offspring of a social group, to more fully
understand the current mating system.
Materials and methods
Study area and sampling
Wild boars were studied in the Chaˆ teauvillain-Arc-en-
Barrois national forest in the north-east of France (481020N,
41560E). The forest covers 8500 ha and is surrounded by
other private and communal forests (2500 ha), and all are
mixed broadleaved deciduous woodlands (for more details
on study area see Gaillard et al., 1992; Kaminski et al.,
2005). All methods and protocols used for our research
were approved by the French Agricultural and Forest
Administration.
Wild boars have been studied and monitored in the
national forest since 1983 by a team from the ‘Office
National de la Chasse et de la Faune Sauvage’ (ONCFS,
National Office of Hunting and Wildlife). Boars were
caught in a network of box traps (Jullien et al., 1988) and
corral traps (Vassant & Brandt, 1995), baited with maize
and intensively used from March to September. Trap sizes
allowed simultaneous multiple captures and/or recaptures
of several animals at the same time (1–30 individuals). Sex
and age of all boars were recorded at first capture. Age was
estimated according to the tooth-eruption procedure de-
scribed by Matschke (1967), validated and adjusted to our
study area (Baubet et al., 1994). Three age classes were
distinguished: young ( 12 months), subadults (individuals
from 13 to 24 months) and adults (over 24 months). Wild
boars have a very short generation time, females becoming
reproductive as young as 12 months.
We obtained samples from two different sources between
October 1999 and February 2002: captured individuals
and hunted individuals. Captured animals (n=214) were
handled for only a few minutes in the early morning.
Animals were tagged with numbered plastic ear tags for
individual identification. Animals caught together in the
same trap were considered as belonging to the same group
and they received the same coloured set of tags, with
individual shape designs for adults and subadults. A punch
was used to collect small pieces of ear tissue, which was then
stored in 70% ethanol solution for subsequent genetic
analyses. To analyze relationships in female groups, we used
five family groups that were intensively followed for demo-
graphic and dynamic data (group composition, fusion–
fission events, death by hunting or disease) as well as for
space occupation over the 3-year sampling period (ONCFS,
unpubl. data).
The second part of our sampling (n= 274) came from
managed hunting in the national forest. This area has been
divided into 27 distinct plots (600–800 ha plot
1
). Hunting
occurred each year from October to February, in one or two
plots each day, and on almost every weekend. The demo-
graphy of this population has been followed by records from
hunting bags since 1978 (ONCFS, unpubl. data). During
our study period, wild boars were at high population
density. Each hunted animal and each catch were indexed
to a plot, allowing the description of individual and family-
group movements over time.
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London2
Socio-genetic structure of wild boar P. Chantal et al.
A total of 488 individuals, sampled by specific trapping
and by hunting, were studied in: 1999 (12 subadult females,
and six adult females with 52 piglets); 2000 (33 subadult
females, and 20 adult females with 102 piglets); 2001 (15
subadult females, and eight adult females with 81 winter-
born piglets). Additional hunt animals (n=159) were
sampled throughout the study area, with a substantial
number of tagged individuals helpful to reconstruct family-
group and to estimate group relatedness values.
Microsatellite loci and genetic structure
analysis
Twelve microsatellite loci were used (Robic et al., 1996). All
of these dinucleotide loci were highly polymorphic, un-
linked, and without null-alleles. Total DNA was extracted
by Qiamp minikit (Qiagen, Valencia, CA, USA). PCR
amplification was performed according to a standard proto-
col in two ways. For 289 individuals (59% of the sample),
PCR reactions were performed with a
33
P-dATP (Dallas
et al., 1995). Each PCR was run in a 10 mL volume contain-
ing 30–50 ng of DNA solution, 75 mM of each dNTP except
for dATP (5 mM), 0.04 mLofa
33
P-dATP, 0.10 mM of each
primer, 1 reaction buffer (50 mM KCl, 0.1% Triton X-
100, 10 mM Tris-HCl, pH 9.0), 1.5 mM MgCl
2
and 0.05 U of
Taq polymerase (Qiagen) using a PCR T1 thermal cycler
(Biometra, Goettingem, Germany). The thermal cycle pro-
file was an initial denaturation of 3 min at 94 1C; 30
amplification cycles of denaturation for 30 s at 94 1C,
annealing for 30 s at 53–61 1C (according the TM of the
locus) and extension for 45 s at 72 1C; and a final extension
for 5 min at 72 1C. PCR products were electrophoresed on a
standard 6% acrylamide sequencing gel and polymorphism
revealed by autoradiography. Genotypes were determined
at all loci. In the other part of the sample (n=199), only the
seven most polymorphic loci were examined. Genotypes
were produced by Antagene (Lyon, France), using fluores-
cent primers and an automated ABI Prism 310 sequencer
(Applied Biosystems, Foster city, CA, USA). A set of 15 in-
dividuals previously analyzed using radiolabel method were
also analyzed with the ABI system to validate consistency of
results. Allele size estimates were obtained by comparison
with a standard, using Genescan software.
Population differentiation between captured and hunted
samples, deviations from Hardy–Weinberg proportions
(HWE) and linkage equilibrium between loci were tested by
a Markov chain method (5000 steps of dememorization, 100
batches, 1000 iterations per batch) using Genepop 3.1c
(Raymond & Rousset, 1995). These tests were performed
on adults and subadults known to be alive in 2001, to avoid
use of non-independent genotypes caused by sampling
closely related boars. The number of alleles per locus (A),
observed (H
O
) and unbiased expected (H
e
) heterozygosi-
ties, and the likely frequency of null alleles (Nulls) were
computed using Cervus 2.0 (Marshall et al., 1998). The
population was tested for recent bottleneck effects (hetero-
zygote excess) with Bottleneck 1.2.02 (Cornuet & Luikart,
1996) assuming 10% multi-step mutations (Piry, Luikart &
Cornuet, 1999).
We used Wright’s F-statistics to analyze within and
between group social structures. We measured the departure
of heterozygosity from the expectation under random mat-
ing at the level of social groups by estimating F
IS
, and F
IT
at the level of the whole population. F
ST
was obtained as
a measure of genetic differentiation among social groups.
F-statistics were estimated according to Weir & Cockerham
(1984), with f=F
IS
and y=F
ST
. Their departure from the
null hypothesis (no population differentiation for F
ST
, and
random mating for F
IS
and F
IT
) was tested using 2500
permutations as implemented in Genetix 3.07 (Belkhir et al.,
1996–1997). F-statistics for piglets and adults were inter-
preted differently, because the former reflects the population
before dispersal and the latter after dispersal (Dobson, 2007).
To analyze the spatial genetic structure within the forest,
a spatial autocorrelation analysis was performed using
Spagedi 1.2 (Hardy & Vekemans, 2002). As wild boar males
disperse and females are philopatric, we expected related-
ness among females to be inversely related to distance.
Statistics were calculated for diploid multi-locus individual
genotypes for different classes of geographical distances
(individuals belonging to the same group (intra-group),
individuals caught in the same field (intra-field) or at 1500,
3000, 6000 and 11 000 m of distance. Spatial analyses were
carried out at the individual level using the kinship coeffi-
cient of Loiselle et al. (1995). Standard error (SE) for each
distance class was estimated by jackknifing over loci. P-
values were obtained by performing 10 000 permutations of
spatial locations of individuals within each class.
Relationships and parentage
Average relatedness (R) among piglets, adults and groups
was estimated using Relatedness 5.0.1 (Queller & Good-
night, 1989). To account for potential bias of related
animals in the estimation of population allele frequencies,
we calculated R-values using global allele frequencies esti-
mated over the entire sample of 488 wild boars. For R
estimations, groups were weighted equally and SEs were
obtained by jackknifing over groups or sites.
Parentage analyses were conducted using Cervus 2.0
(Marshall et al., 1998). We restricted paternity analyses to
cases where both mothers (adult or subadult females) and
their offspring were genotyped. As the wild boar could be a
facultative cooperative breeding species (Kaminski et al.,
2005), we first verified the absence of mismatches between
mother–offspring pair genotypes. All reproductive males
(410 months old, 50 kg) sampled in the studied area
during the breeding season (December–March) were con-
sidered as potential candidate fathers. Males sampled out of
the breeding season at Chaˆ teauvillain-Arc-en-Barrois were
also considered as father candidates when their weight was
compatible with a reproductive status during the last breed-
ing season. A special effort was made by hunters to sample
large males from hunting bags in 2000 and 2001; however,
low returns were obtained in 1999 and 2000 (Table 1). We
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London 3
Socio-genetic structure of wild boarP. Chantal et al.
ran different simulations of Cervus keeping the following
parameters constant: 10 000 cycles, 100 candidate parents,
90% of loci typed and 1% of loci mistyped. Genetic
paternities were assigned when the combined score given
was 95% as the strict confidence interval, and 80% as the
relaxed level.
Considering litters for which no male was assigned, we
reconstructed the father genotype using Gerud 1.0 (Jones,
2001). This program estimated the minimum number of
males (if mothers were known) contributing to a progeny
array and reconstructed those genotypes based on multi-
locus data. When multiple solutions were found, we ranked
the solutions by likelihood using priority score based on
both allele frequencies and Mendelian segregation.
Results
Microsatellite loci and genetic analysis
Genetic differentiation between captured and hunted sam-
ples was not significant (P=0.075), particularly given that
our sample size was sufficiently large for the statistical test
performed to have a high statistical power (n=214 and 274,
respectively). Thus, individuals were pooled for analyses.
The mean level of genetic diversity (H
e
) in our wild boar
population was 0.56 (range of H
e
=0.21–0.87) and a mean
number of alleles of 6.154 (Table 2). Exact tests for HWE
and linkage were performed on adults and subadults known
to be alive in 2001. No pair of loci departed significantly
from linkage disequilibrium after correcting for multiple
tests (adults; k=55, a=0.05; P= 0.021–0.96).
Departures of random mating within social groups (and
from H–W expectations of heterozygosity) as measured by
F
IT
were not significant over the whole population
(F
IT
=0.008, P=0.21). Genetic structure was assessed for
social groups composed of adults (mainly females for 1999
and 2000) and piglet cohorts separately (Table 3). Except for
1999, offspring exhibited a greater degree of heterozygote
excess compared to adults. In all years, offspring exhibited a
high degree of inter-group genetic differentiation (F
ST
of
0.15, 0.18 and 0.23, respectively, for 1999, 2000 and 2001).
Deviations of inbreeding from that expected under random
mating within social groups (F
IS
=f) across years and for
offspring ranged from 0.045 to 0.041, while for adults it
varied from 0.009 to 0.1. In the bottleneck analysis, tests
for heterozygote excess were significant for both the sign test
(P=0.008) and the standardized differences test (P=0.004).
Genetic spatial structure
We analyzed average relatedness values (R-values) for each
area when more than three animals were trapped together
(n=19 among the 27 areas), and R-values for females and
males separately. R-values were not significant from zero
for either males or females (R
m
= 0.037 0.086 and R
f
=
0.019 0.112). The mean R-value within area was
0.042 0.078 (range from 0.069 to 0.146). Adult between-
group differentiation was only estimated in 2001 and re-
sulted in a low but significant F
ST
(0.017 0.005, Po0.01).
Table 1 Demographic data from hunting bags and males sampled
Hunting
season
N
national forest
(8500 ha) N
m
N
m-father
N
m-sampled
(%)
1998–1999 546 271 (49.6%) 107 (39.5%) 10 (9%)
1999–2000 549 272 (49.5%) 96 (35.3%) 36 (37.5%)
2000–2001 535 263 (49.2%) 68 (25.9%) 56 (82%)
N
national forest
represents the number of individuals (males and
females) shot during the hunting season in the national forest, N
m
is
the total of males and their percentage from the total in brackets, N
m-
father
is the number of large males as candidate father and their
percentage from total of males in brackets, N
m-sampled
is the number
of males really sampled for genetic analyses and their percentage
from total of candidate males.
Table 2 Characteristics of the microsatellite loci polymorphism analyzed in the Cha
ˆteauvillain-Arc-en-Barrois wild boar Sus scrofa population
Locus Chromosome AH
o
H
e
HWE PNulls P
excl
S0155 1 2 0.20 0.21 0.37 0.023 0.09
S0226
a
2 5 0.54 0.50 0.06 0.043 0.25
SW240 2 8 0.68 0.71 0.19 0.029 0.47
S0002
a
3 10 0.77 0.77 0.19 0.001 0.58
S0227
a
4 5 0.48 0.48 0.62 0.018 0.26
IGF1
a
5 11 0.80 0.77 0.02 0.014 0.57
S0005
a
5 16 0.83 0.87 0.11 0.023 0.75
SW122
a
6 5 0.29 0.28 0.69 0.021 0.27
S0178
a
8 10 0.78 0.78 0.02 0.0001 0.59
SW911 9 3 0.62 0.66 0.83 0.020 0.36
S0215 13 2 0.22 0.21 0.98 0.001 0.09
S0218 X 2 0 0.38 –
Mean 6.15 4.52 0.564 0.22 0.567 0.24
a
Loci used to amplify all the dataset.
The chromosome location is indicated; Ais the number of alleles observed for each locus; the mean, expected (H
e
) and observed (H
o
)
heterozygosity, the probability of H
o
given H
e
(HWE P), the likely frequency of null alleles (Nulls) and the probability of parentage exclusion (P
excl
).
The sample size is 488.
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London4
Socio-genetic structure of wild boar P. Chantal et al.
Spatial autocorrelation analysis revealed the same pat-
tern of female structure for all 3 years (Fig. 1a for pooled
data): females showed significant negative spatial correla-
tion values (slope=0.005, P=0.04). F
ij
for females were
only significantly different from zero at small scale. For
males, there was no evidence of spatial structuring (Fig.
1b), with all F
ij
values being not significantly different
from zero.
Table 3 Genetic variance within and among generations of wild boar Sus scrofa
Number of
groups
Number of
individuals Sex ratio (F/M) F
IS
(f)R
1999 parents
aa
33
a
0.086 0.003
0.002 0.025
1999 piglets 12 52 0.94 0.041 0.002
0.05 0.024 (0.39 0.147)
b
2000 parents
a
39 1.6 0.10 0.004
0.001 0.027
2000 piglets 14 102 0.67 0.0230.001 0.003 0.02 (0.42 0.094)
2001 parents 19 167 0.96 0.0090.027 0.004 0.026
2001 piglets 11 81 0.81 0.045 0.002 0.017 0.029 (0.46 0.089)
a
Females older than 12 months (adults and subadults) and males older than 2 years were all putative parents and were pooled (all caught during
the breeding season, between October and March).
b
Average relatedness values within groups of piglets were reported between brackets.
Po0.05;
Po0.01.
Standard errors of F-statistics and average relatedness coefficients were obtained by jackknifing over loci.
Figure 1 Pattern of spatial autocorrelation ob-
served at the forest level for wild boar Sus
scrofa females and males for the three studied
years. Indices represent kinship coefficients
(Loiselle et al., 1995), given for each point,
1SE (jackknifed over loci). Indices are given
for individuals caught together (intra-group),
caught in the same field (intra-field) and accord-
ing to the logarithm of the distance separating
them (at 1500, 3000, 6000 and 11 000 m).
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London 5
Socio-genetic structure of wild boarP. Chantal et al.
Relatedness in sampled groups
Individuals, both adults and piglets, were sometimes caught
together in the same trap. Overall, we caught 40 groups
composed of at least three individuals per catch (n= 280
individuals). Captured group size averaged 7.05 individuals
(range=3–21). Average relatedness coefficients observed
for individuals within a group ranged from 0.05 0.08 to
0.88 0.07. Except in four cases where the average related-
ness within groups were o0.1, our data confirmed that
piglets caught in the same trap belonged to the same litter
or were highly related (average relatedness within groups of
0.303 0.176, n=36).
Relationships in female groups
Five family groups were followed from 1999 to 2000 for
family A (all individuals were hunted in 2000) and from 1999
to 2001 for the other families. Individuals were distributed in
three to four overlapping generations (Fig. 2). Almost all
groups were characterized by high individual survival be-
tween years. The group organization was largely conserved,
but with an elevation of female status (e.g. family C). Some
groups exhibited a dynamic organization, with a fission–
fusion pattern (e.g. families B, C and D). Despite regular
follow-up, we lost track of some females that disappeared
due to death or dispersal. Old females were rarely sampled
because they were difficult to capture.
The animals of five groups were analyzed using all
microsatellite loci (n=128 boars). Relationships among
individuals (Fig. 2) were obtained, or checked when they
were known, from genotypic data. Average relatedness
among individuals within family groups ranged from 0.063
(group D, 40 individuals) to 0.31 (group B, 23 individuals).
We observed three separate cases of new female arrival (two
adults and one subadult, Fig. 2) into an already constituted
social group. At the end of 1999, the group E was composed
of two subadult females and their 6–8 month piglets. These
two females were highly related with a value of 0.38. Both
females founded their own group during 2000 and an adult
female with her piglets joined one of these groups. Accord-
ing to age and distribution of alleles among genotypes, she
could have been the mother of the female she joined, but not
of the other female. Another unknown adult female joined
this group next year but was never caught. We compared the
relatedness level between her 3 offspring and both females
of the same group and, according to this indirect estima-
tion, she was more related to the adult female she joined
(R
offspringfemale1
=0.38) than the other (R
offspringfemale2
=
0.09). In group B, a subadult female joined the group with
her piglets in 1999. According to her genotype, she was
potentially the daughter of the old female, and the sister of
some females of her cohort.
Male genetic contribution
The total parentage exclusion probability was, respectively,
0.968 and 0.998 for the first and the second parent (Table 2).
We captured and genotyped 21 litters (five in 1999, eight in
2000 and eight in 2001) and their putative mother. Seven
piglets that did not match with the putative mother geno-
type were eliminated from these analyses. Litter size ranged
from two to six piglets (85 individuals, mean= 4.051 1.21).
Average relatedness among littermates (R
x
= 0.505 0.173)
and between mother–offspring (R
x
= 0.47 0.082) were not
Figure 2 Demography and R-values of five groups of wild boars Sus
scrofa sampled between 1999 and 2001 (&, male; , female; , never
caught for genetics; , unknown female joining the group). Individual
disappearance between 2 years was due to emigration, natural death
or hunting.
o1 year 42 years
Ontogeny: ? ??
Piglet Subadult Adult
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London6
Socio-genetic structure of wild boar P. Chantal et al.
different from the expected value of 0.5 in both cases (t-tests,
NS). For these 21 litters, paternity was assigned to 44 piglets
(451%). For the most stringent analysis, only 12 piglets
(14%) were assigned with 95% confidence. Rates of assign-
ment were always lower than those expected from simula-
tions. This can be due to a higher error rate or to a low
proportion of individuals sampled as true candidate fathers
in the population. From the 44 paternity assignments, 23
concerned at least two piglets of the same litter and corre-
sponded to 10 litters. Of these, five litters were assigned to
only one father while the other five litters were assigned to
more than one father. However, at 95% confidence, only
one litter had two different fathers assigned, four litters had
both piglets (and three piglets in one case) assigned the same
father. Only one male was assigned for the same breeding
season as father of two piglets from two different litters.
Results of the reconstruction of the father’s genotype
using Gerud 1.0 were more conservative; they were compa-
tible with single paternity in 19 litters. In the other two cases,
five piglets exhibited genotypes compatible with their
mother and one putative father, but the last piglet had an
incompatible allele with these two individuals for at least
three loci. Moreover, the reconstructed father genotypes
were compared to verify if a male sired several litters or
not during the same breeding season. When comparisons
were made for litter genotypes from the same family-groups
(12 sows, n=53 piglets), we found that different males
(11 different genotypes) each fertilized one female in the
group. In only one case, a male genotype was compatible as
father of two litters of separate females within the same
social group. These results are consistent with our Cervus
analyses. We obtained the same results when comparing
fertilization of synchronized (o10 days between estrous
peaks) females from different but adjacent areas (four sows,
n=16 piglets).
Discussion
Group structure
In Chaˆ teauvillain-Arc-en-Barrois, the average group size of
captured individuals was seven, with a high degree of
relatedness among individuals caught in the same trap,
especially among adult females. Consistent with previous
reports, our study confirms the wild boar’s matrilineal
structure of social groups (Kaminski et al., 2005) for up to
three generations in our sample (Fig. 2). Aggregation of
related individuals in family groups, as well as female
philopatry, structured the population genetically (Fig. 1
and strong F
ST
). High relatedness undoubtedly facilitates
the social cohesion observed in wild boars, in terms of
spatial and temporal associations, and of social interactions
among group members (Kaminski et al., 2005). For old
females, benefits of association with their daughter(s) likely
include the opportunity to be ‘helped’ by these close kin in
terms of thermoregulation during cold periods and ‘baby-
sitting’ piglets (Delcroix et al., 1990).
Changes in female social groups occurred in the wild
temporarily due to parturitions and/or to the departure of
subadult females (in 20% of cases, Kaminski et al., 2005).
A pattern of fission–fusion was observed during the study
period, as it has occurred often during long-term studies
conducted in the same population (ONCFS, unpubl. data)
and in other semi-natural populations (Delcroix et al.,
1990). Moreover we observed the arrival of new females
into already established groups, and with no ensuing ag-
gressive interactions (Fig. 2). In these cases, we showed that
new individuals were closely related to some members of
the group. Similar dynamic structure was reported for feral
pigs by Gabor et al. (1999) and Spencer et al. (2005). Such
a dynamic pattern was also genetically documented in the
African elephant and, in this highly matriarchal society,
maternal kinship predicts the fusion of social groups
(Archie, Moss & Alberts, 2006).
Factors influencing boar group-structure
Hunting practices may strongly modify wild boar socio-
genetic structure over the short and long term. During the
hunting season, hunting may facilitate changes in home
range (Sodeikat & Pohlmeyer, 2003) and in group structure
(Maillard & Fournier, 1995). Hunting may also influence
population size, with genetic long-term consequences as
shown by our bottleneck results. Tests for bottleneck effects
were significant in tests, indicating that the wild boar
population at Chaˆ teauvillain-Arc-en-Barrois had been re-
duced in the recent past. In addition to direct observation,
demography of this population is well known from hunting
bags. Records show that the population remained stable at
low density before 1988. Then, the population size quickly
increased from this date to a maximum of more than 1000
individuals shot per year, until a crash was recorded in 1993.
This crash caused the Hunting Federation to change the
hunting rules locally to preserve the population. Since this
change, the population increased up to the present, reaching
a high density of individuals again as elsewhere in Europe
(Monz ´
on & Bento, 2004).
Mating system
According to results from both observation and hunting
records, single paternity of wild boar litters was found to
be common. Litters were consistent with a single father in
19 cases, as opposed to two litters with multiple father
assignments. Monopaternity in feral pigs was reported by
Hampton, Pluske & Spencer (2004) in Australian popula-
tions and one rare case of multiple paternity was observed
by Delgado et al. (2008) in a Mediterranean region. Rearing
of piglets is likely cooperative in this species (Kaminski
et al., 2005), making misidentification between offspring
and their putative mother a distinct possibility. However,
in domestic pigs, it has been observed that sows mate with
different males and often produce litters of mixed paternity
(Aguilera-Reyes et al., 2006; Drake, Fraser & Weary, 2008).
Multiple paternities were also observed in a large proportion
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London 7
Socio-genetic structure of wild boarP. Chantal et al.
of feral sows from another Australian population (Spencer
et al., 2005). Thus, we tentatively conclude the possibility of
multiple paternity in this population of wild boars, but at a
very low rate during our study period.
The wild boar mating system has been described as
polygynous, with a male dominance hierarchy (Beuerle,
1975; Martys, 1991). Our data showed a low level of
polygyny inside social groups, numerous males contributing
to the next generation (12 sows mated with 11 different
males). For the sows that were receptive at the same time
and living in adjacent areas, we found that their piglets had
different fathers. According to genetic models (Chesser,
1991a,b), female natal philopatry and polygyny may coe-
volve to promote increased coancestry within social lineages
in small family groups. Polygyny in wild boars has also been
assumed because of individual distribution in forest and the
increase in male aggression in winter that coincides with
competition for females (Beuerle, 1975; Goulding, 1998).
However, we were unable to confirm strict monopolization
of females by highly polygynous males, as observed by
Hampton et al. (2004) for feral pigs.
Several factors might explain the absence of complete
polygyny encountered in this wild boar population. The first
one is the absence of dominant males. From 1993 to 2003,
hunting rules prohibited the killing of females weighing
more than 50 kg. According to the hunting records, killed
animals were either young animals of both sexes or adult
males, almost all males were shot before reaching 3 years
old (ONCFS), which is in accordance with results from
long-term studies in Europe (Fruzinski & Labudzki, 2002).
Hunting pressure may have acted as a regulating factor on
genetic diversity by increasing male turnover. A conse-
quence of hunting selection may be the progressive disap-
pearance of old and dominant males, which might
monopolize females during the breeding season. Hunting
regulations lead to selective effects that could be dramatic
for harvested populations. Recent reviews (Coltman, 2007;
Milner, Nilsen & Andreassen, 2007) have called attention to
the potential selective effects of sport hunting on wild
ungulates, particularly for trophy individuals. Recently,
wildlife managers at Chaˆ teauvillain-Arc-en-Barrois consid-
ered such evidence, and amended hunting rules to include
harvesting of large females.
In conclusion, even though a low level of multiple
paternity may occur, the social and mating system of the
wild boar at Chaˆ teauvillain-Arc-en-Barrois is probably best
described as polygynous, as opposed to polygynandrous.
Nonetheless, the degree of polygyny was not high, and lower
levels of polygyny produce lower degrees of genetic differ-
entiation among social groups. Yet wild boars exhibited
significant genetic structure among social groups, probably
due to female philopatry. In addition, single paternity of
litters, which seemed to be most common, produced litters
composed of full rather than half siblings. Male dispersal
and female fidelity to dynamic social groups produced a
remarkable social system in which cooperative reproduction
by females and high genetic diversity are both possible
(Dobson, 2007).
Acknowledgments
This study was supported financially and logistically by the
Office National de la Chasse et de la Faune Sauvage
(ONCFS). We wish to thank Jacques Vassant, the ‘Fe
´de
´ra-
tion des chasseurs de Haute-Marne,’ and all students and
volunteers for their help in the field. We are very grateful to
Prof. Monique Monnerot who welcomed C.P. into her
laboratory and to T. Do Chui and N. Dennebouy for
technical help. We also wish to thank N. Busquet, N.
Chaline, A Nettel-Hernanz, L. Say and J.M. Gaillard for
constructive comments. The genetic analysis was supported
by the grant no. 2000/6 between ONCFS and the University
Paris 13, LEEC. This work was conducted under France
legal requirements.
References
Aguilera-Reyes, U., Zavala-P ´
aramo, G., Valdez-Alarc ´
on,
J.J., Cano-Camacho, H., Garc ´
ıa-L ´
opez, G. & Pescador-
Salas, N. (2006). Multiple mating and paternity determi-
nations in domestic swine (Sus scrofa). Anim. Res. 55,
409–417.
Archie, E.A., Moss, C.J. & Alberts, S. (2006). The ties that
bind: genetic relatedness predicts the fission and fusion of
social groups in wild African elephants. Proc. Roy. Soc.
Lond. Ser. B 273, 513–522.
Baubet, E., Brandt, S., Jullien, J.M. & Vassant, J. (1994).
Valeur de la denture pour la de
´termination de l’aˆ ge chez le
sanglier (Sus scrofa). Gibier Faune Sauvage 11, 119–132.
Belkhir, K., Borsa, P., Chikhi, L., Goudet, J. & Bonhomme,
F. (1996-1997). GENETIX 3.07, Windows
TM
software for
population genetics. Laboratoire Ge
´nome et Populations,
Montpellier II University, Montpellier, France.
Beuerle, W. (1975). Field observations of aggressive sexual
behaviour in European wild hog (Sus scrofa L.). Zeit.
Tierpsych. 39, 211–258.
Boitani, L., Mattei, L., Nonis, D. & Corsi, F. (1994). Spatial
and activity patterns of wild boars in Tuscany, Italy.
J. Mammal. 75, 600–612.
Brandt, S., Vassant, J. & Jullien, J.M. (1998). Domaine vital
diurne des sangliers en foreˆ t de Chaˆ teauvillain - Arc-en-
Barrois. Bull. Mens. Office Chasse 234, 4–11.
Burton, C. (2002). Microsatellite analysis of multiple pater-
nity and male reproductive success in the promiscuous
snowshoe hare. Can. J. Zool. 80, 1948–1956.
Calenge, C., Maillard, D., Vassant, J. & Brandt, S. (2002).
Summer and hunting season home ranges of wild boar (Sus
scrofa) in two habitats in France. Game Wildl. Sci. 19,
281–301.
Carling, M.D., Wiseman, P.A. & Byersa, J.A. (2003).
Microsatellite analysis reveals multiple paternity in a
population of wild pronghorn antelopes (Antilocapra
americana). J. Mammal. 84, 1237–1243.
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London8
Socio-genetic structure of wild boar P. Chantal et al.
Carranza, J. (1996). Sexual selection for male body mass and
the evolution of litter size in mammals. Am. Nat. 148,
81–100.
Chesser, R.K. (1991a). Gene diversity and female philopatry.
Genetics 127, 437–447.
Chesser, R.K. (1991b). Influence of gene flow and breeding
tactics on gene diversity within populations. Genetics 129,
573–583.
Clutton-Brock, T.H. (1989). Mammalian mating systems.
Proc. Roy. Soc. Lond. Ser. B 236, 339–372.
Coltman, D.W. (2007). Molecular ecological approaches to
studying the evolutionary impact of selective harvesting in
wildlife. Mol. Ecol. 17, 221–235.
Cornuet, J.M. & Luikart, G. (1996). Description and power
analysis of two tests for detecting recent population
bottlenecks from allele frequency data. Genetics 144,
2001–2014.
Dallas, J., Dod, B., Boursot, P., Prager, E.M. & Bonhomme,
F. (1995). Population subdivision and gene flow in Danish
house mice. Mol. Ecol. 4, 311–320.
Delcroix, I., Mauget, R. & Signoret, J.P. (1990). Existence of
synchronisation of reproduction at the level of the social
group of the European wild boar (Sus scrofa). J. Reprod.
Fertil. 89, 613–617.
Delgado, R., Fernande
´z-Llario, P., Azevedo, M., Beja-
Pereira, A. & Santos, P. (2008). Paternity assessment in
free-ranging wild boar (Sus scrofa) – are littermates
full-sibs? Mammal Biol. 73, 169–176.
Dobson, F.S. (1982). Competition for mates and predominant
juvenile male dispersal in mammals. Anim. Behav. 30,
1183–1192.
Dobson, F.S. & Zinner, B. (2003). Social groups, genetic
structure, and conservation. In: Animal Behavior and Wild-
life Conservation: 211–228. Festa-Bianchet, M. & Apollo-
nia, M. (Eds). Washington D.C.: Island Press.
Dobson, F.S. (2007). Gene dynamics and social behaviour. In
Rodent societies: 163–172. Wolff, J.O. & Sherman, P.W.
(Eds). Illinois: University of Chicago Press.
Drake, A., Fraser, D. & Weary, D.M. (2008). Parent–
offspring resource allocation in domestic pigs. Behav.
Ecol. Sociobiol. 62, 309–319.
Emlen, S.T. & Oring, L.W. (1977). Ecology, sexual selection,
and the evolution of mating systems. Science 197, 215–223.
Fruzinski, B. & Labudzki, L. (2002). Management of wild
boar in Poland. Z. Jagdwiss. 48, 201–207.
Gabor, T.M., Hellengren, E.C., Van Den Bussche, R.A. &
Silvy, N.J. (1999). Demography, socio-spatial behaviour
and genetics of feral pigs (Sus scrofa) in a semi-arid
environment. J. Zool. (Lond.) 247, 311–322.
Gaillard, J.-M., Pontier, D., Brandt, S., Jullien, J.-M. &
Allaine
´, D. (1992). Sex differentiation in postnatal growth
rate: a test in a wild boar population. Oecologia 90,
167–171.
Goulding, M.J. (1998). Current Status and Potential Impact of
Wild Boar (Sus scrofa) in the English Countryside: a Risk
Assessment. Report to Conservation Management
Division C, Ministry of Agriculture, Fisheries and Food.
March 1998.
Greenwood, P.J. (1980). Mating systems, philopatry and
dispersal in birds and mammals. Anim. Behav. 28,
1140–1162.
Hampton, J., Pluske, J. & Spencer, P. (2004). A preliminary
genetic study of the social biology of feral pigs in south-
western Australia and the implication for management.
Wildl. Res. 31, 375–381.
Hardy, O.J. & Vekemans, X. (2002). SPAGeDi: a versatile
computer program to analyse spatial genetic structure at
the individual or population levels. Mol. Ecol. Notes 2,
618–620.
Hoogland, J.L. (1998). Why do female Gunnison’s prairie
dogs copulate with more than one male? Anim. Behav. 55,
351–359.
Jones, A.G. (2001). Gerud 1.0: a computer program for
the reconstruction of parental genotypes from progeny
arrays using multilocus DNA data. Mol. Ecol. Notes 1,
215–218.
Jullien, J.M., Vassant, J., Delorme, D. & Brandt, S. (1988).
Techniques de capture de sangliers. Bull. Mens. Office
Chasse 122, 28–35.
Kaminski, G., Brandt, S., Baubet, E. & Baudoin, C. (2005).
Life-history patterns in female wild boars (Sus scrofa L.,
1758): mother–daughter postweaning associations. Can.
J. Zool. 83, 474–480.
Komers, P.E. & Brotherton, P.N. (1997). Female space use is
the best predictor of monogamy in mammals. Proc. Roy.
Soc. Lond. Ser. B 264, 1261–1270.
Lacey, E.A., Wieczorek, J.R. & Tucker, P.K. (1997). Male
mating behaviour and patterns of sperm precedence in
Arctic ground squirrels. Anim. Behav. 53, 767–779.
Lane, J.E., Boutin, S., Gunn, M.R., Slate, J. & Coltman,
D.W. (2007). Relatedness of mates does not predict
patterns of parentage in North American red squirrels.
Anim. Behav. 74, 611–619.
Loiselle, B.A., Sork, V.L., Nason, J. & Graham, C. (1995).
Spatial genetic structure of a tropical understory shrub,
Psychotria officinalis (Rubiaceae). Am. J. Botany 82,
1420–1425.
Maillard, D. & Fournier, P. (1995). Effects of shooting with
hounds on size of resting range of wild boar (Sus scrofa L.)
groups in Mediterranean habitat. J. Mount. Ecol. 3,
102–107.
Marshall, T.C., Slate, J., Kruuk, L.E. & Pemberton, J.M.
(1998). Statistical confidence for likelihood-based paternity
inference in natural populations. Mol. Ecol. 7, 639–655.
Martys, M.F. (1991). Social organisation and behaviour in
the Suidae and Tassuidae. In Biology of Suidae: 65–77.
Barrett, R.H. & Spitz, F. (Eds). Toulouse: Institut de
Recherche sur les Grands Mammife
`res.
Matschke, G.H. (1967). Aging European wild hogs by denti-
tion. J. Wildl. Mgmt. 31, 109–113.
Mauget, R., Campan, R., Spitz, F., Dardaillon, M., Goneau,
G. & Pe
´pin, D. (1984). Synthe
`se des connaissances actuelles
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London 9
Socio-genetic structure of wild boarP. Chantal et al.
sur la biologie du sanglier, perspectives de recherche. Symp.
Int. sur le sanglier, Colloque INRA 22, 15–50.
Milner, J.M., Nilsen, E.B. & Andreassen, H.P. (2007). De-
mographic side effects of selective hunting in ungulates and
carnivores. Conserv. Biol. 21, 36–47.
Monz ´
on, A. & Bento, P. (2004). An analysis of the hunting
pressure on wild boar (Sus scrofa) in the Tr ´
as-os-Montes
region of northern Portugal. In Wild Boar Research
(2002). A selection and edited papers from the ‘‘4th
International Wild boar Symposium’’. Fonseca, C.,
Herrero, J., Lu ´
ıs, A. & Soares, A.M. (Eds). Galemys 16,
253–262.
Parker, P.G. & Waite, T.A. (1997). Mating systems, effective
population size, and conservation of natural populations.
In Behavioral approaches to conservation in the wild:
243–261. Clemmons, J.R. & Buchholz, R. (Eds). Cam-
bridge: Cambridge University Press.
Piry, S., Luikart, G. & Cornuet, J.M. (1999). Bottleneck: a
computer program for detecting recent reductions in the
effective population size using allele frequency data.
J. Hered. 90, 502–503.
Queller, D.C. & Goodnight, K.F. (1989). Estimating related-
ness using genetic markers. Evolution 43, 258–275.
Raymond, M. & Rousset, F. (1995). Genepop. version 1.2
population genetics software for exact tests and ecumeni-
cism. J. Hered. 86, 248–249.
Robic, A., Riquet, J. & Yerle, M., Milan, D., Lahbib-Mansais,
Y., Dubut-Fontana, C. & Gellin, J. (1996). Porcine linkage
and cytogenetic maps integrated by regional mapping of 100
microsatellites on somatic cell hybrid panel. Mammal Gen-
ome 7, 438–45.
Servanty, S., Gaillard, J.-M., Allaine
´, D., Brandt, S. &
Baubet, E. (2007). Litter size and fetal sex ratio adjustment
in a highly polytocous species: the wild boar. Behav. Ecol.
18, 427–432.
Sodeikat, G. & Pohlmeyer, K. (2003). Escape movements of
family groups of wild boar Sus scrofa influenced by drive
hunts in Lower Saxony, Germany. Wildl. Biol. 9, 43–49.
Spencer, P., Lapidge, S., Hampton, J. & Pluske, J. (2005). The
socio-genetic structure of a controlled feral pig population.
Wildl. Res. 32, 297–304.
Sugg, D.W. & Chesser, R.K. (1994). Effective population
sizes with multiple paternity. Genetics 137, 1147–1155.
Sugg, D.W., Chesser, R.K., Dobson, F.S. & Hoogland, J.L.
(1996). Population genetics meets behavioral ecology.
Trends Ecol Evol 11, 338–342.
Vassant, J. & Brandt, S. (1995). Adaptation du pie
´geage par
enclos-pie
`ge de type corral pour la capture de compagnies
de sangliers (Sus scrofa). Gibier Faune Sauvage, Game
Wildl. 12, 51–61.
Vernesi, C., Crestanello, B. & Pecchioli, E., Tartari, D.,
Caramelli, D., Hauffe, H. & Bertorelle, G. (2003). The
genetic impact of demographic decline and reintroduction
in the wild boar (Sus scrofa): a microsatellite study. Mol.
Ecol. 12, 585–595.
Weir, B.S. & Cockerham, C.C. (1984). Estimating Fstatistics
for the analysis of population structure. Evolution 38,
1358–1370.
Journal of Zoology ]] (2009) 1–10 c2009 The Authors. Journal compilation c2009 The Zoological Society of London10
Socio-genetic structure of wild boar P. Chantal et al.
