Early phases of a successful invasion: mitochondrial
phylogeography of the common genet (Genetta genetta)
within the Mediterranean Basin
Philippe Gaubert Æ José A. Godoy Æ
Irene del Cerro Æ Francisco Palomares
Abstract The Mediterranean Basin, connected by
cultural exchanges since prehistoric times, provides
an outstanding framework to study species translocations. We address here the early phases of the
successful invasion of the common genet (Genetta
genetta), a small carnivoran supposedly introduced
from Africa to Europe during historical times, by
assessing mitochondrial nucleotide variability in 134
individuals from its native and invasive ranges. We
identify four lineages within the native species range
[northern Algeria, Peninsular Arabia, southern
Africa and western Africa + Maghreb (including
northern Algeria)], in contradiction with morphological taxonomy. We propose that the cooccurrence in Maghreb of two divergent lineages
(autochthonous and western African) is due to
P. Gaubert I. del Cerro F. Palomares
Departamento de Biologı́a de la Conservación, Estación
Biológica de Doñana, Pabellón del Perù, Avda. Marı́a
Luisa s/n, 41013 Sevilla, Spain
P. Gaubert (&)
UR IRD 131 – UMS MNHN 403, Département Milieux et
Peuplements Aquatiques, Muséum National d’Histoire
Naturelle, 43 rue Cuvier, 75005 Paris, France
e-mail: gaubert@mnhn.fr
J. A. Godoy I. del Cerro
Laboratorio de Ecologı́a Molecular, Estación Biológica de
Donaña, Pabellón del Perù, Avda. Marı́a Luisa s/n, 41013
Sevilla, Spain
secondary contact through intermittent permeability
of the Saharan belt during the Plio-Pleistocene.
Estimates of coalescence time and genetic diversity,
in concert with other available evidences in the
literature, indicate that the origin of European
populations of common genets is in Maghreb,
possibly restricted to northern Algeria. The autochthonous mitochondrial lineage of Maghreb was the
only contributor to the European pool, suggesting
that translocations were associated to a cultural
constraint such as a local use of the species, which
might have artificially excluded the western African
lineage. Haplotype network and nested clade analysis
(NCA) provide evidence for independent events of
introductions throughout Spain (Andalucia, Cataluñ a,
and the Balearic Isl.)—and, to a lesser extent,
Portugal—acting as a ‘translocation hotspot’. Due to
the reduced number of northern Algerian individuals
belonging to the autochthonous mitochondrial lineage of Maghreb, it remains impossible to test
hypotheses of historical translocations, although a
main contribution of the Moors is likely. Our
demographic analyses support a scenario of very
recent introduction of a reduced number of individuals in Europe followed by rapid population
expansion. We suggest that an exceptional combination of factors including multiple translocations,
human-driven propagation across natural barriers,
and natural processes of colonization allowed by a
wide ecological tolerance, promoted the successful
spread of the common genet into Europe.
123
Keywords Carnivora Historical demography
Introduced species Mediterranean
MtDNA Phylogeography
Introduction
The Mediterranean Basin, connected by cultural
exchanges since prehistoric times, provides an outstanding framework to study species translocations,
notably in mammals (Dobson 1998; Gippoliti and
Amori 2006). After the faunal migrations between
Maghreb and Europe that occurred during the
Messinian crisis (ca. 7–5.3 Mya; van der Made et al.
2006), the Strait of Gibraltar acted as a biogeographic
barrier to dispersal (Castella et al. 2000; Dobson and
Wright 2000; Juste et al. 2004). It is only recently
that the biodiversity structure of the Mediterranean
Basin was profoundly modified through translocations of non-native species (Alcover 1980; Vigne
1992; Marra 2005; Gippoliti and Amori 2006),
notably mammals, occurring between Maghreb and
south-western Europe from ca. 9,000 years ago
(Dobson 1998; Vignes 1999; Anderung et al. 2005;
Modolo et al. 2005; Beja-Pereira et al. 2006;
Gippoliti and Amori 2006).
One of the main translocation flows into Europe
took place during the Moor invasion of Iberian
Peninsula, starting 711 A.D. and lasting ca. five
centuries (Wenner 1980; Dobson 1998; Cymbron et al.
1999; Ramon-Laca 2003). A legend relates that after
the defeat of Moor armies near Poitiers, France, in
732 A.D., the King’s Majordomo Charles Martel
found in the loot a great quantity of furs belonging to
an African small carnivoran, the common genet
(Genetta genetta) (Perrot 1820). This long stood as
the main evidence supporting the hypothesis of an
introduction of the species through Moor conquerors,
together with the dating of the only known archaeological remains of common genets in Europe to the
early 13th century Almohad levels in Portugal
(Morales 1994). Gsell (1913), and later Amigues
(1999), proposed an alternative scenario based on
mentions by antique Greek authors of a ‘‘weasel from
Tartessos’’ (south-western Iberian Peninsula), suggesting that the common genet was introduced in Europe
from Greek Lybian colonies during the 6th century
B.C. A third scenario was proposed by Schauenberg
(1966), who argued instead for the natural presence of
123
G. genetta in south-western Europe as a relict population from the Messinian period, despite the absence
of Plio-Pleistocene remains.
The common genet, which now spreads from
Portugal to continental Spain, Mallorca, Cabrera and
Ibiza (Balearic Isl.), south-western France and Italy
(Gaubert et al. 2008; Delibes and Gaubert in press),
represents, together with the Egyptian mongoose
Herpestes ichneumon (Palomares in press), a rare
example of successful translocation of a wild African
carnivoran into Europe. Poorly addressed but critical
to the understanding of the eco-evolutionary mechanisms that underlie such ecological successes are the
processes occurring during the early phases of
invasion, namely how, when and from where species
were introduced (Stepien et al. 2002; Muñ oz-Fuentes
et al. 2006; Roman and Darling 2007; Hayes and
Barry in press). We herein investigate the phylogeography of G. genetta in both its native and
invasive ranges using fragments of two mitochondrial
genes (cytochrome b and control region). We give
special emphasis to European populations, with the
aim of (i) reconstructing the evolutionary history of
Mediterranean populations during Plio-Pleistocene
climatic variations, (ii) tracing back the geographic
origin of European common genets and (iii) assessing
translocation patterns and historical demography
within the invasive range of the species. By doing
so, we attempt to characterize the early phases of
invasion of the common genets in Europe in the
historical context of the Mediterranean Basin.
Materials and methods
Geographic and nucleotide coverage
Genets are elusive small carnivorans that remain
extremely difficult to collect due to their nocturnal,
solitary way of life and their low density in the wild
(e.g. Gaubert et al. 2002). We examined nucleotide
variation in a total of 134 common genets from all the
regions invaded in Europe, including the Balearic Isl.,
and also most of its native range, covering Maghreb,
the Arabian Peninsula, western and southern Africa
(Appendix I). Given that outgroup taxon permutations did not influence ingroup topology (data not
shown), we only presented phylogenetic trees rooted
with G. maculata.
We amplified two fragments of mitochondrial
DNA (mtDNA), including (i) 403 bp of cytochrome b
(cytb) and (ii) 494–499 bp spanning the left domain
of the control region (CR1), after previously
described protocols and primers (Gaubert and Begg
2007); CR1 alignment (with outgroup) was 500 bp
long and included six indels. Shorter fragments were
obtained for 10 museum or badly preserved samples
from Maghreb and western Africa using the protocol
of Palomares et al. (2002). In the latter case, we
sequenced (i) 278 bp of cytb following previously
described primers and PCR cycles (Gaubert et al.
2004a, b), and (ii) 202 bp of CR1 partially covering
the domain of extended termination associated
sequences (Gaubert and Begg 2007) with a pair of
new primers: HVGg-H 50 CAC GAT ATA CAT AGT
ATG YCT T 30 - HVGg-L 50 GAA ATT CTT TTT
AAA CTA TTC CTT 30 (annealing temperature:
58°C). PCR products were directly purified from
reaction components by ultrafiltration using Montage
PCR filter units (Millipore Corp., Billerica, MA) or
isolated from a 1.5% agarose gel through the MinElute
Gel Extraction Kit (Qiagen S.A., Courtaboeuf,
France). We sequenced the purified products in both
forward and reverse directions in an ABI Prism 3100
Genetic Analyzer v. 2.0 (Applied Biosystems, Foster
City, CA) automated DNA sequence following manufacturer’s recommendations. All the sequences
reported in this paper have been deposited in the
GenBank database (accession nos. EF371617EF371704).
Phylogeographic analysis in the native and
invasive ranges of G. genetta
Sequences were aligned by eye using BioEdit 5.0.6
(Hall 1999). Models of molecular evolution for
statistical phylogenetic analyses were selected from
alignments pruned from indels and incomplete
nucleotide sequences using ModelGenerator (Keane
et al. 2004) and the Bayesian Information Criterion
(Posada and Buckley 2004). We performed model
selection with four gamma categories for the two
separated data sets (all individuals) and a concatenated matrix (cytb + CR1) consisting of 900 bp for a
geographically representative subset of 38 individuals
including all the variable cytb sequences (Fig. 1). The
best-fitting model selected for cytb and cytb + CR1
was HKY (Hasegawa et al. 1985) with a proportion
of invariable sites (I), and TrN (Tamura and Nei
1993) + I for CR1.
We ran maximum likelihood (ML) analysis through
PhyML (Guindon and Gascuel 2003), injecting model
parameters as calculated in ModelGenerator. Node
support was assessed using nonparametric bootstrapping with 500 pseudoreplicates. Bayesian analyses
were ran with MrBayes 3.1 (Ronquist et al. 2005).
Indels resulting from the CR1 alignment were coded as
binary characters using GapCoder (Young and Healy
2003) before running Bayesian analysis under a model
that mixed HKY + I (nucleotides) and ‘‘Restriction
Site’’ (three indel characters), an approximation of the
F81 model (Ronquist et al. 2005). For the concatenated
analysis we attributed models respective to each
partition, unlinking parameters and allowing partitions
to evolve under different rates. We used default priors
and fixed a gamma distribution divided into five
categories. We ran four Monte Carlo Markov chains
(MCMC) simultaneously for 2,000,000 Metropoliscoupled generations in two independent runs. Convergence between runs was estimated visually by
examining the plot ‘‘number of generations vs. log
likelihood values’’ and using the Potential Scale
Reduction Factor (PSRF) provided in the ‘‘sump’’
output of MrBayes. Burnin was fixed to 25,000 for each
data set with chains sampled every 10 generations. The
final majority-rule consensus phylograms were based
on 312,173 (cytb), 258,560 (CR1), and 225,341
(cytb + CR1) trees.
Coalescence time estimates for the main lineages
were calculated from a cytb + CR1 submatrix consisting of sequences without ambiguities (33 in total;
outgroup excluded). We used MCMCcoal v. 1.1.
(Yang and Yoder 1999; Yang 2002), which estimates
ancestral population sizes by extracting information
from conflicts among gene tree topologies and
coalescent times using a MCMC algorithm. This
method assumes a molecular clock (Jukes Cantor
model) and a known species (or population) tree
(Rannala and Yang 2003). ‘‘Speciation time’’ is
expressed as s = age of divergence 9 l, where l is
the mutation rate per site per year. We used a
divergence time of 3.55 Myr for the root of the
G. genetta tree as the midpoint between the first
geological record of the genus Genetta (4.1 Myr;
Werdelin 2003) and a molecular calibration estimate
obtained at the species node from a mitochondrial
phylogeny (3.1 Myr; Gaubert and Begg 2007). We
123
SOUTH-WESTERN
EUROPE
Maghreb
ARABIAN
PENINSULA
Sahara
Mr
France GgE149
Algeria GgT316
Se
AS
Portugal GgT477
Clade I
96
Om
western
Africa
Ma
Be
Ye
Ch
Spain (Baleares, Cabrera Isl.) GgT511
AFRICA
Spain (Baleares, Majorca Isl.) GgT514
Spain (continent) GgT408
0.99
Italy GgT540
Spain (Andalucia) GgT438
Algeria GgE91
Algeria GgE88
Na
Algeria GgT150
southern
Africa
SA
Algeria GgT317
Spain (Baleares, Ibiza Isl.) GgT487
Saudi Arabia GgC81
-
Clade II
1.00
Saudi Arabia GgT96
Clade III
1.00
88
-
Oman GgC52
0.95
99
-
Yemen GgT104
100
Saudi Arabia GgT502
Saudi Arabia GgT504
South Africa GgE19
South Africa GgE107
South Africa GgT98
100
100
Namibia GgE89
1.00
Namibia GgT380
1.00
Clade IV
75
0.99
Algeria GgE92
Tunisia GgT161
Benin GgT304
Benin GgT306
Benin (+ Chad) GgT298 (+ GgT327)
Morocco GgT120
0.05
Morocco GgT121
Algeria GgT539
98
95
0.95
Morocco GgT543
1.00
Mali GgE16
Mauritania GgE81
Senegal GgE65
Senegal GgE75
Fig. 1 Phylogeography of Genetta genetta inferred from
Bayesian analysis of cytochrome b + left domain of control
region (900 bp + 3 indels) under a model combining
HKY + I and ‘‘Restriction Site’’ (see ‘Materials and methods’). Values above and below nodes indicate bootstrap indices
C75% (ML analysis) and Bayesian posterior probabilities
C0.95, respectively. Scale bar corresponds to 5% sequence
divergence. The outgroup species (G. maculata) was removed
for convenience of visibility. Specimen numbers refer to
Appendix I. Countries sampled across sub-Saharan Africa and
Arabian Peninsula are indicated as follows: Mr: Mauritania;
Se: Senegal; Ma: Mali; Be: Benin; Ch: Chad; Na: Namibia;
SA: South Africa; AS: Saudi Arabia; Ye: Yemen; Om: Oman.
Detailed geographic sampling for Maghreb and south-western
Europe is given in Fig. 2
then calculated l from sG. genetta and applied the
mutation rate to all s values to obtain coalescence time
estimates along the tree. We used a burnin of 10,000
generations for a total of 100,000 samples taken every
two generations, and executed two independent runs
using different random number seeds. Priors for the
five-tuning variables were given values yielding a
range for the five acceptance proportions between 0.1
and 0.7. The shape parameters a of the gamma
distribution were fixed to values C1 (Yang 2002).
For both mtDNA fragments, number of polymorphic sites and haplotypes, haplotype and nucleotide
123
diversity, and genetic differentiation among lineages
estimated from haplotype and nucleotide statistics
(using Monte Carlo permutation test; 10,000 permutations) were calculated through DnaSP v. 4.10
(Rozas et al. 2003). Average number of pairwise
differences was estimated through MEGA v. 3.1
(Kumar et al. 2004).
Haplotype network and demographic history
within the Mediterranean border
We assessed genealogical relationships within populations from the Mediterranean border (Clade I; see
Fig. 1) on the basis of CR1 nucleotide variability for
a total of 100 individuals (cytb provided only one
variable site). We used (i) Network v. 4.1.1.2 (www.
fluxus-engineering.com) and the Median Joining
method (Bandelt et al. 1999), with parameter e fixed
to 0 (minimized production of median networks), and
(ii) TCS v. 1.21 (Clement et al. 2000), applying the
statistical parsimony procedure (Templeton et al.
1992) and a 95% cut-off for mutational connections.
In order to assess the impact of irregular sampling
effort across the area covered by Clade I on genetic
diversity estimates, we used the rarefaction analysis
(Simberloff 1972) applied to haplotype frequencies
(e.g. Bernatchez et al. 1989). This method is based on
the assumption that a greater sampling effort is likely
to yield a higher diversity of haplotypes at greater
frequencies. We used Rarefaction Calculator (J.
Brzustowski, http://www.biology.ualberta.ca/jbrzusto/
rarefact.php) to estimate the expected average number
of haplotypes in 42 hypothetical sub-samples
(equivalent to the number of locale in Europe;
Appendix I) of five individuals (i.e. the number of
northern Algerian individuals belonging to Clade I)
taken from the European sample set (95 individuals).
By doing so, we were able to assess whether the
number of haplotypes found in northern Algeria was
likely dependant on the sampling effort (value within
the range of expected diversity from the European
sample set) or not (higher or lower value). We also
used the program to estimate the distribution curve of
expected number of haplotypes with increasing
sample size. We calculated the rarefaction curve
using sub-samples randomly taken from of our
complete sample set, and changing their size incrementally from 1 to 100 (100 constituting the total
number of individuals in Clade I) (Formia et al.
2007). We then ran EstimateS v.8.0.0 (Colwell 2006)
to calculate the finite total number of haplotypes
within Clade I using the Michaelis–Menten asymptotic function (Colwell and Coddington 1994). The
program computes two different diversity estimators—MMRuns and MMMeans—from the estimated
Michaelis–Menten asymptote (based here on 100
individuals). We used them both to provide an
interval of values, MMRuns (averages over 100
randomization runs for each pooling level) tending to
over-estimate finite total number relative to
MMMeans (estimates for each sample pooling level
without randomization) (Colwell et al. 2004; Colwell
2006).
Geographical associations at different hierarchical
levels of the resulting CR1 haplotype network were
tested using the nested clade analysis (NCA) method
(Templeton et al. 1995; Templeton 1998), as implemented in GeoDis 2.5 (Posada and Templeton 2006).
In the particular case of invasive species, NCA might
indicate independent translocations to separate areas
under historical events such as ‘‘fragmentation’’ or
‘‘colonization’’ (see Templeton 1998). We applied
the standard, n-step nesting rules (Templeton et al.
1987; Templeton and Sing 1993) to define clade
hierarchy within the intraspecific network. Test for
geographical association of nested clades was done
by defining a location pairwise distance matrix
(Posada and Templeton 2006). We used Encarta
(Microsoft, Redmond, WA) to calculate distances
between localities, taking into account the geographic
shape of south-western Europe. We then built a
second distance matrix where we arbitrarily multiplied by 10 the pairwise distance values that involved
localities separated by the Mediterranean Sea, in an
attempt at accounting for impassable geographic
areas. Randomization of clade distance distribution
was performed through 1,000 permutations to obtain
statistical inferences at the 5% threshold. We used the
updated version of the GeoDis inference key (11 Nov
2005) to infer population structure and history at each
clade nesting level (Templeton 1998; Posada and
Templeton 2006).
Because NCA was demonstrated to yield a high
frequency of false positive under specific demographic
conditions (Panchal and Beaumont 2007; Petit 2008a;
but see Templeton 2008), we assessed historical
scenarios through alternative, independent methods,
whenever it was possible (Garrick et al. 2008; but see
123
France
F7
F8
5
EUROPE
S10
0
F6
F4
F5
F3
S9
Italy
I1
F2
S8
F9
F1
S7
S4
P2
S5
Portugal
S6
S3
S2
S11
1
Spain
P1
4
S13
3
Balearic
Isl.
S12
2
S1
1
2
A5
A3 A4
A1
A6
T2
A2
Tunisia
MAGHREB
M2
T1
Morocco
Algeria
3
M1
Hap 11
S1
Hap 12
Hap 17
2-1
1-8
A6
Hap 9
S3
S1
S1
8
S11
1 Hap 13
1-4
3-2
1-5
S4
Hap 8
1-6
A4 A5
F6
1-1
Hap 3
Hap 2
F3
F6
F7
F8
Hap 1
P1
P2
S1
S2
S3
S5
S7
S8
S9
S10
0
S11
1 S12
2 F1
F2
F4
1-7
F5
Hap 16
S6
F9
F3
A3
1-3
S13
3
P1
I1
P2
2-2
1-9
Hap 15
Hap 18
A4
F2
S6
1-2
F1
Hap 6
3-1
S6
F9
Hap 7
Hap 4
Fig. 2 Geographic sampling of Genetta genetta for Europe
and Maghreb, and haplotype network within Clade I inferred
from sequences of the left domain of the control region (CR1).
Localities representing Clade I are indicated in green tones.
Clade IV haplotypes from Maghreb are provided (in grey) but
were not included into the network because they belong to a
distinct lineage, non-sister group to Clade I (see Fig. 1). Codes
for regions sampled are detailed in Table 3. Numbers
correspond to: 1—Strait of Gibraltar (Moor invasion, starting
711 A.D.); 2—Carthage (814 to 146 B.C.); 3—Cyrenaica
123
(eastern coastal region of Lybia; former Greek colonies, ca.
700 to 525 B.C.); 4—Tartessos kingdom (ca. 1000 B.C. to
600 A.D.); 5—Poitiers, France (northernmost battle of Moors,
732 A.D.). Orange circles in the network correspond to
reconstructed, hypothetical haplotypes. Each connection
between points represents a single mutation (except in
Hap 11 and 12, which are linked to the rest of the network
by eight mutations). Nested clade design is such that dashed,
red and blue lines enclose 1-, 2- and 3-step clades, respectively.
The total cladogram regroups Clades 3-1 and 3-2
Petit 2008b). Given that the sample set from northern
Algeria was reduced, we investigated demographic
history within Clade I by concentrating on European
populations. We examined three groups, namely
Europe + northern Algeria (Clade I), Europe and
‘‘Europe without Clade 3-2’’ (i.e. excluding a divergent lineage possibly introduced independently; see
Fig. 2 and ‘Results’). We assessed demographic
expansion using (i) the Tajima selective neutrality test
(Tajima 1993) and (ii) the Fu neutrality test (Fu 1997),
to remove the potential effects of genetic hitchhiking
and background selection (Lessa et al. 2003). The two
tests were run with Arlequin v. 3.01 (Excoffier et al.
2005) using 4,000 permutations. We also examined
mismatch distributions (pairwise sequence differences; Rogers and Harpending 1992) and frequency
spectra (allelic frequency distribution at a site; Tajima
1989) through DnaSP to evaluate expanding population scenarios. Smoothness of observed distributions
was calculated through the raggedness statistics (r;
Harpending 1994) using coalescent simulations (1,000
replicates).
Results
Phylogeographic analysis
The cytb and CR1 alignments yielded 27 and 41
variable sites, of which 19 and 29 (including three
indels) were phylogenetically informative, respectively. We obtained similar tree topologies through
ML and Bayesian analyses. The combined analysis
suggested the existence of four well-supported geographic lineages within the common genet (Fig. 1),
including south-western Europe + northern Algeria
(Clade I), Arabian Peninsula (Clade II), southern
Africa (Clade III), and western Africa + Maghreb
[including Algeria] (Clade IV). CR1 fragment sizes
were clade-specific, yielding 499, 495 and 494 bp for
Clade I, Clades III–IV and Clade II, respectively.
Clade I was sister-group to the other clades; the node
grouping (Clade II, (Clade III, Clade IV)) was supported in the Bayesian analysis only.
Assuming a 3.55 Myr time of origin for G. genetta,
the mutation rate per site per year l for cytb + CR1
was 1.12 (0.88 - 1.36) 9 10-9. Using this rate, the
split between all four major lineages could be dated
back to Pliocene (Table 1). Coalescence times within
Table 1 Coalescence times of lineages along the phylogenetic
tree of Genetta genetta
Clade
Coalescence time (Myr)
CI
(I, (II, (III, IV)))
(II, (III, IV))
3.55a
3.47
–
(2.86–4.42)
(III, IV)
I
3.32
0.43
(2.74–4.23)
(0.35–0.55)
II
III
IV
1.53
2.67
0.13
(1.26–1.94)
(2.20–3.40)
(0.11–0.17)
Numbering of clades refers to Fig. 1. Coalescence time is
given with 95% confidence interval (CI)
a
Mean time of origin fixed at the G. genetta node (see
‘Materials and methods’)
each lineage yielded more precise estimates (i.e.
narrower CI values), with southern Africa and Arabian
Peninsula clades coalescing in the Late Pliocene and
Early Pleistocene, respectively. Extant populations
from Europe + northern Algeria and western Africa
coalesced more recently (Late Pleistocene).
Examination of genetic differentiation within
G. genetta (Table 2) revealed that Clade I had a
mitochondrial diversity significantly lower than those
calculated for Arabian and sub-Saharan lineages,
including the partly sympatric ‘western African
+ Maghreb’ clade (data not shown). The number of
mtDNA polymorphic sites and haplotypes was lower
in the northern Algerian pool (Clade I) than in its
European counterpart, with two cytb haplotypes
detected in Europe for only one in northern Algeria.
However, this picture is most probably due to the low
number of samples available for the latter region (see
‘Discussion’). On the other hand, nucleotide diversity,
haplotype diversity and mean pairwise differences in
CR1, the most variable gene fragment, were much
higher for northern Algeria. Genetic differentiation
between Algerian Clade I and European populations
was not significant in that case.
Haplotype network and demographic history
within Clade I
Both methods of genealogical reconstruction yielded
the same CR1 haplotype network for common genets
from the Mediterranean border (Clade I). A total of
18 haplotypes were found, three of them restricted to
northern Algeria (Fig. 2 and Table 3). Hap 8 was
123
Incomplete nucleotide sequences removed
a
Numbering of clades refers to Fig. 1; cytb and CR1 go for cytochrome b and left domain of control region, respectively. 95% CI and standard error are given between parentheses
(haplotype diversity and mean pairwise difference, respectively)
0.923 (±0.394) 3.263 (±0.924)
1.611 (±0.698) 5.333 (±1.394)
0.01111
0.722 (±0.312) 0.917 (±0.143) 0.00400
6
5
14
5
495
9
403
3.600 (±1.403) 3.000 (±1.308)
Clade IVa
9
0.01285
0.154 (±0.247) 0.901 (±0.084) 0.00111
9
2
9
2
495
19
13
Clade IV
403
3.933 (±1.186) 3.222 (±1.125)
0.00652
0.00606
0.800 (±0.321) 0.600 (±0.343) 0.00893
0.600 (±0.421) 0.750 (±0.220) 0.00976
4
2
3
3
9
5
6
10
494
495
403
5
Clade III
403
5
6
Clade II
9
2.400 (±0.954)
_
0.00481
0.900 (±0.316) _
_
4
1
6
0
499
403
5
North Algeria
5
0.154 (±0.149) 1.457 (±0.378)
0.250 (±0.246) 1.328 (±0.338)
0.00292
0.00266
0.154 (±0.247) 0.630 (±0.108) 0.00038
0.250 (±0.353) 0.590 (±0.114) 0.00062
18
15
2
2
26
23
1
1
499
499
403
403
100
95
13
8
cytb
CR1
CR1
cytb
CR1
cytb
CR1
cytb
CR1
cytb
CRI
Clade I
Europe
cytb
cytb
CR1
Nucleotide diversity Mean pairwise difference
N sequences N sites (alignment) N polymorphic sites N haplotypes Haplotype diversity
Table 2 Mitochondrial diversity among the main lineages of Genetta genetta
123
shared between Algeria and Europe (one specimen
from Cataluñ a, Spain). The number of mutations
separating the different haplotypes was generally low
(1–4), with the exception of a cluster restricted to
Andalucia, southern Spain (Hap 11 and 12), which
was nine mutations distant from its putative Algerian
relatives (it also showed a unique cytb haplotype).
Hap 3 was the most frequent and widespread across
Europe (60%), including the Balearic Isl. of Mallorca
(where Hap 3 co-occurred with the unshared Hap 13)
and Cabrera (Hap 3 = 100%). The common genets
from Ibiza (Balearic Isl.) had a single, unshared
haplotype (Hap 14), not found on the continents.
Populations from western and central France consisted of two exclusive haplotypes (Hap 1 and 2).
Results from the rarefaction analysis yielded an
expected number of haplotypes for northern Algeria
of 2.73 (±1.88 [95% CI]). The number of haplotypes
detected in Spain (41 individuals; 10 haplotypes),
Italy (1; 1), and northern Algeria (5; 4) fell within the
95% confidence limits of the rarefaction curve,
whereas France (35; 5) and Portugal (18; 2) had
lower diversities than expected (7–14 and 4–10
expected haplotypes, respectively). The rarefaction
curve showed that expected haplotype diversity
increased with sample size (Fig. 3). Michaelis–Menten
estimators yielded values of 22,26 (MMRuns) and
21,45 (MMMeans), thus approximating the total finite
number of 22–23 haplotypes within Clade I.
The nested clade structure resulting from the CR1
haplotype network consisted of four levels (Fig. 2).
Significant geographical associations were revealed at
two clade levels (Table 4). Both distance matrices
yielded similar scenario inferences (see ‘Materials and
methods’). Clade 1-3 (northern Algeria, Portugal,
Cataluñ a, Ibiza Isl.) had a geographical pattern compatible with allopatric fragmentation, or long distance
colonization if steps 19–20 of the chain of inference
were considered inapplicable (absence of terrestrial
habitats between haplotypes). Restricted gene flow
with isolation by distance was inferred for two 1-step
clades (1-1: western France; 1-2: continental Europe
plus Mallorca and Cabrera Isl.), whereas geographical
associations within Clade 2-1 (Algeria and Europe) fit
the pattern expected under restricted gene flow/
dispersal with some long distance dispersal.
Tests of demographic history yielded significant D
(Tajima’s) and Fs (Fu’s) statistics in every CR1
haplotype partition (Table 5), thus suggesting genuine
Samples
N Hap
1
Hap
2
Hap
3
Hap
4
Hap
5
France
Languedoc-Roussillon (F1)
Midi-Pyrénées (F2)
35 9
5
3
2
20
3
2
2
2
4
Aquitaine (F3)
Limousin (F4)
Auvergne (F5)
Poitou-Charentes (F6)
11 2
1
2
4 2
Pays-de-la-Loire (F7)
Centre (F8)
PACA (F9)
Italy
4 4
1 1
4
1
Piemonte (I1)
Spain
Hap
6
Hap
7
Hap
8
Hap
9
Hap
10
Hap
11
Hap
12
Hap
13
Hap
14
2
1
1
3
1
1
1
1
4
1
1
1
Hap
15
1
1
5
1
Madrid (S3)
Castilla-y-Leon (S4)
Aragon (S5)
Cataluña (S6)
2
2
2
4
1
Pais Vasco (S7)
Cantabria (S8)
2
1
2
1
Asturias (S9)
Galicia (S10)
Baleares, Mallorca Isl. (S11)
Baleares, Cabrera Isl. (S12)
2
3
3
7
2
3
2
7
4
18
5
15
3
3
2
12
1
1
1
2
2
2
1
1
1
4
2
531
123
1
3
8
1
5
1
1
1
2
Andalucia (S1)
Extremadura (S2)
Algeria
Algiers (A3)
Hap
18
1
1
26
Porto – Braga – Viana de Castelo 13
(P2)
Hap
17
9
1
2
1
41
Baleares, Ibiza Isl. (S13)
Portugal
Beja (P1)
Hap
16
Phylogeography of the common genet
Table 3 Sample set of Genetta genetta within Clade I and distribution of CR1 haplotypes by main regions
P. Gaubert et al.
demographic expansion within Clade I and Europe.
Mismatch analysis revealed a slightly bimodal distribution of pairwise nucleotide differences within
Clade I and Europe, and a unimodal distribution
when the divergent lineage from southern Spain
(Clade 3-2) was excluded (Fig. 4). Although the
typical ‘smooth peak-like’ distribution characterizing
population expansion was not found, our mismatch
pattern could fit a scenario of very recent population
expansion, as suggested by mutation class distributions, which showed an excess of 1–2 step mutations
relative to the frequency expected under a model of
constant population size (Fig. 4).
1
Hap
12
Hap
13
Hap
14
Hap
15
Hap
16
1
Hap
17
Hap
18
532
123
1
1
Jijel (A5)
El Kala (A6)
Codes for samples and numbering of haplotypes refer to Fig. 2. Sample sizes are given under the ‘‘N’’ column
1
1
2
Zeralda (A4)
Samples
Table 3 continued
N Hap
1
Hap
2
Hap
3
Hap
4
Hap
5
Hap
6
Hap
7
Hap
8
Hap
9
Hap
10
Hap
11
Discussion
Phylogeography of the common genet in its native
range: the Saharan belt as an intermittent
geographic barrier
The mitochondrial phylogeographic analysis identified
four geographic clusters within the native range of the
common genet: northern Algeria, Arabian Peninsula,
southern Africa and western Africa + Maghreb
(including Algeria), with the two latter clades strongly
supported as sister-groups (Fig. 1). Our results are not
concordant with craniometrical analyses that grouped
‘‘western Africa – Maghreb - Europe’’ and ‘‘southern + eastern Africa - Arabian Peninsula’’ in two
distinct clusters (Crawford-Cabral 1981). Despite the
co-occurrence of two divergent haplogroups in
Maghreb, the morphological homogeneity of
G. genetta in this region is remarkable (Delibes and
Gaubert in press), especially since western African
common genets show a different pelage from those of
the Mediterranean rim (Schlawe 1981). We here
suggest that previous taxonomic works based on
pelage similarity (Schlawe 1980) were likely misled
by the adaptability of coat pattern to environmental
conditions. Indeed, one of us (PG) could observe from
museum collections that genets under similar climatic
conditions tended to exhibit similar coat pattern (e.g.
spots dark, ground pale greyish and long hair under
Mediterranean climate in G. genetta from Maghreb
and Europe, and the South African small-spotted genet
G. felina).
The alternation of periods of isolation and permeability of the Saharan belt since the Pliocene
Phylogeography of the common genet
24
21
18
n haplotypes
Fig. 3 Rarefaction curve of
CR1 haplotype diversity
within Clade I (95%
confidence interval (vertical
grey bars) = standard
deviation 9 1.96). The top
dashed lines indicate the
total finite number of
haplotypes calculated from
two asymptote diversity
estimators using EstimateS
(MMRuns = 22,26;
MMMeans = 21,45). The
five countries represented in
Clade I are plotted as black
circles
533
15
12
Spain
9
6
France
N Algeria
3
Portugal
0
0
25
50
75
100
125
150
175
200
n samples
(Thomas 1979; Thomas et al. 1982) may be responsible for the complex scenario we evidenced in
common genets from Maghreb, combining an allopatric isolation that generated a homogeneous
haplogroup (Clade I) and a subsequent arrival of
western African migrants (Clade IV). The times of
coalescence for Clades I and IV were unexpectedly
recent (Late Pleistocene), which suggests smaller
long term effective population sizes possibly due to
the accumulation of hyper-arid cycles during PlioPleistocene. We propose that recurrent bottleneck
events affected northern Algerian haplotypes (Clade I) during refuge-like forest contractions following
arid phases in Maghreb (Harris et al. 2002; Cosson
et al. 2005; Modolo et al. 2005), coincident with the
estimated time of coalescence of Clade I (350,000–
550,000 years), and as suggested by a significantly
lower mitochondrial diversity compared to subSaharan and Arabian lineages. On the other hand,
the presence of haplotypes from western Africa argue
for a recent, northward passage of common genets
through the Saharan belt, possibly during the major
extension of the palaeomonsoon system (Dobson and
Wright 2000), which induced a semi-humid climate
all over Sahara (Kuper and Krö pelin 2006) and a
vegetation—in its western part—similar to that of the
modern Mediterranean biome ca. 6,000 B.P. (Jolly
et al. 1998).
On the geographic origin of the European
common genet
The question of the geographic origin of the European common genet has long excited the interest and
imagination of zoologists. As a consequence, some
preconceived ideas have emerged in the community,
although very few have been based on archaeozoological or historical evidences. By reviewing the
literature available in a variety of fields and taking
into account our own results, we intend to clarify the
possible scenarios of introduction of the species in
south-western Europe:
(i) The paleontological and phylogenetic evidence: despite the presence of numerous excavation
sites in south-western Europe, and notably in Spain,
the genus Genetta has never been recorded (Kurten
2007); more generally, the frequency of Viverridae
fossils is extremely low compared to other carnivoran families (Montoya et al. 2001). The fossil
described under Genetta plesictoides from Cyprus
caves (Bate 1903)—the dating of which remains
ambiguous (Boekschoeten and Sondaar 1972)— has
been unequivocally attributed by Wolsan and Morlo
(1997) to Herpestides antiquus, a Viverridae Miocene fossil present in Europe. On the other hand,
the genus Genetta is mentioned in the Early
Pliocene of Kenya (Werdelin 2003) and the Late
123
534
P. Gaubert et al.
Table 4 Results of the nested clade analysis conducted within Genetta genetta Clade I
Clades
Geographic distance analysis
1–1
Hap 1 (int)
Hap 2 (tip)
1-2
Hap 3 (int)
Hap 4 (tip)
Hap 6 (tip)
Hap 7 (tip)
1–3
Hap 8 (int)
Hap 14 (tip)
Hap 15 (tip)
Hap 16 (tip)
2-1
1-1
1-2
1-3
1-4
1-6
1-7
(tip)
(int)
(int)
(int)
(tip)
(tip)
1-8 (int)
Dc = 177; Dn = 169
Interior versus tip clade test
Chain of inference
I-TDc = 177L; I-TDn = 38
1-2-3-4-No: restricted gene flow with
isolation by distance
I-TDc = 469L; I-TDn = 33
1-2-3-4-No: restricted gene flow with
isolation by distance
I-TDc = 371; I-TDn = -25
1-19-No: allopatric fragmentation
1-(19-20)-2-3-5-15-No: long distance
colonization (short branch length
between haplotypes)
I-TDc = 455L; I-TDn = -38
1-2-3-5-6-7-Yes: restricted gene
flow/dispersal but with some
long distance dispersal
L
Dc = 0S; Dn = 130
Dc = 610; Dn = 611
Dc = 225S; Dn = 627
Dc = 0S; Dn = 496
Dc = 0; Dn = 492
Dc = 496; Dn = 582
Dc = 0 S ; Dn = 453S
Dc = 335; Dn = 842L
Dc = 0; Dn = 509
Dc
Dc
Dc
Dc
Dc
Dc
=
=
=
=
=
=
S
162 ; Dn = 687
608S; Dn = 630S
606; Dn = 708L
422; Dn = 558S
0; Dn = 708
0; Dn = 585
Dc = 0; Dn = 1089
Numbering of clades corresponds to Fig. 2, with interior (int) and tip clades indicated. Geographic distance analysis was conducted
using a distance matrix, with the clade distance Dc as the geographical spread of a clade, and the nested clade distance Dn as the
distance of a clade from the geographical centre of the nested clade (Templeton 1998). The interior versus tip clade distances were
calculated as the difference between the average interior distance and the average tip distance (I-TDc and I-TDn). Significant values at
the 5% level are indicated by S (significantly small) or L (significantly large). Chain of inference refers to the steps in the key of
Templeton (11Nov05); we represented an alternative chain for Clades 1–3 where steps 19–20 were considered inapplicable (absence
of terrestrial habitats between haplotypes)
Table 5 Tests of population growth within Genetta genetta
Clade I
Clade I
Europe
Europea
Tajima’s
test (D)
Fu’s
test (Fs)
Raggedness
statistic (r)
-2.126*
-2.090*
-1.839*
-11.229*
-8.149*
-8.629*
0.178*
0.207*
0.277*
Statistically significant values are followed by an asterisk
a
Divergent lineage from southern Spain excluded (Clade 3-2;
see Fig. 2 and ‘Results’)
Pliocene of South Africa (Hendey 1974) and Morocco
(Geraads 1997). In accordance with the fossil data,
molecular phylogenetic analysis coupled to reconstruction of ancestral geographic areas strongly confirmed
123
the African origin of all the species belonging to the
genus Genetta (Gaubert and Cordeiro-Estrela 2006).
(ii) The archaeozoological evidence: the only
record of the common genet in Europe was reported
from the first quarter of the XIIIth century in Baixo
Alentejo, Portugal (Morales 1994), which corresponds to the Almohad levels of Mértola, the last
Berber dynasty who occupied the Iberian Peninsula.
On the other hand, remains of common genets were
found earlier in Maghreb, in the Late Pleistocene of
Morocco (Ouchaou and Amani 2002) and Algeria
(Romer 1928), where they were associated to hunting
sites, thus supporting the hypothesis of a recent event
of translocation from Maghreb to Europe. The
assertions of a large number of authors as to the
use of the species as a possible ‘pre-cat’ against rats
and snakes in Egypt and in Maghreb are not based on
Phylogeography of the common genet
0.4
Clade I (Europe + northern Algeria)
0.3
Frequency
Frequency
0.3
0.2
0.1
0.2
0.1
0.0
0.0
0
5
10
15
Pairwise differences
20
25
0
20
40
Segregating sites
Europe
0.3
Frequency
0.3
0.2
0.2
0.1
0.1
0.0
0.0
0
5
10
15
20
25
0
10
Pairwise differences
20
30
40
50
40
50
Segregating sites
0.5
0.4
0.4
Europe*
Frequency
Frequency
50
0.4
0.4
Frequency
Fig. 4 Mismatch (left) and
segregating site (right)
distributions of CR1
haplotypes of Genetta
genetta within Clade I and
Europe. In mismatch
analysis, expected
frequency is based on a
population growth-decline
model and is represented by
a continuous line. Observed
frequency is indicated as a
dotted line. In segregating
site analysis, expected
distribution is based on a
constant population size
model.
* European haplotypes
excluding a divergent
lineage from southern Spain
(Clade 3-2; see Fig. 2 and
‘Results’)
535
0.3
0.2
0.3
0.2
0.1
0.1
0.0
0.0
0
5
10
15
Pairwise differences
any concrete evidences (see Osborn and Osbornova
1998).
(iii) The historical evidence or why not an origin
from the Levant? Several Mediterranean civilizations
were designated as having possibly contributed to
the introduction of the common genet in Europe; yet
again concrete grounds are missing. The most
elaborated scenario to date, based on Cyrenaican
coinage and Heredotus’ description of the Lybian
fauna, suggested that the species was introduced by
the Greeks from their Libyan colonies (Amigues
1999). This widens the debate to a possible spread
by the first great sailors of the Mediterranean, the
Phoenicians, whom were initially based in the
coastal Middle East (current Lebanon and Israel).
However, previous mentions of common genets in
the region (Belon du Mans 1557; Tristram 1866;
Tristram 1884) were hoaxes (Schlawe 1981) or
misidentifications with another small carnivoran
species (the marbled polecat Vormela peregusna;
Kock 1983). Genuine records of the species simply
20
0
10
20
30
Segregating sites
do not exist in the Middle-East (Y. Yom-Tov, in
litt., 1999).
(iv) Evidence from mtDNA analysis: The mtDNA
data showed European genets to be closest relatives
of northern Algerian genets, together forming a clade
(Clade I) for which the estimation of coalescence
time in Late Pleistocene (Table 1) ruled out the
hypothesis of a natural crossing of G. genetta through
the Gibraltar passage during the Messinian crisis
(Schauenberg 1966). Rarefaction analysis suggested
that we probably detected most of the haplotypes
present in the European sample, with a good proportion (78–81%) of the total haplotype diversity
(22–23) estimated for Clade I (Fig. 3). This leaves
little room for unsampled haplotypes from additional
divergent sources and strongly suggests that the
mitochondrial source population from which the
European common genets were introduced was from
Maghreb, with a possible restriction to northern
Algeria (based on our sample set; see Fig. 2).
Furthermore, nucleotide diversity, haplotype diversity
123
536
and mean pairwise differences in CR1 were much
higher for northern Algeria than for Europe, supporting the former as source for the latter. Finally, CR1
mismatch distribution indicated that population
expansion in northern Algeria predated that in Europe
(data not shown), confirming that the northern Algerian population is older (Brandli et al. 2005). Similar
molecular studies involving translocations of micromammals identified the geographic sources of European populations in Morocco or Tunisia, but in every
case samples from Algeria were missing (Libois et al.
2001; Michaux et al. 2003; Brandli et al. 2005;
Cosson et al. 2005).
Translocation scenarios and historical
demography of the European common genets
The results from the rarefaction analysis confirmed
that, although our sample set covered a broad range
of Maghreb (15 samples in total, distributed between
Clades I and IV), Clade I haplotypes may have gone
undetected in Morocco and Tunisia. Indeed, we found
four haplotypes in northern Algeria, a number that
fell within the range of expected number of haplotypes estimated from the European pool using a subsample fixed to five individuals (2.73 ± 1.88)
(Fig. 3). On the other hand, the probability of
presence of haplotypes from Clade IV in Europe,
with 95 individuals representing ca. 4/5 of the total
expected Clade I haplotype diversity, can be considered low. The restriction of the source of European
common genets to Clade I haplotypes, despite the cooccurrence of a divergent lineage (Clade IV) in the
whole Maghreb, may appear unexpected given the
successive, numerous exchanges that occurred
between Maghreb and Europe since ca. 10,000 years
B.C. (Arnaiz-Villena et al. 1999; Scozzari et al.
2001; Rodrı́guez-Ariza and Moya 2005; Casas et al.
2006). The prevalence of Clade I haplotypes, whether
or not the original pool is actually restricted to
northern Algeria, suggests either that translocations
were associated to a cultural constraint, such as a
local use of the species by peculiar people, or that the
crossing of the Mediterranean was effective before
the migrant lineage from West Africa (Clade IV)
reached Maghreb. The latter option would imply
however that the event occurred accidentally across
the Strait of Gibraltar (e.g. through rafting) earlier
than 110,000–170,000 years ago, and thus was not
123
P. Gaubert et al.
renewed later; an unlikely hypothesis given the low
support from biogeographical data in the region
(Dobson 1998; Cosson et al. 2005), the genetic
diversity detected in Europe (lower than in northern
Algeria, but not diagnostic of a strong founder event),
the presence of common genets in Balearic Isl., and
the genetic evidence for multiple introductions (see
below). We would thus opt for recent events of
introduction conditioned by cultural constraint,
although this scenario still suffers from a lack of
archaeozoological evidence relative to the possible
use of the common genet against pests in the
Mediterranean Basin (Kingdon 1977; Amigues
1999). Interestingly, an example of a culturally
dependant introduction was recently reported for the
Algerian hedgehog Atelerix algirus, which was
introduced by the Almohads from Morocco to the
Mediterranean islands and the coastal Iberian Peninsula for meat consumption and medicine (Morales
and Rofes 2008).
Strong evidence for independent events of translocation resided in the co-occurrence of two very
divergent mitochondrial lineages in Europe (Clades
3-1 and 3-2; Fig. 2). Clade 3-2, which appeared
endemic to Andalucia, southern Spain, was characterized by unique cytb and CR1 haplotypes, indicative
of a northern African origin from an unsampled
genetic pool belonging to Clade I. Additional samples are needed, notably from former Cyrenaica
(Greek Lybian colonies), to assess whether
the persistence of Clade 3-2 in Andalucia, near the
ancient Tartessian kingdom’s boundaries, is the
signature of the introduction of common genets by
Greeks during the 6th century B.C. (Amigues 1999).
However, the presence of the species in contemporaneous Lybia (Ufnagl 1972) requires an urgent
re-assessment. In accordance with our hypothesis of
multiple introductions, NCA based on CR1 variation
suggested a second level of independent translocation
events into Europe from Clade 3-1, affecting both
continental Europe (Cataluñ a, Portugal) and the
Balearic Isl. (Ibiza vs. Mallorca and Cabrera)
(Fig. 2). Hap 8 was shared between Zeralda and
Jijel, Algeria, and the Parque Natural del Montnegre i
Corredor, Cataluñ a, Spain, suggestive of a direct
event of translocation between northern Algeria and
Cataluñ a. Haplotype distribution between the three
Balearic Islands suggested that common genets from
Ibiza, with an exclusive haplotype (Hap 14), had a
Phylogeography of the common genet
different translocation history from Mallorca and
Cabrera, where Hap 3—the most frequent on the
European continent—was highly dominant. Traces of
human passages have been reported in Mallorca since
8,000 years (Vignes 1999), but settlements started the
third millennium B.C., whereas Ibiza was not occupied until ca. 2,000 B.C. and was probably deserted
between ca. 1,300 and 700 B.C. (Bellard 1995). The
specificity of the peopling of Ibiza, linked to the great
influence of Carthaginians (Phoenician colony from
Tunisia) on the island, has been demonstrated by
genetic investigations on both humans and introduced
animals (Cosson et al. 2005; Tomàs et al. 2006). In
our case, common genets from Ibiza might have been
subject to an independent event of translocation,
directly originating from Maghreb, whereas the
Mallorcan population was possibly introduced later
from continental Europe. Remarkably, common genets from Ibiza have a significantly smaller body size
than other Mediterranean populations (Delibes 1977),
suggesting in situ adaptation to ecological constraints
of the island made possible by an earlier period of
arrival than in Mallorca and Cabrera (a founder effect
can also be envisaged). The omnipresence of Hap 3
in Cabrera, which is dominant in Mallorca, fit with
the recent introduction of the common genet from
Cabrera to Mallorca as a biocontrol agent against
rabbits (Delibes 1977).
Spain, with a higher genetic diversity and a
number of haplotypes falling within the 95% confidence limits of the rarefaction curve, may be
considered as the ‘translocation hotspot’ of European
common genets where several, independent introductions would have occurred at successive periods. On
the other hand, France and Portugal, both with lower
than expected haplotype diversities, are likely to
represent countries where the species arrived subsequently, either naturally or through human mediation
(coupled to a possible, former introduction in Portugal).
A main contribution of the Moor invaders to the
introduction and geographic expansion of the species
in Europe is a tempting hypothesis given that (i)
genetic diversity was higher in the regions occupied
by the Almohads from 1146 A.D. (southern and
eastern coasts of Spain and the Balearic Isl.; Morales
and Rofes 2008), (ii) the current species range fits
with the maximal progression of Moors’ settlements
into south-western France (Merdrignac and Mérienne
2003), and (iii) the only archaezoological records
537
linking the common genets to humans were found in
Algeria, Morocco, and one Almohad site in Europe.
Unfortunately, it remains impossible to reject or
validate hypotheses of historical translocations—by
Carthaginians, Greeks, Moors or else. Indeed, the
ideal approach for relating translocation events to
specific time periods would be to combine an
exhaustive screening of haplotype diversity to accurate estimates of divergence time. Dating of such
recent divergence is hampered by numerous factors,
including absence of monophyly due to retention of
ancestral polymorphism, overestimation of divergence time due to coalescence within the ancestral
population, and huge variance among loci due to the
stochastic nature of coalescence processes. All these
hurdles render the direct application of a molecular
clock to data on a single locus totally inappropriate
(Edwards and Beerli 2000; Arbogast et al. 2002). As
a consequence, we expect that at least some dating
events proposed for the recently introduced populations across the Mediterranean Basin are likely overestimated. It is thus especially surprising that those
estimates that predated by far evidences of human
crossing of the Strait of Gibraltar or fossil records
were not envisaged under the light of these expected
estimation biases and limitations (Guiller et al. 2001;
Libois et al. 2001; Michaux et al. 2003; Brandli et al.
2005; Cosson et al. 2005). Recent methods that
calculate simultaneously times of divergence, population sizes and migration rates, were designed to
circumvent this bias (e.g. Nielsen and Wakeley
2001). However, the low representation of Clade 1
in Maghreb (five individuals from N Algeria), did not
allow us to obtain reliable results from those methods
(data not shown).
In our study, Hap 3 was the commonest and most
widespread CR1 haplotype in Europe, possibly leading to the wrong impression that it may constitute the
ancestral haplotype from which northern Algerian
haplotypes were derived (see Fig. 2). Indeed, network
analysis makes the initial assumption that sampling
effort is distributed homogeneously across the study
area (Templeton 2004), the distribution of the sample
sizes directly affecting the distribution of haplotype
frequency. In our case, sampling activities were
unbalanced, with a sample set from northern Algeria
(Clade I) reduced to five individuals, contra 95 for
south-western Europe, where field work was easier to
plan. It is thus expected that a greater sample set from
123
538
Maghreb would lead to a much more accurate picture
of haplotype frequency and distribution in this region,
including higher haplotype frequencies that would
balance Hap 3 supremacy in Europe. Haplotype
network patterns may also be affected by multiple
introductions or human-mediated spread in Europe,
which can contribute to artefactually increase the
geographic coverage and frequency of a common
haplotype (suspected for Hap 3). Finally, although
the star-like pattern centered on Hap 3 may look
similar to what is observed in populations having
expanded from Pleistocene refugia (e.g. Michaux
et al. 2003), we find useful to mention that a similar
pattern has also been used to justify bottleneck events
related to recent introductions (e.g. Cosson et al.
2005).
NCA inferred contemporary processes of restricted
gene flow with isolation by distance/dispersal as
shaping the geographic structure of European common genets at two clade levels, but made no
inference of range expansion. On the contrary, our
demographic analyses supported a scenario of very
recent expansion within south-western Europe,
excluding the second haplogroup (Clade 3-2) that
possibly originated from a different translocation
event. Although we cannot completely rule out the
possibility of biases induced by accelerated rates of
evolution that may occur during invasive processes
(Lee 2002), the great number of haplotypes but the
low value of nucleotide diversity (Table 2), the
mismatch distribution and frequency spectrum (with
an excess of 1–2 steps mutations), the tests of
population growth (Table 5 and Fig. 4), and the starlike topology observed in the spanning network
(Fig. 2) all strongly suggest that the genetic structure
observed in European common genets results from
the very recent introductions of a reduced number of
individuals followed by rapid population expansion
(Avise 2000; Michaux et al. 2003). One basic
prediction underlying NCA is that older haplotypes
(interior clades) are geographically more widespread
than younger haplotypes (tips) (Templeton et al.
1995). In our case, NCA inference might have been
misled by the effect of the human-mediated translocation of the species, resulting for instance in the
widespread distribution of Hap 3 (interior) all over
Europe (Clades 1-2 and 2-1; Table 2 and Fig. 2). In
fact, the effect of a population bottleneck followed by
rapid geographic spread of a single haplotype has
123
P. Gaubert et al.
been reported as a possible cause of NCA failure in
detecting range expansion (Paulo et al. 2002). Importantly, restricted gene flow with isolation by distance
is one of the inferences that NCA may yield as false
positive at high frequencies (Panchal and Beaumont
2007; but see Templeton 2008). Although NCA gave
results consistent with prior expectations of a scenario of multiple introductions of the common genet
into Europe (see above), we cannot consider as
reliable the inference on restricted gene flow with
isolation by distance without the use of additional
loci and cross-validations through other methods
(Panchal and Beaumont 2007; Garrick et al. 2008).
Conclusion
We suggest that the exceptional combination of
several factors may have promoted the successful
spread of the common genet from Maghreb to
Europe, including multiple translocations and efficient propagule vectors such as human-driven
propagation across natural barriers (Ellstrand and
Schierenbeck 2000; Stepien et al. 2002; Roman and
Darling 2007), but also natural processes of colonization allowed by the wide ecological tolerance of the
species (Gaubert et al. 2008; Delibes and Gaubert in
press). It remains remarkable that supposedly independent introductions originated only from one out of
two of the mitochondrial lineages occurring in
Maghreb, and further investigations—using nuclear
markers and an extended sample from Maghreb—are
required to improve our understanding of the translocation processes and of this maternal geographic
imprint (Gaubert et al. in press). Remarkably, the
establishment of the common genet in south-western
Europe represents a historical, probably involuntary,
case of Pleistocene re-wilding (see Donlan 2005) in
Europe of a relictual North African lineage, which is
now potentially subject to habitat loss and introgression in its native range (see Masseti et al. 2008 for a
similar example in fallow deers from the Mediterranean
Basin). Indeed, Genetta genetta is an anthropochorous
species (i.e. described from introduced populations;
Gippoliti and Amori 2006) now protected in its whole
European range. Given the mitochondrial homogeneity
observed between European and northern Algerian
populations, we propose that the subspecies name
G. g. genetta (Linnaeus 1756), described from El
Phylogeography of the common genet
539
Pardo, near Madrid, Spain, be applied to all the
representatives of Clade I, with afra (Maghreb), balearica (Mallorca Isl.) and isabelae (Ibiza Isl.) as
synonyms. Possible introgression events with the
western African lineage (Clade IV) in Maghreb (involving afra) will have to be further investigated.
Acknowledgments We deeply thank the following people
for having contributed to the sampling effort: N. Aı̈t-Ameur,
A. Arrizabalaga Blanch, K. Ba, J.J. Bafaluy Zoriguel,
O. Berdion, M. Beucher, J.-C. Boisguerin, R. Bouhraoua,
M. Boukheroufa, F. Bourguemestre, E. Brandt, J.-M. Cassiède,
F. Catzeflis, M. Colyn, G. Coste, C. Crémière, J. Cuisin,
F. Cuzin, X. Domingo, G. Dominguez, M.-F. Faure,
F. Ferrandon, C. and P. Fournier, I.R. Fraile, A. Galat-Luong,
J. Garrigue, A. Gerbaud, M. Gouichiche, L. Granjon,
D. Guérineau, C. Gutierez, B. Hamou, E.H. Harley,
M. Hubert, A. Kitchener, E. Le Nuz, F. Léger, F. Llimona,
P. Lluch, J.V. Lopez Bao, F. Lopez-Giraldez, A. Loureiro,
E. Martinez Nevado, L. Matringe, J. Mayné, B. Mellier,
J.-D. Méric, A. Olivier, M.C. Otero, J.-P. Paillat, L. Parpal,
M. Pelven, S. Peres, A. Petit, L. Picco, J. Placer Lopez,
D. Portier, G. Pottier, J.-J. Ranouil, H. Rguibi, A.G.P. de
Santayana, J. Seon, P. Sierra, H. Sitek, P.J. Taylor, M. Tranier,
G.M. Vacas, J.-P. Vacher, G. Van Laere, P. Vercammen,
G. Veron, B. Vilatte, S. de Vries, S. Yepes, J. Zabala,
I. Zuberogoitia. The following institutions allowed us to access
their tissue banks: Breeding Centre for Endangered Arabian
Wildlife, Sharjah (United Arab Emirates); Consorcio de
Recuperacio de la Fauna, Illes Baleares (Spain); Instituto da
Conservação da Natureza, Porto (Portugal); Museu de
Granollers – Ciencies Naturals, Barcelona (Spain); Museo
Nacional de Ciencias Naturales, Madrid (Spain); Muséum
National d’Histoire Naturelle, Paris (France); Parc de
Collserola, Cataluña (Spain). We are grateful to Ana Piriz
and the whole staff of Laboratorio de Ecologia Molecular,
Estació n Bioló gica de Doñ ana, for lab work assistance and
fruitful discussions. Two anonymous reviewers provided useful
comments on the early version of the manuscript. Arnaud
Fontanet and Géraldine Veron played a significant role in fundraising the project. This work was funded by the European
Commission 6th PCRDT ‘‘EPISARS’’ (FP6-2003-SSP-2SARS; no. 51163).
Appendix I
Detailed list of the samples and sequences used in this study
Country (n)
Clade I
France (35)
Locale of collection
Sample n°
Type of sample
cytb
CR1
Vendée
Vendée
GgE151
GgE154
Deux-Sèvres
Deux-Sèvres
GgE01
GgE76
Tissue
Tissue
Hair
(a)
(a)
(a)
(a)
EF371655 (a)
(a)
Tissue
(a)
Loire-Atlantique
Maine-et-Loire
Charente-Maritime
Charente-Maritime
GgE153
GgT528
GgE142
GgE148
Tissue
Tissue
Hair
Hair
(a)
(a)
(a)
(a)
(a)
(a)
(a)
EF371664 (b)
(b)
Dordogne
Dordogne
Creuse
Indre
GgE43
GgE145
GgE149
GgE77’
Tissue
Tissue
Tissue
Hair
(a)
(a)
EF371617 (a)
(a)
(c)
(c)
EF371665 (c)
(a)
Allier
Cantal
Lot-et-Garonne
Lot-et-Garonne
GgT489
GgT488
GgE49
GgE61
Tissue
Tissue
Tissue
Tissue
(a)
(a)
(a)
(a)
(c)
(c)
(c)
(c)
Gironde
Gironde
GgE48
GgE63
Tissue
Tissue
(a)
(a)
(a)
(a)
Landes
Landes
Landes
GgE50
GgE51
GgE58
Tissue
Hair
Tissue
(a)
(a)
(a)
(c)
(c)
(c)
123
540
P. Gaubert et al.
Detailed list of the samples and sequences used in this study
Country (n)
Italy (1)
Spain (41)
123
Locale of collection
Sample n°
Type of sample
cytb
CR1
Pyrénées-Orientales
Pyrénées-Orientales
Hautes-Pyrénées
Hautes-Pyrénées
Pyrénées Atlantiques
Pyrénées Atlantiques
GgT294
(a)
(a)
(a)
(a)
(a)
(a)
EF371660 (d)
GgT115
GgT113
Hair
Hair
Hair
Hair
Hair
Hair
Aveyron
Gard
Hérault
Hérault
Bouches-du-Rhô ne
Bouches-du-Rhô ne
Bouches-du-Rhône
Alpes de Haute Provence
Piemont
GgE150
GgT293
Tissue
Tissue
(a)
(a)
EF371661 (r)
(c)
GgC9
GgT307
GgT133
GgT134
GgT292
GgT118
GgT540
DNA
(a)
(c)
Tissue
Hair
Hair
Hair
Hair
Tissue
(a)
(a)
(a)
(a)
(a)
EF371618 (a)
(c)
(d)
(d)
(c)
(d)
EF371666 (c)
Aragon
Aragon
Pais Basco
Pais Basco
Asturias
GgT95
GgT371
GgT362
GgT369
GgT373
Hair
Tissue
Hair
Hair
Tissue
Asturias
GgT396
Hair
(a)
(a)
(a)
(a)
(a)
(a)
(c)
(c)
(c)
(c)
(c)
(c)
Galicia
Galicia
Galicia
Cantabria
Extremadura
GgT388
GgT389
GgT446
GgT377
Tissue
Tissue
Tissue
Tissue
Hair
(a)
(a)
(a)
(a)
(a)
(c)
(c)
(c)
(c)
(c)
Tissue
Tissue
Tissue
Tissue
Hair
(a)
(a)
(a)
(a)
(e)
GgT407
GgE147
GgT114
(d)
(c)
(c)
(c)
(c)
Cataluña
Cataluña
Cataluña
Cataluña
Madrid
GgT530
GgT393
GgT449
GgT493
GgT498
GgT408
Madrid
GgT481
Tissue
EF371619 (a)
(a)
EF371667 (h)
(c)
Castilla-y-Leon
Castilla-y-Leon
Andalucia
GgT370
GgT391
GgE08
Tissue
Tissue
Hair
(a)
(a)
(a)
(h)
(h)
(c)
Andalucia
Andalucia
GgE09
GgE10
Hair
Hair
(a)
(a)
(c)
(c)
Andalucia
Andalucia
Andalucia
Andalucia
GgT378
GgT432
GgT438
GgT452
Hair
Tissue
Tissue
Hair
(a)
(a)
EF371620 (b)
(b)
EF371656 (i)
(c)
EF371668 (j)
EF371657 (k)
Andalucia
Mallorca Isl.
Mallorca Isl.
GgT459
GgT514
GgT515
Tissue
Blood
Blood
(a)
EF371621 (a)
(a)
(c)
EF371669 (c)
(c)
EF371662 (e)
EF371663 (f)
(g)
Phylogeography of the common genet
541
Detailed list of the samples and sequences used in this study
Country (n)
Portugal (18)
Algeria (5)
Clade II
Saudi Arabia (4)
Yemen (4)
Locale of collection
Sample n°
Type of sample
cytb
CR1
Mallorca Isl.
Cabrera Isl.
GgT516
GgT511
Blood
Blood
(a)
EF371622 (a)
EF371658 (l)
EF371670 (c)
Cabrera Isl.
Cabrera Isl.
GgT513
GgT517
Blood
Blood
(a)
(a)
(c)
(c)
Cabrera
Cabrera
Cabrera
Cabrera
GgT518
GgT523
GgT526
GgT527
Blood
Blood
Blood
Blood
(a)
(a)
(a)
(a)
(c)
(c)
(c)
(c)
GgT483
GgT484
GgT485
GgT487
GgT431
Tissue
Tissue
Tissue
Tissue
Tissue
(a)
(a)
(a)
EF371623 (a)
(a)
(m)
(m)
(m)
EF371671 (m)
(n)
Beja
Beja
Beja
GgT464
GgT467
GgT473
Tissue
Tissue
Tissue
(a)
(a)
(a)
(c)
(n)
(c)
Beja
Douro river
Douro river
Viana do Castelo
GgT480
GgT468
GgT470
GgT463
Tissue
Tissue
Tissue
Tissue
(a)
(a)
(a)
(a)
(c)
(c)
(c)
(c)
Viana do Castelo
Viana do Castelo
GgT475
GgT477
Tissue
Tissue
(a)
EF371624 (a)
(c)
EF371672 (c)
Braga
Braga
GgT465
GgT478
Tissue
Tissue
(a)
(a)
(c)
(c)
Braga
Braga
Minho
Porto
GgT479
GgT471
GgT469
GgT472
Tissue
Tissue
Tissue
Tissue
(a)
(a)
(a)
(a)
EF371659 (n)
(c)
(c)
(c)
–
–
Alger
Zeralda
Zeralda
Jijel
El Kala
GgT466
GgT476
GgE91
GgT316
GgT317
GgE88
GgT150
Tissue
Tissue
Hair
Tissue
Tissue
Hair
Hair
(a)
(a)
EF371626
EF371627
EF371628
EF371625
EF371629
(c)
(c)
EF371674
EF371675
EF371676
EF371673
EF371677
Djeddah
Captivity (Sharjah Desert Park,
United Arab Emirates)
Eastern Saudi Arabia
Taif
GgC81
GgT96
DNA
Hair
EF371644 (a)
EF371645 (a)
EF371692 (a)
EF371693 (b)
GgT502
GgT504
Hair
Hair
EF371646 (a)
EF371647 (a)
EF371694 (b)
EF371695 (b)
Taiz
Taiz
Taiz
Taiz
GgT101
GgT102
GgT103
GgT104
Hair
Hair
Hair
Hair
(b)
(b)
(b)
EF371648 (b)
(c)
(c)
(c)
EF371696 (c)
Ibiza
Ibiza
Ibiza
Ibiza
Beja
Isl.
Isl.
Isl.
Isl.
Isl.
Isl.
Isl.
Isl.
(a)
(a)
(a)
(a)
(a)
(o)
(p)
(g)
(g)
(q)
123
542
P. Gaubert et al.
Detailed list of the samples and sequences used in this study
Country (n)
Oman (1)
Locale of collection
Sample n° Type of sample
cytb
CR1
Salallah
GgC52
Tissue
EF371649 (c)
EF371697 (d)
GgE89
GgT380
GgE19
GgE107
Hair
Hair
Tissue
Hair
EF371639
EF371640
EF371642
EF371641
GgT98
Hair
EF371643 (c)
Batna
GgE92
Hair
Clade III
Namibia (2)
Okahandja
–
South Africa (3) KwaZulu-Natal
Captivity (Sharjah Desert Park,
United Arab Emirates)
Captivity (Sharjah Desert Park,
United Arab Emirates)
Clade IV
Algeria (2)
(a) EF371687
(a) EF371688
(b) EF371690
(b) EF371689
EF371632 (a)
(a)
(a)
(b)
(b)
EF371691 (b)
EF371680 (a)
Tlemcen
GgT539
Tissue
EF371633 (b)
Morocco (3)
Marrakech
Marrakech
Missoura
GgT120
GgT121
GgT543
Tissue
Dried ear (market)*
Tissue
EF371630 (c) EF371678 (c)
EF371651 (b) EF371699 (b)
EF371631 (b) EF371679 (f)
Tunisia (5)
Zagouan
Kroumiria
GgT161
GgT323
Tissue
Footpad (museum)*
EF371634 (d)
(d)
EF371682 (a)
(d)
Kroumiria
Kroumiria
GgT324
GgT325
Footpad (museum)*
Footpad (museum)*
(d)
(d)
(d)
(d)
Kroumiria
GgT326
Footpad (museum)*
Tentane
Emnal’here
Dakar
Dakar
GgE81
GgE16
GgE65
GgE75
Tissue
Tissue
Tissue
Tissue
EF371635 (e) EF371683 (e)
EF371636 (b) EF371684 (f)
EF371638 (b) EF371686 (g)
EF371637 (b) EF371685 (g)
Smoked tissue (market)*
EF371652 (b)
Mauritania (1)
Mali (1)
Senegal (2)
Benin (3)
Chad (2)
–
–
–
N’Djamena
N’Djamena
GgT298
GgT304
GgT306
GgT327
GgT328
(d)
EF371681 (b)
(a)
EF371700 (h)
Smoked tissue (market)* EF371653 (b) EF371701 (a)
Smoked tissue (market)* EF371654 (b) EF371702 (a)
Footpad (museum)*
(b)
EF371703 (h)
Footpad (museum)*
(b)
EF371704 (i)
GenBank accession numbers are given under cytb (cytochrome b) and CR1 (left domain of control region). Letters between
parentheses indicate identical haplotypes respective to each clade. Samples in bold were used for the combined cytb + CR1 analysis
(see Fig. 1)
* Samples extracted using ancient DNA protocol
Note: Clades I–IV correspond to Fig. 1
References
Alcover JA (1980) Note on the origin of the present mammalian fauna from the Balearic and Pityusic islands. Misc
Zool 6:141–149
Amigues S (1999) Les belettes de Tartessos. Anthropozool
29:55–64
Anderung C, Bouwman A, Persson P et al (2005) Prehistoric
contacts over the Straits of Gibraltar indicated by genetic
analysis of Iberian Bronze Age cattle. Proc Natl Acad Sci
102:8431–8435
Arbogast BS, Edwards SV, Wakeley J et al (2002) Estimating
divergence times from molecular data on phylogenetic
123
and population genetic timescales. Ann Rev Ecol Syst
33:707–740
Arnaiz-Villena A, Martinez-Laso J, Alonso-Garcia J (1999)
Iberia: population genetics, anthropology, and linguistics.
Hum Biol 71:725–743
Avise JC (2000) Phylogeography: the history and formation of
species. Harvard University Press, Cambridge
Bandelt H-J, Forster P, Rö hl A (1999) Median-joining networks for inferring intraspecific phylogenies. Mol Biol
Evol 16:37–48
Bate DMA (1903) On an extinct species of genet (Genetta
plesictoides, sp. n.) from the Pleistocene of Cyprus. Proc
Zool Soc Lond 2:121–124
Phylogeography of the common genet
Beja-Pereira A, Caramelli D, Lalueza-Fox C et al (2006) The
origin of European cattle: evidence from modern and
ancient DNA. Proc Natl Acad Sci USA 103:8113–8118
Bellard CG (1995) The first colonization of Ibiza and Formentera (Balearic Islands, Spain): some more islands out
of the stream? World Archaeol 26:442–445
Belon du Mans P (1557) Portraits d’oiseaux, animaux, serpens,
herbes, arbres, hommes et femmes d’Arabie & d’Egypte
observés par P. Belon. Le tout enrichi de quatrains pour
plus facile connaissance des oiseaux & autres portraits. G.
Cavellat, Paris
Bernatchez L, Dodson JJ, Boivin S (1989) Population bottlenecks: influence on mitochondrial DNA diversity and its
effect in coregine stock discrimination. J Fish Biol
35A:233–244
Boekschoeten GJ, Sondaar PY (1972) On the fossil Mammalia
of Cyprus. Proc K Ned Acad Van Wetenschappen Ser C
Biol Med Sci 75:306–338
Brandli L, Handley L-JL, Vogel P et al (2005) Evolutionary
history of the greater white-toothed shrew (Crocidura
russula) inferred from analysis of mtDNA, Y, and X
chromosome markers. Mol Phylogenet Evol 37:832–844
Casas MJ, Hagelberg E, Fregel R et al (2006) Human mitochondrial DNA diversity in an archaeological site in
al-Andalus: genetic impact of migrations from North Africa
in medieval Spain. Am J Phys Anthropol 131:539–551
Castella V, Ruedi M, Excoffier L et al (2000) Is the Gibralatr
Strait a barrier to gene flow for the bat Myotis myotis
(Chiroptera: Vespertillonidae)? Mol Ecol 9:1761–1772
Clement M, Posada D, Crandall KA (2000) TCS: a computer
program to estimate gene genealogies. Mol Ecol 9:1657–
1659
Colwell RK (2006) EstimateS: statistical estimation of species
richness and shared species from samples. The University
of Connecticut, Storrs
Colwell RK, Coddington JA (1994) Estimating terrestrial
biodiversity through extrapolation. Philos Trans R Soc B
345:101–118
Colwell RK, Mao CX, Chang J (2004) Interpolating, extrapolating, and
comparing
incidence-based
species
accumulation curves. Ecology 85:2717–2727
Cosson J-F, Hutterer R, Libois R et al (2005) Phylogeographical
footprints of the Strait of Gibraltar and quaternary fluctuations in the western Mediterranean: a case study with the
greater white-toothed shrew, Crocidura russula (Mammalia: Soricidae). Mol Ecol 14:1151–1162
Crawford-Cabral J (1981) The classification of the genets
(Carnivora, Viverridae, genus Genetta). Bol Soc Port
Ciênc Nat 20:97–114
Cymbron T, Loftus RT, Malheiro MI et al (1999) Mitochondrail sequence variation suggests an African influence
in Portuguese cattle. Proc R Soc Lond B 266:597–603
Delibes M (1977) Sobre las Ginetas de la Isla de Ibiza (Genetta
genetta isabelae n. ssp.). Doñana. Acta Vertebr 4:139–160
Delibes M, Gaubert P Genetta genetta Linneaus. In: Kingdon
JS, Hoffmann M (eds) The mammals of Africa, vol 5.
Carnivora, Pholidota, Perissodactyla. Academic Press,
Amsterdam (in press)
Dobson M (1998) Mammal distributions in the western
Mediterranean: the role of human intervention. Mamm
Rev 28:77–88
543
Dobson M, Wright A (2000) Faunal relationships and zoogeographical affinities of mammals in north-west Africa. J
Biogeogr 27:417–424
Donlan J (2005) Re-wilding North America. Nature 436:913–
914
Edwards SV, Beerli P (2000) Gene divergence, population
divergence, and the variance in coalescence time in phylogeographic studies. Evolution 54:1839–1854
Ellstrand NC, Schierenbeck KA (2000) Hybridization as a
stimulus for the evolution of invasiveness in plants? Proc
Natl Acad Sci 97:7043–7050
Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: an
integrated software package for population genetics data
analysis. Evol Bioinf Online 1:47–50
Formia A, Broderick AC, Glen F et al (2007) Genetic composition of the Ascension Island green turtle rookery
based on mitochondrial DNA: implications for sampling
and diversity. ESR 3:145–158
Fu YX (1997) Statistical tests of neutrality of mutations against
population growth, hitchhiking and background selection.
Genetics 147:915–925
Garrick RC, Dyer RJ, Beheregaray LB et al (2008) Babies and
bathwater: a comment on the premature obituary for
nested clade phylogeographical analysis. Mol Ecol
17:1401–1403
Gaubert P, Begg CM (2007) Re-assessed molecular phylogeny
and evolutionary scenario within genets (Carnivora,
Viverridae, Genettinae). Mol Phylogenet Evol 44:920–
927
Gaubert P, Cordeiro-Estrela P (2006) Phylogenetic systematics
and tempo of evolution of the Viverrinae (Mammalia,
Carnivora, Viverridae) within feliformians: implications
for faunal exchanges between Asia and Africa. Mol
Phylogenet Evol 41:266–278
Gaubert P, del Cerro I, Palomares F et al. Development and
characterization of 11 microsatellite loci in a historically
introduced carnivoran, the common genet (Genetta genetta). Mol Ecol Resour (in press)
Gaubert P, Fernandes CA, Bruford MW et al (2004a) Genets
(Carnivora, Viverridae) in Africa: an evolutionary synthesis based on cytochrome b sequences and
morphological characters. Biol J Linn Soc 81:589–610
Gaubert P, Tranier M, Delmas A-S et al (2004b) First molecular evidence for reassessing phylogenetic affinities
between genets (Genetta) and the enigmatic genet-like
taxa Osbornictis, Poiana and Prionodon (Carnivora,
Viverridae). Zool Scr 33:117–129
Gaubert P, Veron G, Colyn M et al (2002) A reassessment of
the distributional range of the rare Genetta johnstoni
(Viverridae, Carnivora), with some newly discovered
specimens. Mamm Rev 32:132–144
Gaubert P, Jiguet F, Bayle P et al (2008) Has the common
genet (Genetta genetta) spread into south-eastern France
and Italy? Ital J Zool 75:43–57
Geraads D (1997) Carnivores du Pliocène terminal de Ahl al
Oughlam (Casablanca, Maroc). Geobios 30:127–164
Gippoliti S, Amori G (2006) Ancient introductions of mammals in the Mediterranean Basin and their implications for
conservation. Mamm Rev 36:37–48
Gsell S (1913) Histoire ancienne de l’Afrique du Nord.
Hachette, Paris
123
544
Guiller A, Coutellec-Vreto MA, Madec L et al (2001) Evolutionary history of the land snail Helix aspersa in the
Western Mediterranean: preliminary results inferred from
mitochondrial DNA sequences. Mol Ecol 10:81–87
Guindon S, Gascuel O (2003) A simple, fast and accurate
method to estimate large phylogenies by maximum-likelihood. Syst Biol 52:696–704
Hall TA (1999) BioEdit: a user-friendly biological sequence
alignment and analysis program for Windows 95/98/NT.
Nucleic Acids Symp 41:95–98
Harpending HC (1994) Signature of ancient population growth
in a low-resolution mitochondrial DNA mismatch distribution. Hum Biol 66:591–601
Harris DJ, Carranza S, Arnold EN et al (2002) Complex biogeographical distribution of genetic variation within
Podarcis wall lizards across the Strait of Gibraltar.
J Biogeogr 29:1257–1262
Hasegawa M, Kishino H, Yano T-A (1985) Dating of the
human-ape splitting by a molecular clock of mitochondrial DNA. J Mol Evol 22:160–174
Hayes K, Barry S Are there any consistent predictors of
invasion success? Biol Invasions (in press)
Hendey QB (1974) The late Cenozoic Carnivora of the
southwestern Cape province. Ann S Afr Mus 63:1–369
Jolly D, Harrison SP, Damnati B et al (1998) Simulated climate
and biomes of Africa during the late Quaternary: comparison with pollen and lake status data. Quat Sci Rev
17:629–657
Juste J, Ibañ ez C, Muñ oz J et al (2004) Mitochondrial phylogeography of the long-eared bats (Plecotus) in the
Mediterranean Palaearctic and Atlantic Islands. Mol
Phylogenet Evol 31:1114–1126
Keane TM, Naughton TJ, McInerney JO (2004) ModelGenerator: amino acid and nucleotide substitution model
selection. National University of Ireland, Maynooth
Kingdon J (1977) East African Mammals: an atlas of evolution
in Africa. Academic Press, London
Kock D (1983) Identifizierung der Palastina-Genetten von J.
Aharoni als Vormela peregusna (Guldenstaedt, 1770). Z
Saugetierkd 48:381–383
Kumar S, Tamura K, Nei M (2004) MEGA3: integrated software for molecular evolutionary genetics analysis and
sequence alignment. Brief Bioinf 5:150–163
Kuper R, Kröpelin S (2006) Climate-controlled Holocene
occupation in the Sahara: motor of Africa’s evolution.
Science 313:803–807
Kurten B (2007) Pleistocene mammals of Europe. Transaction
Pub, New Brunswick, London
Lee CE (2002) Evolutionary genetics of invasive species.
TREE 17:386–391
Lessa EP, Cook JA, Patton JL (2003) Genetic footprints of
demographic expansion in North America, but not
Amazonia, during the late Quaternary. Proc Natl Acad Sci
100:10331–10334
Libois RM, Michaux JR, Ramalhinho MG et al (2001) On the
origin and systematics of the northern African wood
mouse (Apodemus sylvaticus) populations: a comparative
study of mtDNA restriction patterns. Can J Zool 79:1503–
1511
Marra AC (2005) Pleistocene mammals of Mediterranean
islands. Quat Int 129:5–14
123
P. Gaubert et al.
Masseti M, Pecchioli E, Vernesi C (2008) Phylogeography of
the last surviving populations of Rhodian and Anatolian
fallow deer (Dama dama dama L., 1758). Biol J Linn Soc
93:835–844
Merdrignac B, Mérienne P (2003) Le monde au Moyen-Â ge.
Editions Ouest-France, Rennes, France
Michaux JR, Magnanou E, Paradis E et al (2003) Mitochondrial phylogeography of the woodmouse (Apodemus
sylvaticus) in the Western Palearctic region. Mol Ecol
12:685–697
Modolo L, Salzburger W, Martin RD (2005) Phylogeography
of Barbary macaques (Macaca sylvanus) and the origin
of the Gibraltar colony. Proc Natl Acad Sci 102:7392–
7397
Montoya P, Alcala L, Morales J (2001) First find of a
Viverridae (Carnivora, Mammalia) in the Upper Miocene
of the Teruel Basin (Spain). Bol R Soc Esp Hist Nat Geol
96:101–109
Morales A (1994) Earliest genets in Europe. Nature 370:512–
513
Morales A, Rofes J (2008) Early evidence for the Algerian
hedgehog in Europe. J Zool 274:9–12
Muñ oz-Fuentes V, Green AJ, Sorenson MD et al (2006) The
ruddy duck Oxyura jamaicensis in Europe: natural colonization or human introduction? Mol Ecol 15:1441–1453
Nielsen R, Wakeley J (2001) Distinguishing migration from
isolation: a Markov chain Monte Carlo approach. Genetics 158:885–896
Osborn DJ, Osbornova J (1998) The mammals of ancient
Egypt. Aris & Phillips Ltd, Warminster
Ouchaou B, Amani F (2002) Les carnivores des gisements
néolithiques et protohistoriques du nord du Maroc.
Quaternaire 13:79–87
Palomares F Herpestes ichneumon. In: Kingdon JS, Hoffmann
M (eds) The mammals of Africa, vol 5. Carnivores,
Pangolins, Equids and Rhinos. Academic Press, Amsterdam (in press)
Palomares F, Godoy JA, Piriz A et al (2002) Faecal genetic
analysis to determine the presence and distribution of
elusive carnivores: design and feasibility for the Iberian
lynx. Mol Ecol 11:2171–2182
Panchal M, Beaumont MA (2007) The automation and evaluation of nested clade phylogeographic analysis.
Evolution 61:1466–1480
Paulo OS, Jordan WC, Bruford MW et al (2002) Using nested
clade analysis to assess the history of colonization and the
persistence of populations of an Iberian lizard. Mol Ecol
11:809–819
Perrot A-M (1820) Collection historique des ordres de chevalerie civils et militaires, existant chez les différents
peuples du monde, suivie d’un tableau chronologique des
ordres éteints.. Aimé André, Paris
Petit RJ (2008a) The coup de grace for the nested clade phylogeographic analysis? Mol Ecol 17:516–518
Petit RJ (2008b) On the falsifiability of the nested clade phylogeographic analysis method. Mol Ecol 17:1404–1404
Posada D, Buckley TR (2004) Empirical problems of the
hierarchical likelihood ratio test for model selection. Syst
Biol 53:949–962
Posada D, Templeton A (2006) GeoDis 2.5. Universidad de
Vigo, Spain
Phylogeography of the common genet
Ramon-Laca L (2003) The introduction of cultivated citrus to
Europe via northern Africa and the Iberian Peninsula.
Econ Bot 57:502–514
Rannala B, Yang Z (2003) Bayes estimation of species divergence times and ancestral population sizes using DNA
sequences from multiple loci. Genetics 164:1645–1656
Rodrı́guez-Ariza M, Moya E (2005) On the origin and
domestication of Olea europaea L. (olive) in Andalucı́a,
Spain, based on the biogeographical distribution of its
finds. Veg Hist Arch 14:551–561
Rogers AR, Harpending H (1992) Population growth makes
waves in the distribution of pairwise genetic differences.
Mol Biol Evol 9:552–569
Roman J, Darling JA (2007) Paradox lost: genetic diversity and
the success of aquatic invasions. TREE 22:454–464
Romer AS (1928) Pleistocene mammals of Algeria: fauna of
the Paleolithic station of Mechta el Arbi. Logan Mus Bull
2:79–163
Ronquist F, Huelsenbeck JP, van der Mark P (2005) MrBayes
3.1 manual. Florida State University, Tallahassee
Rozas J, Sánchez-DelBarrio JC, Messegyer X et al (2003)
DnaSP, DNA polymorphism analyses by the coalescent
and other methods. Bioinformatics 19:2496–2497
Schauenberg P (1966) La Genette vulgaire (Genetta genetta
L.), répartition géographique en Europe. Mammalia
30:371–396
Schlawe L (1980) Zur geographischen Verbreitung der
Ginsterkatzen, Gattung Genetta G. CUVIER, 1816
(Mammalia, Carnivora, Viverridae). Faun Abh Mus
Tierkde Dresden 7:147–161
Schlawe L (1981) Material, Fundorte, Text- und Bildquellen
als Grundlagen für eine Artenliste zur Revision der
Gattung Genetta G. CUVIER, 1816. Zool Abh Mus
Tierkde Dresden 37:85–182
Scozzari R, Cruciani F, Pangrazio A et al (2001) Human Ychromosome variation in the western Mediterranean area:
implications for the peopling of the region. Hum Immunol
62:871–884
Simberloff D (1972) Properties of the rarefaction diversity
measurement. Am Nat 106:414–418
Stepien CA, Taylor CD, Dabrowska KA (2002) Genetic variability and phylogeographical patterns of a nonindigenous
species invasion: a comparison of exotic vs. native zebra
and quagga mussel populations. J Evol Biol 15:314–328
Tajima F (1989) The effect of change in population size on
DNA polymorphism. Genetics 123:597–601
Tajima F (1993) Simple method for testing the molecular
evolutionary clock hypothesis. Genetics 135:599–607
Tamura K, Nei M (1993) Estimation of the number of nucleotide substitutions in the control region of mitochondrial
DNA in humans and chimpanzees. Mol Biol Evol 10:512–
526
Templeton AR (1998) Nested clade analyses of phylogeographic data: testing hypotheses about gene flow and
population history. Mol Ecol 7:381–397
Templeton AR (2004) Statistical phylogeography: methods of
evaluating and minimizing inference errors. Mol Ecol
13:789–809
Templeton AR (2008) Nested clade analysis: an extensively
validated method for strong phylogeographic inference.
Mol Ecol 17:1877–1880
545
Templeton AR, Sing CF (1993) A cladistic analysis of phenotypic associations with haplotypes inferred from
restriction endonuclease mapping. IV. Nested analyses
with cladogram uncertainty and recombination. Genetics
134:659–669
Templeton AR, Boerwinkle E, Sing CF (1987) A cladistic
analysis of phenotypic associations with haplotypes
inferred from restriction endonuclease mapping. I. Basic
theory and an analysis of alcohol dehydrogenase activity
in Drosophila. Genetics 117:343–351
Templeton AR, Crandall KA, Sing CF (1992) A cladistic
analysis of phenotypic associations with haplotypes
inferred from restriction endonuclease mapping and DNA
sequence data. III. Cladogram estimation. Genetics
132:619–633
Templeton AR, Routman E, Phillips CA (1995) Separating
population structure from population history: a cladistic
analysis of the geographical distribution of mitochondrial
DNA haplotypes in the tiger salamander, Ambystoma
tigrinum. Genetics 140:767–782
Thomas H (1979) Le rô le de barrière écologique de la ceinture
saharo-arabique au Miocène: arguments paléontologiques.
Bull Mus Nat Hist Nat 4-è s Sect C 1:127–135
Thomas H, Bernor R, Jaeger J-J (1982) Origines du peuplement mammalien en Afrique du Nord durant le Miocène
terminal. Geobios 15:283–297
Tomàs C, Jiménez G, Picornell A et al (2006) Differential
maternal and paternal contributions to the genetic pool of
Ibiza Island, Balearic Archipelago. Am J Phys Anthropol
129:268–278
Tristram HB (1866) Report on the mammals of Palestine. Proc
Zool Soc Lond 1866:84–93
Tristram HB (1884) The survey of western Palestine. The
fauna and flora of Palestine. The Committee of the
Palestine Exploration Fund, London
Ufnagl E (1972) Lybian mammals. The Olander Press, New
York
van der Made J, Morales J, Montoya P (2006) Late Miocene
turnover in the Spanish mammal record in relation to
palaeoclimate and the Messinian Salinity Crisis. Pal Pal
Pal 238:228–246
Vigne J-D (1992) Zooarchaeology and the biogeographical
history of the mammals of Corsica and Sardinia, since the
last Ice Age. Mamm Rev 22:87–96
Vignes J-D (1999) The large ‘‘true’’ Mediterranean islands as
a model for the HoIocene human impact on the European vertebrate fauna? Recent data and new reflections.
In: Benecke N (ed) The Holocene history of the European vertebrate fauna. Modern aspects of research.
Workshop, Berlin, Germany, pp 295–322. 6–9 April
1998
Wenner MW (1980) The Arab/Muslim presence in medieval
central Europe. Int J M East Stud 12:59–79
Werdelin L (2003) Mio-Pliocene carnivora from Lothagam,
Kenya. In: Leakey MG, Harris JM (eds) Lothagam: the
dawn of humanity in Africa. Columbia University Press,
New York
Wolsan M, Morlo MO (1997) The status of ‘Plesictis’ croizeti,
‘Plesictis’ gracilis and ‘Lutra’ minor: synonyms of the
early Miocene viverrid Herpestides antiquus (Mammalia,
Carnivora). Bull Nat Hist Mus Geol 53:1–9
123
546
Yang Z (2002) MCMCcoal. Markov Chain Monte Carlo
Coalescent program. Department of Biology, University
College, London
Yang Z, Yoder AD (1999) Estimation of the transition/transversion rate bias and species sampling. J Mol Evol
48:274–283
123
P. Gaubert et al.
Young ND, Healy J (2003) GapCoder automates the use of
indel characters in phylogenetic analysis. BMC Bioinfo
4:6