Howling from the past: historical

rspb.royalsocietypublishing.org phylogeography and diversity losses

in European grey wolves
Christophe Dufresnes1,4, Christian Miquel5, Nadège Remollino1,
Research François Biollaz1,6, Nicolas Salamin2,3, Pierre Taberlet5,† and Luca Fumagalli1,†
Cite this article: Dufresnes C, Miquel C, 1
Laboratory for Conservation Biology, Department of Ecology and Evolution, 2Department of Ecology and
Remollino N, Biollaz F, Salamin N, Taberlet P, Evolution, and 3Department of Computational Biology University of Lausanne, Biophore Building, CH-1015
Fumagalli L. 2018 Howling from the past: Lausanne, Switzerland
4
Department of Animal and Plant Sciences, University of Sheffield, Alfred Denny Building, Western Bank,
historical phylogeography and diversity losses
Sheffield S10 2TN, UK
in European grey wolves. Proc. R. Soc. B 285: 5
Laboratoire d’Écologie Alpine (LECA), UMR5553, BP53, 38041 Grenoble, Cedex 9, France
6
20181148. Route Pra de Louetse 32, 1968 Mase, Switzerland
http://dx.doi.org/10.1098/rspb.2018.1148 LF, 0000-0002-6648-2570

Genetic bottlenecks resulting from human-induced population declines make

alarming symbols for the irreversible loss of our natural legacy worldwide. The
Received: 24 May 2018 grey wolf (Canis lupus) is an iconic example of extreme declines driven by
Accepted: 6 July 2018 anthropogenic factors. Here, we assessed the genetic signatures of 150 years
of wolf persecution throughout the Western Palaearctic by high-throughput
mitochondrial DNA sequencing of historical specimens in an unprecedented
spatio-temporal framework. Despite Late Pleistocene bottlenecks, we show
that historical genetic variation had remained high throughout Europe until
Subject Category: the last several hundred years. In Western Europe, where wolves nearly got
Evolution fully exterminated, diversity dramatically collapsed at the turn of the twentieth
century and recolonization from few homogeneous relict populations induced
Subject Areas: drastic shifts of genetic composition. By contrast, little genetic displacement
evolution, genetics, ecology and steady levels of diversity were maintained in Eastern European regions,
where human persecution had lesser effects on wolf demography. By compar-
ing prehistoric, historic and modern patterns of genetic diversity, our study
Keywords:
hence traces the timeframe and the active human role in the decline of the
conservation, Canis lupus, genetic diversity, grey wolf, an emblematic yet controversial animal which symbolizes the
historical DNA, human-driven declines, complex relationship between human societies and nature conservation.
museum specimens

Author for correspondence: 1. Introduction

Luca Fumagalli Genetic variation is the keystone of the adaptive potential of species and an
e-mail: luca.fumagalli@unil.ch important component of our natural legacy [1,2]. Genetic bottlenecks (i.e. the
loss of genetic variants when populations contract) are one hidden consequence
of the biodiversity crisis worldwide. When conservation efforts pay off, species
recovery and recolonization of former ranges may also reshape the genetic com-
position of populations, with potential detrimental effects such as migration load
(i.e. reduced fitness of migrants due to maladaptation [3]) and expansion load (i.e.
the fixation by drift of slightly deleterious mutations during demographic expan-
sions [4]). Genetically impoverished populations may thus be more susceptible to
extrinsic stresses [5]. Understanding the genetic impacts of past and present popu-
lation dynamics in changing human societies and environments is thus a central
†
These authors contributed equally to this aspect of conservation research. Moreover, quantifying the irreversible losses of
study. diversity can raise awareness among the institutional stakeholders and the
public opinion towards conservation issues.
Electronic supplementary material is available A wide array of studies has inferred genetic bottlenecks from demographic
analyses based on contemporary samples [6–9], yet with the inherent difficulty
online at http://dx.doi.org/10.6084/m9.
to distinguish between recent, anthropogenic losses of diversity versus long-term
figshare.c.4166597 bottlenecks initiated earlier during the Quaternary. In addition, human-mediated

& 2018 The Author(s) Published by the Royal Society. All rights reserved.
extinction events might play a key role in the spatial (re)structur-
ing of biodiversity, with important conservation consequences
[10]. Addressing these issues requires access to initial levels
and distribution of genetic diversity from historical collec-
tions, which is challenging due to their low DNA content
and quality, and often limited availability. Hence, so far
most studies analysing historical DNA in declining species
have remained restricted to narrow geographical regions
and time scales. Yet comprehensive spatio-temporal frame-
works are necessary to attribute a role of anthropogenic
pressures and historical declines in shaping present patterns
of diversity.
Few animals exemplify the contrasted relationship between
society and nature better than the grey wolf (Canis lupus).
Formerly the most widely distributed mammal on Earth, its
persecution up until the twentieth century has led to extreme
declines and extinctions over large parts of the Palaearctic and
Nearctic [11,12]. However, wolves from the Western Palaearctic
went through contrasted histories. The persecution, mainly
due to competition for food resources and cultural prejudice,
was extremely severe in Western Europe, involving hunting/
trapping programmes initiated as early as the Middle Ages
and ultimately leading to the extirpation from England (six-
teenth century), Scotland (1684), Ireland (1786), Denmark
(1772), The Netherlands (1869), as well as Germany, Switzerland
and France (late nineteenth/early twentieth century). The Italian
population (C. l. italicus) only persisted in narrow Apennine
pockets with as low as hundred surviving individuals [13],
while the Iberian population (C. l. signatus) had its range
shrunk down to two areas in the southern and northwest parts
of the peninsula [11,12]. By contrast, the species faced a more
relaxed pressure in Central and Eastern Europe. Despite some
periods of persecution leading to regional population declines
and range contractions, wolves were never extirpated from
Balkanic and Baltic countries. Besides, wolves could easily
migrate from the Russian empire, where hunting programmes
remained relatively moderate and populations maintained sus-
tainable numbers [12]. For instance, the Finnish wolf population
has been under perfusion from the adjacent Russian Karelia.
As for other large carnivores, cultural changes combined
with strong legislative acts are now promoting recovery and
expansion across former ranges. The wolf is legally protected
in most countries and has recolonized the Alpine Arc from
the remnant Italian population. In the east, numbers rapidly
increased over the last two decades in the Baltic States,
Finland and Poland [14,15]. Wolves are now spreading
further west in Sweden and Germany, and are regularly
observed in Austria [14 –16].
The important economic and conservation issues associated
with wolf dynamics have stressed the need for programmes of
genetic monitoring in most European countries. Modern pat-
terns of diversity have thus received considerable attention
[15,17,18], some of them consistent with local genetic bottle-
necks [15]. By contrast, despite a few scattered cases (limited
to Fennoscandia [19,20]), the genetics of European wolf
populations prior and during their decline has never been
addressed. Based on genomic analyses of contemporary speci-
mens, several studies reported ancient (Late Pleistocene) loss of
genetic variation and long-term demographic declines pre-
dating modern times by tens of thousands of years [21–23],
raising questions regarding the actual amount of diversity
remaining in wolf populations prior to their recent extirpation,
and, by extension, the role of human persecution in driving the
detected genetic bottlenecks [24]. Dissecting this role thus 2
requires a direct measure of the patterns of genetic diversity
rspb.royalsocietypublishing.org
during human persecutions, over a comprehensive geographical
and temporal scale, in order to draw general conclusions.
Here, we provide such comparative spatio-temporal frame-
work with an unprecedented resolution by investigating
changes in mitochondrial DNA (mtDNA) diversity across
Western Palaearctic wolves over two centuries of history. The
use of mtDNA, and particularly the hypervariable control
region (CR), allows to capture high levels of polymorphism
over short sequences (see Material and methods), a prerequisite
for informative analyses of degraded DNA from museum-
Proc. R. Soc. B 285: 20181148
prepared specimens. Importantly, the CR has been massively
used in contemporary wolf population genetics from all over
the world (reviewed in [18]), enabling direct comparisons. To
this end, we conducted a phylogeographical analysis of past
and present wolf populations, combining more than 150
museum samples covering an entire zoogeographical region,
with metadata from 20 regional studies. We aimed (i) to
quantify the original amount of diversity, (ii) to quantify the
rate of diversity losses during nineteenth- and twentieth-century
declines and (iii) to assess changes in genetic composition result-
ing from extinction and recolonization events. In particular, we
tested whether the contrasted persecution regimes left differen-
tial genetic signatures between Western and Central/Eastern
Europe.
2. Results
We identified 42 haplotypes of the hypervariable mtDNA CR
(validated by cloning and/or high-throughput sequencing;
see Material and methods) among 155 historical specimens
(complemented by an additional 22 published from Finland;
[20]), covering the nineteenth to early twentieth centuries
(figure 1; electronic supplementary material, figure S1). About
half have since vanished: only 22 of these haplotypes remain
among 1814 modern wolves analysed in Europe and adjacent
regions (figure 1; electronic supplementary material, table S1).
An additional 10 haplotypes found in modern wolves were
not present in our historical samples. Moreover, only historical
haplotypes H3 and H45 matched two of the 19 haplotypes of
prehistoric wolves from the Late Pleistocene and Early Holocene
published in [25] (57 bp, isolated from 24 wolf remains found
across the Western Palaearctic; electronic supplementary
material, figure S2).
Considering the entire Western Palaearctic, genetic diversity
estimates, i.e. haplotype diversity (Hd) and nucleotide diversity
(p), were overall higher in historical (Hd ¼ 0.94 and p ¼ 0.017,
n ¼ 177) than contemporary samples (Hd ¼ 0.78 and p ¼
0.015, n ¼ 1814). Resampling the modern estimates shows
that these differences are statistically robust and independent
from the uneven sample sizes (table 1; electronic supplemen-
tary material, figure S3). The trend was further confirmed by
analysing each population separately (figure 2; electronic
supplementary material, table S1).
Modern wolf populations are highly structured ( pairwise
Fst ¼ 0.50 + 0.30; electronic supplementary material, table S2
and figure S4), in comparison with historical ones (pairwise
Fst ¼ 0.24 + 0.17; electronic supplementary material, table
S2 and figure S3; paired Wilcoxon signed-rank test between
modern and historical Fst: p , 0.001, n ¼ 45 comparisons).
Indeed, most of the major mitochondrial haplogroups used
to historically coexist all over Europe (figure 1).
 (a) 3
historical

rspb.royalsocietypublishing.org
Proc. R. Soc. B 285: 20181148
(b)
modern

range
persistent

recolonized

data:
haplotype
frequency
haplotype
occurrence

Figure 1. Mitochondrial haplotype network with the distribution of haplogroups in (a) historical and (b) modern times. Haplotypes (number: haplotype code) only
found in our historical dataset are shown in shaded circles; historical haplotypes also found in modern datasets are shown in plain circles. The sizes of circles
illustrate the relative frequency of each haplotype in the historical dataset. Haplotypes only found in modern datasets from the Western Palaearctic are shown
by squares. Colours represent the main haplogroups composed of closely related haplotypes. Correspondence with the phylogeographical study by Pilot et al.
[17] is provided in electronic supplementary material, figure S1. The historical map plots the distribution of haplogroups based on our dataset (circles with
white strokes, size being proportional to sample sizes; n ¼ 155) and from the data of [20] for Finland (circle with black stroke; n ¼ 22), covering the
1795– 1945 period. Circles annotated with official country codes (see electronic supplementary material, figure S4 for details) indicate specimens where geographical
information was limited to the country of origin. At this time, wolves used to be distributed all over continental Europe. The modern map plots the current
distribution of haplogroups based on published haplotype frequency data from 20 studies using comparable mtDNA sequences (n ¼ 1814; large circles with
black strokes; see Material and methods). Regions where only the haplotype occurrence was reported (but not their relative frequencies) are also shown
(small circles with white strokes). The current range distinguishes areas where wolves always persisted (plain lines and dashed areas) from where the species
became extinct and which have been recently recolonized (dashed lines).
Table 1. Historical and modern diversity indices (haplotype diversity Hd and nucleotide diversity p) computed from all Western Palaearctic wolves and 4
separately for Western and Eastern Europe.

rspb.royalsocietypublishing.org
Hd p

historical modern (95% CI) historical modern (95% CI)

overall 0.94 0.78 (0.73– 0.83) 0.017 0.015 (0.014 –0.016)

West 0.90 0.28 (0.17– 0.39) 0.016 0.006 (0.004 –0.008)
East 0.90 0.85 (0.81– 0.88) 0.015 0.016 (0.015 –0.017)
95% confidence intervals (CIs) after 10 000 resampling of the modern dataset (to the sample size of the corresponding historical one) are given.

Proc. R. Soc. B 285: 20181148

(a) all West East we independently analysed (see next sections). Note that
n.s. Scandinavian samples (n ¼ 5 historical and 18 modern
* *
1.0 samples) were not attributed to either region (see Discussion).
0.8

0.6 (a) Wolves from Western Europe

Hd

0.4 Historical specimens from Western Europe featured highly

diverse mtDNA variation over the entire region (Hd ¼ 0.90
0.2 and p ¼ 0.016) (table 1; electronic supplementary material,
table S1). Modern estimates are significantly lower in
0
comparison (Hd ¼ 0.28 and p ¼ 0.006; table 1; electronic sup-
historical

modern

historical

modern

historical

modern

plementary material, figure S3). About three-quarters (17 out

of 22) of the original mitotypes vanished from the region.
This is well reflected by population-based comparisons,
which also statistically support a reduction of Hd in modern
(b) 1.0 West Western European wolves (figure 2a; electronic supplementary
East material, table S1). For instance, the Italian wolf population is
0.8
nowadays characterized by a largely predominant private hap-
0.6 lotype (H18) and a second rare haplotype (H11), but at least 10
Hd

different ones used to occur throughout the country. The same

0.4 is true for France and Switzerland, where 15 and 8 haplotypes
0.2 were, respectively, found, before these populations became
extinct. The genetic bottleneck was less drastic in Iberia,
0 where several CR haplotypes have been maintained (electronic
1800 1820 1840 1860 1880 1900 1920 1940
modern

supplementary material, table S1). Analysis of haplotype

year
Figure 2. (a) Comparisons of region-based Hd between historical
(1795 – 1945) and modern times (after 1990) and (b) haplotype diversity
(Hd) over time in Western (blue) and Eastern Europe (red). (a) Comparison
between historical and modern estimates were performed with non-
parametric ANOVAs (10 000 permutations) and were significant for the
whole dataset (n ¼ 10 and 20 historical and modern estimates, respectively;
p ¼ 0.001) and for Western Europe (n ¼ 4 and 5 historical and modern
estimates, respectively; p ¼ 0.016) but not for Eastern Europe (n ¼ 5 and
14 historical and modern estimates, respectively; p ¼ 0.095). On the boxplot:
*p , 0.05, n.s., non-significant. More stringent pairwise comparisons (non-
parametric paired Wilcoxon signed-rank test) were only significant for the
overall dataset (n ¼ 10 comparisons, p ¼ 0.0020) but not for Western
(n ¼ 4 comparisons; p ¼ 0.12) and Eastern Europe (n ¼ 5 comparisons,
p ¼ 0.062). (b) Analyses are based on 30 years sliding windows, compu-
ted only for n  5, including samples with accurate date information only
(n ¼ 110). Western and Eastern areas are also distinguished by circles and
triangles, respectively, and temporal series by empty (historical) and plain
(modern) symbols.
These trends vary between Western Europe (n ¼ 85 his-
torical and 865 modern samples) and Central/Eastern
Europe (n ¼ 84 historical and 931 modern samples), which
diversity (Hd) through time revealed a dramatic collapse at
the turn of the twentieth century (from 0.9 down to 0.3), reach-
ing the extremely low diversity levels recorded across
contemporary individuals (figure 2b).
Population differentiation across Western Europe appears
lower historically (average pairwise Fst ¼ 0.22 + 0.17; elec-
tronic supplementary material, table S2 and figure S4) than at
present (average pairwise Fst ¼ 0.53 + 0.43; electronic sup-
plementary material, table S2); however, this difference was
not significant, potentially due to the little number of compari-
sons (paired Wilcoxon signed-rank test between modern and
historical Fst: p ¼ 0.40, n ¼ 6 comparisons). For instance, the
modern C. l. italicus mtDNA private lineage (H18) used to
occur in northern France. Historical Iberian wolves carried
alleles that are not specific to modern C. l. signatus, but were
also present elsewhere (e.g. H1 in France and Switzerland).
Accordingly, pairwise Fst among the historical and modern
populations are not correlated (Mantel test, p ¼ 0.29), indicat-
ing that the genetic relationships between populations have
changed.
Moreover, we report strong shifts in the genetic composition
of these populations. Modern Western European wolves are
genetically different from the ones originally present (France:
Fst ¼ 0.60; Switzerland: Fst ¼ 0.70; Italy: Fst ¼ 0.73; electronic
supplementary material, table S1 and figure S3). The displace-
ment was moderate in Spain, where several C. l. signatus
haplotypes persisted (Fst ¼ 0.08).
(b) Wolves from Eastern and Central Europe
The mitochondrial diversity of Eastern and Central European
wolves remained high throughout the past centuries (Hd ¼
0.90 and p ¼ 0.015; figure 2). Modern estimates were margin-
ally lower (Hd ¼ 0.85) or similar (p ¼ 0.016) (table 1; electronic
supplementary material, figure S3). Accordingly, population-
based Hd were generally lower in modern than historical popu-
lations, but the difference was not significant (figure 2;
electronic supplementary material, table S1).
The patterns of differentiation between populations
remained stable between historical and modern datasets,
with similar estimates (average pairwise Fst ¼ 0.30 + 0.12
prior 1945 versus Fst ¼ 0.32 + 0.22 among modern popu-
lations; paired Wilcoxon signed-rank test: p ¼ 0.19, n ¼ 10
comparisons Wilcoxon; electronic supplementary material,
table S2 and figure S4) and relationships (significant correlation
between historical versus modern pairwise Fst, p ¼ 0.016).
Accordingly, little shifts in genetic composition were
generally observed (Poland: Fst ¼ 0.14; Turkey: Fst ¼ 0.21;
Caucasus: Fst ¼ 0.02; Russia: Fst ¼ 0.02; electronic supplemen-
tary material, table S1 and figure S4). Finland makes an
exception, with strong shifts in frequencies between the two
main haplotypes H15 and H29 (Fst ¼ 0.39, based on the data
from Jansson et al. [20]).
3. Discussion
Our analyses showed that the genetic diversity of grey wolves
from the Western Palaearctic used to be remarkably higher a
century ago than it is today (figure 2). In parallel, population
differentiation was lower, indicating better historical connec-
tivity and larger effective sizes. These patterns make clear
genetic signatures of the human persecutions faced by the
wolf up until the twentieth century at the continental scale
and suggest that the Late Pleistocene bottleneck that leads
to the loss of prehistoric diversity [17,25] was not solely
responsible. Importantly, these signatures remarkably differ
across Western and Central/Eastern Europe, in light of the
contrasting histories of this controversial species throughout
the Old World.
Before their nearly complete extermination, Western Euro-
pean wolves featured highly diverse and private mtDNA
variation, comparable to the rest of the continent (figure 2; elec-
tronic supplementary material, table S1). Most of this diversity
rapidly vanished at the turn of the twentieth century, not long
before the last wolf packs disappeared from France, Switzer-
land, Germany and much of Italy and Iberia. As one would
expect, these severe bottlenecks, detected in present-day
nuclear data [15,26,27], had far-reaching consequences on the
genetic composition and structure of modern populations.
Wolves that recolonized the Western and Central Alps exclu-
sively originate from the Italian population (C. l. italicus) that
survived in the Apennines [28,29] and are genetically different
from the ones present a century ago (electronic supplementary
material, table S1 and figure S3). The Iberian population
(C. l. signatus), however, retained more diversity [30,31]. Yet,
the mitochondrial identity of these populations appears to be
very recent and partly driven by the strong drift underlying
the decline and recolonization processes; prior to the declines, 5
the C. l. italicus mtDNA private lineage was not restricted to
rspb.royalsocietypublishing.org
Italy; Iberian wolves killed in the nineteenth century carried
non-‘signatus’ alleles. More generally, contemporary popula-
tions share far less haplotypes than before (due to population
disconnection and lower effective population sizes), resulting
in extremely high genetic differentiation.
By contrast, Eastern and Central European historical speci-
mens tell a very different story. Their diversity remained high
throughout the past centuries (figure 2; electronic supplemen-
tary material, table S1), and neither their genetic structure nor
their identity substantially changed. Compared with Western
Proc. R. Soc. B 285: 20181148
Europe, this stability reflects well the relaxed pressure under-
gone by the wolf in Central and Eastern Europe. The only
major change observed in the Finnish population is probably
due to westward migration from Russia to replace wolves
hunted in Finland [20].
Scandinavian samples (here represented by Swedish
wolves) were independently considered due to their eastern
phylogeographical connections but Western-like regimes of
eradication. Surprisingly, our historical set of specimens
featured high diversity (electronic supplementary material,
table S1), while a previous study documented only a single
haplotype fixed prior to 1945 (n ¼ 33, from 1829 to 1939).
Different geographical origins of samples may explain such
discrepancy. In contrast to [19], our samples mostly originate
from northern populations (figure 1), which were the last to
disappear in 1966 (Sweden) and 1976 (Norway). While
southern populations appeared early-bottlenecked [19], north-
ern ones may then still have featured high effective sizes during
the nineteenth-early twentieth centuries, which would explain
their high historical genetic variation.
The genetic patterns documented in our study thus correlate
with the differential pressures faced by Western and Eastern
European wolf populations. The most persecuted Western
European regions suffered strong losses of genetic diversity,
increased genetic structure as well as profound changes in
their genetic nature following recolonization dynamics. By con-
trast, Central and Eastern European areas did not experience
such drastic shifts, presumably as persecutions had lesser effects
on their demography. These general conclusions are in line with
contemporary microsatellite data from local populations from
Iberia and Russian Karelia, which only detected a genetic bottle-
neck in the former [31]. Importantly, given the high diversity
maintained throughout Europe less than 150 years ago, we
can clearly rule out the hypothesis of historically low genetic
variation, a major confounding factor when disentangling
human-driven versus biogeographical losses of diversity
[32–34]. This aspect is particularly relevant for the wolf, since
recent demographic analyses based on modern samples
estimated a long-term decline initiated as early as Late Pleisto-
cene, at least 30 000 years ago [21–23], leading to the biased
conclusion that remnant populations had low diversity to
begin with, notably in Italy [24]. Indeed, many prehistoric
mtDNA haplotypes [25] had already been lost by the nineteenth
century (electronic supplementary material, figure S2), but
human pressures clearly bottlenecked the remaining diversity,
over a remarkably short time scale. Hence, by adding a new
time point between prehistoric [25] and modern eras [17], our
study provides direct evidence for at least two steps in the
decline of the wolf across Europe: a Late Pleistocene bottleneck,
subsequently followed by a rapid, drastic, human-mediated
collapse over the last centuries.
 Our results are well in line with historical DNA analysis of
34 wolf specimens scattered across North America, where only
4 out of 13 mtDNA haplotypes survived among contemporary
populations [35]. Declines in genetic diversity over the last cen-
turies were observed in other mammals decimated by hunting
and poaching, such as Eurasian beavers [36], sea otters [37],
Artic foxes [38], brown bears [39,40], Indian tigers [41], black
rhinoceros [42], Guadalupe fur seals [43], northern elephant
seals [44], Hector’s dolphins [45] and grey whales [46]. In few
other cases, diversity was low to begin with, either due to
older bottlenecks or historically poor diversity (e.g. koalas
[47]; Tasmanian tigers [48]). Changes in genetic structure
were also commonly reported (e.g. range retraction has trig-
gered population differentiation in alpine chipmunks [49]).
Analyses of historical DNA also evidenced reduction of
diversity and/or shifts in genetic composition and structure
among a wide array of other declining animals, including
birds [50 – 53], fishes [54] as well as insects [55,56]. In com-
parison to most of these regional studies however, our
historical phylogeography has the advantage to analyse
fine-scale genetic fluctuations using a large number of histori-
cal samples over an extended biogeographical region and
constitutes one of the most exhaustive spatio-temporal
survey of historical DNA to date. Additionally, all genetic
polymorphisms were validated by cloning and high-
throughput sequencing to cope with post-mortem DNA
damage (see Material and methods). Finally, note that our
findings are based on short, maternally transmitted mtDNA
sequences only, a necessary restriction when dealing with
highly degraded and low-quantity historical DNA, and
patterns of past nuclear diversity in our historical wolves
are yet to be documented.
The discrepancies between wolf historical and modern
phylogeographical patterns, extreme in Western Europe, raise
interesting questions on what should be considered as a refer-
ence for conservation guidelines. Should additional efforts be
invested to restore the original diversity of Western popu-
lations, where modern Eastern lineages used to co-occur?
These would involve improving connectivity at the European
level and/or translocation actions from Eastern European
wolves. At present, the long-range dispersal between the
Balkan and Alpine populations [16,57] represent encouraging
steps towards a natural resilience. More generally, similar
skews are known to have affected several organisms that
experienced recent and rapid range shifts due to complex
decline/recolonization dynamics in response to anthropogenic
impacts and climate change (reviewed in [10]). Phylogeogra-
phical and conservation inferences based on modern samples
might thus not tell the whole story.
For the time being, our phylogeography of grey
wolves from the Western Palaearctic makes a comprehensive
continental-wide reference among the rich wolf population
genetic literature, which modern patterns have been reviewed
and discussed elsewhere [15,17,18] (see electronic supplemen-
tary material, figure S1 for correspondence of haplogroups
with the study of Pilot et al. [17]). Wolf population divisions
are commonly based on distribution data [14], which should
be complemented by molecular studies, pending adequate
geographical coverage and sampling schemes [15] as our
meta-analysis provides for modern mtDNA. Genetic tools are
essential to apprehend recolonization patterns and the effects
of population disconnection, to estimate population sizes as
well as to detect hybridization with dogs. Major threats against
wolves commonly include overharvesting and poaching, con- 6
flicts due to predation on livestock, habitat fragmentation
rspb.royalsocietypublishing.org
and generally low public acceptance resulting, in part, from
lack of knowledge. As such, molecular surveys play a central
part in population monitoring and management to improve
the conservation situation of European wolves [18]. In particu-
lar, the strong genetic structure in Western Europe is of concern
given the amazing dispersal capabilities of the wolf, emphasiz-
ing the need for additional efforts to restore connectivity
between current populations. Even if, like other large carni-
vores, the situation has been improving over the past two
decades [14], the economic, political and cultural aspects of
Proc. R. Soc. B 285: 20181148
wolf management remain a topic of extreme controversy in
most European countries.
4. Material and methods
(a) Genetic analyses of historical specimens
Bone or skin samples (mostly maxilloturbinal bones, and ear or
foot skin sample from mounted specimens or hides, respectively)
were obtained from a total of 250 grey wolf C. lupus specimens
generously provided by 69 European museums and collections
(electronic supplementary material, table S3). With the exception
of one sample dating back to the thirteenth century (stratigraphy-
dated), all other samples mainly covered the nineteenth and early
twentieth centuries (1795 – 1945), when persecution in continental
Europe reached its climax. We selected specimens broadly
originating from the Western Palaearctic region (except North
Africa and the Arabian Peninsula), and for which information
on geographical origin and date were available. We also included
two historical (1938) specimens from the Himalayan wolf
(C. l. chanco), a close monophyletic relative of C. l. lupus [58,59],
to be used as outgroup in the phylogenetic analyses (see the
electronic supplementary material for sample details).
Owing to the low copy number and degraded nature of
historical DNA, special precautions to avoid contamination and
authenticate results were taken during the whole laboratory pro-
cedure. DNA extraction and pre-PCR set-up were conducted in a
physically separated laboratory exclusively devoted to the analy-
sis of museum and low DNA-content samples, where amplified
DNA has never been present. In each extraction and PCR-
set-up session, a maximum of eight samples (including negative
blanks) were handled simultaneously in a dedicated Plexiglas
hood by personnel wearing disposable protective clothing and
masks. All equipment used was UV-illuminated several hours
prior the extraction and working surfaces where bleach-cleaned
before and after each extraction and PCR-set-up rounds. Bone
samples were processed by removing the exterior layer with a
sterile scalpel blade. Samples were subsequently grinded to
powder in a small mortar, which was bleached and exposed to
shortwave UV after each use. Skin samples (approx. 1  1 cm)
were sliced in small fragments with a sterile scalpel blade,
after removing hairs. DNA was then extracted from bone
powder and skin fragments using the QIAamp DNA Mini
Kit (Qiagen).
To facilitate amplification of fragmented DNA, three short
(approx. 111, 119 and 225 bp), partially overlapping sequences
were amplified, resulting in an approximately 290 bp portion
of the left domain of the mtDNA CR and approximately 27 bp
of the adjacent Pro-tRNA gene (length excluding primers), with
the following primer pairs: L15995 [60]/HW1 (50 -CAAACCA
TTAATGCACGACG-30 ); LW1 (50 -CGTCGTGCATTAATGG
TTTG-30 )/HW2 (50 -TCGAGGCATGGTGATTAAG-30 ); LW1/
H16498 [61]. This highly variable CR portion is the marker of
choice given that it has been massively used in wolf population
genetics from all over the world, enabling the inclusion of
previously published data (reviewed in [18]; see the next section).
Amplifications were performed in 25 ml reactions containing
2 mM MgCl2, 0.2 mM of each dNTP, 0.5 mM of each primer,
200 mg ml21 of bovine serum albumin (Roche), 1 Gold buffer
(Applied Biosystems), 1 unit AmpliTaq Gold DNA polymerase
(Applied Biosystems) and 10 ml of template. Cycling conditions
were: initial denaturation at 958C for 3 min; 55 cycles at 958C for
30 s, 548C for 30 s and 728C for 1 min. Between two and seven
independent PCR amplifications were done for each fragment,
and only samples with at least two positive amplifications per
fragment were retained for subsequent analyses (n ¼ 155). Bone
samples had a significantly higher rate of successful amplification
for all three fragments (x 2 (1, n ¼ 252) ¼ 11.27, p , 0.001) than
museum-prepared skins, as expected [62].
We followed recommended guidelines for historical sample
DNA sequencing, in particular to account for artefactual vari-
ation stemming from post-mortem DNA damage (e.g.
deamination; [25,63]). An adequate procedure is to sequence
with high coverage products from different replicate PCRs
using high-throughput sequencing or cloning techniques, true
haplotypes being present in a majority of sequencing reads, as
opposed to miscodings. We implemented this approach as fol-
lows. First, we sequenced (both directions) two independent
PCR replicates for each CR fragment for all of the 155 retained
specimens, by Sanger technology on an ABI3100 platform
(Applied Biosystems), after purification with the QIAquick PCR
Purification kit (Qiagen). Second, we double-verified every
single haplotype identified in two ways: 454 pyrosequencing
and/or cloning.
For 454 pyrosequencing, the three PCR primer pairs were
modified by the addition of specific 5–8 bp tags on the 50 -end to
allow the assignment of sequence reads to the relevant sample.
After PCR and DNA purification (MinElute PCR purification kit,
Qiagen), all amplicons from the different samples were titrated
using capillary electrophoresis (QIAxcel, Qiagen) and then mixed
together, in equimolar concentrations. This mixture underwent
sequencing on a GS FLX or a GS Junior pyrosequencing system
(Roche), following manufacturer’s instructions. After identifi-
cation of primers and tags, only sequences with perfect match
on tags and a maximum of two errors on primers were taken
into account. Then, after removal of primers and tags, haplotypes
were inferred from the majority-rule consensus of all different
454 reads for each sample. The number of reads per amplicon
across the three analysed fragments ranged from 11 to 1581
(308 on average).
For cloning, we amplified again the three CR fragments and
cloned them with the pGEM-T Easy Vector System (Promega).
For each amplicon, a minimum of eight positive clones were
subsequently sequenced (Sanger).
In total, the double validation of every haplotype involved
either two independent rounds of 454 pyrosequencing (74% of
haplotypes) or one round of 454 pyrosequencing plus one
round of cloning (26% of haplotypes).
(b) Meta-analysis of existing datasets
To include data generated in previous studies, we searched for
correspondence of our CR haplotypes with those published over
the past 20 years in the literature, using the blastn algorithm on
sequences available on GenBank. This task is challenged by the
different sequence overlap between studies. To account for this,
we further downloaded and manually aligned published haplo-
types against our sequences, considering them identical if
sequence overlap was greater than 90%. Below this threshold,
we assigned them to collapsed haplotypes or more conserva-
tively to haplogroups, depending on the analysis. We did not
include data with sequence overlap less than 50%. Once we
established correspondences between haplotypes and
haplogroups, we computed their proportions in the different 7
geographical regions where such data were available. Corre-
spondences are provided in electronic supplementary material,
rspb.royalsocietypublishing.org
table S4. In total, we could include modern data from 20
countries/populations, comprehensively investigated by
regional studies [16,19,20,26 – 28,30,64 – 76], representing a total
of 1814 modern wolves (electronic supplementary material,
table S1). Note that such data are mostly commissioned by
national authorities in charge of wolf management and are
therefore primarily available and comparable by country,
which sometimes do not reflect actual populations. In addition,
we also included the published genetic data obtained from
historical samples (nineteenth and twentieth centuries) available
Proc. R. Soc. B 285: 20181148
for Finland (n ¼ 22) [20].
(c) Phylogenetic and population genetic analyses
Given the little divergence between CR sequences in wolves, we
opted for a statistical network to represent relationships between
haplotypes, accounting for indel polymorphism (TCS) [77].
We also performed maximum-likelihood reconstruction with
PHYML [78], using a HKY þ G þ I model of sequence evolution
(JMODELTEST [79]), and tested branch support by 1000 bootstrap
replicates. We ran a second phylogenetic tree to compare our CR
haplotypes with the prehistoric ones obtained from ancient DNA
of 24 Western Palaearctic wolves dated to the Late Pleistocene/
Early Holocene (57 bp) [25]. To quantify the genetic displacement
between modern and historical diversity, we performed a principal
component analysis (in PCAGEN, http://www2.unil.ch/popgen/
softwares/pcagen.htm) on CR haplotype frequencies for popu-
lations defined according to published datasets (usually countries,
table 1), with a minimum sample size of five individuals. We further
computed pairwise Fst in ARLEQUIN v. 3.5 [80].
We used haplotype diversity (Hd) as the main measure of
genetic variation among European regions and populations for his-
torical (before 1945) and modern (after 1990) times. This metric,
P 
computed as nðn  1Þ 1  i x2i , where n corresponds to the
sample size and xi the frequency of allele i, accounts for the relative
allele frequencies (not only the number of alleles). Importantly, Hd
is the most appropriate for European wolves as they do not show
any phylogenetic structure, but rather frequency differences of
population-specific, yet paraphyletic haplotypes (e.g. Italian and
Iberian lineages). We also computed nucleotide diversity (p),
which accounts for sequence divergence, for each European
region/population in both modern and historical periods. We com-
pared Hd, p and Fst between modern and historical datasets by
paired, non-parametric Wilcoxon signed-rank tests; note that
results should be taken with caution for Fst given the non-indepen-
dence of pairwise genetic distances between populations.
Because our modern datasets feature greater sample sizes than
the historical ones, we performed a resampling procedure to stat-
istically test differences of Hd and p between epochs. To this
end, we randomly subset the modern datasets with the sample
sizes of the historical ones (10 000 replicates), and compared
whether historical estimates fall within the 95% distribution of
the resampled modern estimates.
Temporal trends of Hd were calculated by sliding window
analyses for historical samples with accurate date information.
We performed separate analyses for Western populations where
persecutions led to total or nearly total extermination of the wolf
(France, Switzerland, Spain, Portugal and Italy; n ¼ 47), and East-
ern populations where persecution had lesser demographic effects
balanced by immigration from the east (Balkanic and Baltic
countries, Poland, Czech Republic, Slovakia, Hungary, Romania,
Finland, Russia, Turkey and Caucasus; n ¼ 80). The sliding
window analyses were performed on 30 years spans, allowing to
have meaningful sample sizes over each span. Only sample sizes
of n  5 were considered.
Data accessibility. Haplotype sequences are deposited in GenBank under Acknowledgements. We warmly thank the numerous museums 8
accession nos MH559214– MH559266. (and museum curators and technicians) and collections from all
Authors’ contributions. L.F. and P.T. designed the study, supervised by across Europe that provided historical samples for this study (detailed

L.F. L.F., C.M., N.R. and F.B. conducted labwork. L.F., C.D. and
N.S. conducted the analyses. C.D. and L.F. drafted the manuscript,
which was subsequently improved by all co-authors.
Competing interests. We declare we have no competing interests.
Funding. This study was funded by the University of Lausanne.
rspb.royalsocietypublishing.org
list available in electronic supplementary material, table S3). We also
thank K. Parker and C. Ossola for help in the laboratory, T. Vincent
for information about museum collections and S. Schmid for useful
comments on previous versions of the manuscript.

References

Proc. R. Soc. B 285: 20181148

1. Frankham R, Ballou JD, Briscoe DA. 2010
Introduction to conservation genetics. Cambridge,
UK: Cambridge University Press.
2. Frankham R. 2005 Genetics and extinction. Biol.
Conserv. 126, 131–140. (doi:10.1016/j.biocon.2005.
05.002)
3. Bolnick DL, Nosil P. 2007 Natural selection in
populations subject to a migration load. Evolution
61, 2229 –2243. (doi:10.1111/j.1558-5646.2007.
00179.x)
4. Peisch S, Dupanloup I, Kirkpatrick M, Excoffier L.
2015 On the accumulation of deleterious mutations
during range expansions. Mol. Ecol. 22,
5972–5982. (doi:10.1111/mec.12524)
5. Dufresnes C, Perrin N. 2015 Effect of biogeographic
history on population vulnerability in European
amphibians. Conserv. Biol. 29, 1235 –1241. (doi:10.
1111/cobi.12490)
6. Beaumont MA. 2010 Approximate Bayesian
computation in evolution and ecology. Annu. Rev.
Ecol. Evol. Syst. 41, 379–406. (doi:10.1146/
annurev-ecolsys-102209-144621)
7. Schraiber JG, Akey JM. 2015 Methods and models
for unravelling human evolutionary history. Nat.
Rev. Genet. 16, 727–740. (doi:10.1038/nrg4005)
8. Excoffier L, Dupanloup I, Huerta-Sanchez E, Sousa
VC, Foll M. 2013 Robust demographic inference
from genomic and SNP data. PLoS Genet. 9,
e1003905. (doi:10.1371/journal.pgen.1003905)
9. Li H, Durbin R. 2011 Inference of human population
history from individual whole-genome sequences.
Nature 475, 493 –496. (doi:10.1038/nature10231)
10. Waters JM, Grosser S. 2016 Managing shifting
species: ancient DNA reveals conservation
conundrums in a dynamic world. BioEssays 38,
1177–1184. (doi:10.1002/bies.201600044)
11. Zimen E. 1978 Der wolf-mythos und verhalten.
Munich, Germany: Meyster, Wien.
12. Mech DL, Boitani L. 2003 Wolves: behavior, ecology, and
conservation. Chicago, IL: University of Chicago Press.
13. Breitenmoser U. 1998 Large predators in the
Alps: the fall and rise of man’s competitors. Biol.
Conserv. 83, 279 –289. (doi:10.1016/S0006-
3207(97)00084-0)
14. Chapron G et al. 2014 Recovery of large carnivores
in Europe’s modern human-dominated landscapes.
Science 346, 1517 –1519. (doi:10.1126/science.
1257553)
15. Hindrikson M et al. 2017 Wolf population genetics
in Europe: a systematic review, meta-analysis and
suggestions for conservation and management. Biol.
Rev. 92, 1601–1629. (doi:10.1111/brv.12298)
16. Rauer G, Knauer F, Fumagalli L. 2013. Austria—
meeting place of three wolf populations: melting pot
or black hole? In Wolf Conservation in a Human
Dominated Landscape (eds H Potocnik, N Razen, J
Mulej, M Jelencic, I Bertoncelj), p. 44. Postojna,
Slovenia: University of Ljubljana, Biotechnical Faculty.
17. Pilot M, Branicki W, Jedrzejewski W, Goszczynski J,
Jedrzejewska B, Dykyy I, Shkvyrya M, Tsingarska E.
2010 Phylogeographic history of grey wolves in
Europe. BMC Evol. Biol. 10, 104. (doi:10.1186/1471-
2148-10-104)
18. De Groot GA et al. 2016 Decades of population
genetic research reveal the need for harmonization
of molecular markers: the grey wolf Canis lupus as a
case study. Mammal Rev. 46, 44– 59. (doi:10.1111/
mam.12052)
19. Flagstad O, Walker CW, Vila C, Sundqvist A-K, Fernholm
B, Hufthammer AK, Wiig O, Koyola I, Ellegren H. 2003
Two centuries of the Scandinavian wolf population:
patterns of genetic variability and migration during an
era of dramatic decline. Mol. Ecol. 12, 869–880.
(doi:10.1046/j.1365-294X.2003.01784.x)
20. Jansson E, Harmoinen J, Ruokonen M, Aspi J. 2014
Living on the edge: reconstructing the genetic
history of the Finnish wolf population. BMC Evol.
Biol. 14, 64. (doi:10.1186/1471-2148-14-64)
21. Pilot M, Greco C, vonHoldt BM, Jedrzejewska B,
Randi E, Jedrzejewski W, Sidorovich VE, Ostrander
EA, Wayne RK. 2014 Genome-wide signatures of
population bottlenecks and diversifying selection in
European wolves. Heredity 112, 428–442. (doi:10.
1038/hdy.2013.122)
22. Fan Z et al. 2016 Worldwide patterns of
genomic variation and admixture in gray
wolves. Genome Res. 26, 163–173. (doi:10.1101/gr.
197517.115)
23. Ersmark E et al. 2016 From the past to the present:
wolf phylogeography and demographic history
based on the mitochondrial control region. Front.
Ecol. Evol. 4, 134. (doi:10.3389/fevo.2016.00134)
24. Montana L et al. 2017 Combining phylogenetic and
demographic inferences to assess the origin of the
genetic diversity in an isolated wolf population.
PLoS ONE 12, e0176560. (doi:10.1371/journal.pone.
0176560)
25. Stiller M et al. 2006 Patterns of nucleotide
misincorporations during enzymatic amplification
and direct large-scale sequencing of ancient DNA.
Proc. Natl Acad. Sci. USA 103, 13 578–13 584.
(doi:10.1073/pnas.0605327103)
26. Randi E, Lucchini V, Christensen MF, Mucci N, Funk
SM, Dolf G, Loeschcke V. 2000 Mitochondrial DNA
variability in Italian and East European wolves:
detecting the consequences of small population size
and hybridization. Conserv. Biol. 14, 464–473.
(doi:10.1046/j.1523-1739.2000.98280.x)
27. Lucchini V, Galov A, Randi E. 2004 Evidence of
genetic distinction and long-term population
decline in wolves (Canis lupus) in the Italian
Apennines. Mol. Ecol. 13, 523–536. (doi:10.1046/j.
1365-294X.2004.02077.x)
28. Valière N, Fumagalli L, Gielly L, Miquel C, Lequette
B, Poulle ML, Weber JM, Arlettaz R, Taberlet P.
2003 Long-distance wolf recolonization of France
and Switzerland inferred from non-invasive genetic
sampling over a period of 10 years. Anim. Conserv.
6, 83 –92. (doi:10.1017/S1367943003003111)
29. Fabbri E et al. 2007 From the Apennines to the
Alps: colonization genetics of the naturally
expanding Italian wolf (Canis lupus) population.
Mol. Ecol. 16, 1661–1671. (doi:10.1111/j.1365-
294X.2007.03262.x)
30. Vila C, Amorim IR, Leonard JA, Posada D, Castroviejo
J, Petrucci-Fonseca F, Crandall KA, Ellegren H,
Wayne RK. 1999 Mitochondrial DNA
phylogeography and population history of the grey
wolf Canis lupus. Mol. Ecol. 8, 2089–2103. (doi:10.
1046/j.1365-294x.1999.00825.x)
31. Sastre N, Vila C, Salinas M, Bologov VV, Urios V,
Sanchez A, Francino O, Ramirez O. 2011 Signatures
of demographic bottlenecks in European wolf
populations. Conserv. Genet. 12, 701–712. (doi:10.
1007/s10592-010-0177-6)
32. Matocq MD, Villablanca FX. 2001 Low genetic
diversity in an endangered species: recent or historic
pattern? Biol. Conserv. 98, 61 –68. (doi:10.1016/
S0006-3207(00)00142-7)
33. Leonard JA. 2008 Ancient DNA applications for
wildlife conservation. Mol. Ecol. 17, 4186–4196.
(doi:10.1111/j.1365-294X.2008.03891.x)
34. Orlando L, Cooper A. 2014 Using ancient DNA to
understand evolutionary and ecological processes.
Annu. Rev. Ecol. Evol. Syst. 45, 573 –598. (doi:10.
1146/annurev-ecolsys-120213-091712)
35. Leonard JA, Vila C, Wayne RK. 2005 Legacy lost:
genetic variability and population size of extirpated
US grey wolves (Canis lupus). Mol. Ecol. 14, 9–17.
(doi:10.1111/j.1365-294X.2004.02389.x)
36. Horn S et al. 2014 Ancient mitochondrial DNA and
the genetic history of Eurasian beaver (Castor fiver)
in Europe. Mol. Ecol. 23, 1717– 1729. (doi:10.1111/
mec.12691)
37. Larson S, Jameson RJ, Etnier M, Fleming M, Bentzen
P. 2002 Loss of genetic diversity in sea otters
(Enhydra lutris) associated with the fur trade of the
18th and 19th centuries. Mol. Ecol. 11, 1899–1903.
(doi:10.1046/j.1365-294X.2002.01599.x)
38. Nyström V, Angerbjörn A, Dalen L. 2006 Genetic
consequences of a demographic bottleneck in the
Scandinavian arctic fox. Oikos 114, 84 –94. (doi:10.
1111/j.2006.0030-1299.14701.x)
39. Miller CR, Waits LP, Joyce P. 2006 Phylogeography
and mitochondrial diversity of extirpated brown
bear (Ursus arctos) populations in the contiguous
United States and Mexico. Mol. Ecol. 15, 4477 –
4485. (doi:10.1111/j.1365-294X.2006.03097.x)
40. Xenikoudakis G, Ersmark E, Tison JL, Waits L,
Kindberg J, Swenson JE, Dalen L. 2015
Consequences of a demographic bottleneck on
genetic structure and variation in the Scandinavian
brown bear. Mol. Ecol. 24, 3441 –3454. (doi:10.
1111/mec.13239)
41. Mondol S, Bruford MW, Ramakrishnan U. 2013
Demographic loss, genetic structure and the
conservation implications for Indian tigers.
Proc. R. Soc. B 280, 20130496. (doi:10.1098/rspb.
2013.0496)
42. Moodley Y et al. 2017 Extinctions, genetic erosion
and conservation options for the black rhinoceros
(Diceros bicornis). Sci. Rep. 7, 41417. (doi:10.1038/
srep41417)
43. Weber DS, Stewart BS, Lehman N. 2004 Genetic
consequences of a severe population bottleneck in
the Guadalupe fur seal (Arctocephalus townsendi).
J. Hered. 95, 144–153. (doi:10.1093/jhered/esh018)
44. Hoelzel RA, Fleischer RC, Campagna C, Le Bœuf BJ,
Alvord G. 2002 Impact of population bottleneck on
symmetry and genetic diversity in the northern
elephant seal. J. Evol. Biol. 15, 567 –575. (doi:10.
1046/j.1420-9101.2002.00419.x)
45. Pichler FB, Baker CS. 2000 Loss of genetic diversity
in the endemic Hector’s dolphin due to fisheries-
related mortality. Proc. R. Soc. Lond. B 267,
97 –102. (doi:10.1098/rspb.2000.0972)
46. Alter SE, Rynes E, Palumbi SR. 2007 DNA evidence
for historic population size and past ecosystem
impacts of gray whales. Proc. Natl Acad. Sci. USA
104, 15 162–15 167. (doi:10.1073/pnas.
0706056104)
47. Tsangaras K, Avila-Arcos M, Ishida Y, Helgen KM,
Roca AL, Greenwood AD. 2012 Historically low
mitochondrial DNA diversity in koalas (Phascolarctos
cinereus). BMC Genet. 13, 92. (doi:10.1186/1471-
2156-13-92)
48. Menzies BR, Renfree MB, Heider T, Mayer F,
Hildebrandt TB, Pask AJ. 2012 Limited genetic
diversity preceded extinction of the Tasmanian
tiger. PLoS ONE 7, e35433. (doi:10.1371/journal.
pone.0035433)
49. Bi K, Linderoth T, Vanderpool D, Good JM, Nielsen
R, Moritz C. 2013 Unlocking the vault: next
generation museum population genomics. Mol. Ecol.
22, 6018–6032. (doi:10.1111/mec.12516)
50. Glenn TC, Stephan W, Braun MJ. 1999 Effects of a
population bottleneck on whooping crane
mitochondrial DNA variation. Conserv. Biol.
13, 1097–1107. (doi:10.1046/j.1523-1739.1999.
97527.x)
51. Taylor SS, Jamieson IG, Wallis GP. 2007 Historic and
contemporary levels of genetic variation in two New
Zealand passerines with different histories of
decline. J. Evol. Biol. 20, 2035–2047. (doi:10.1111/
j.1420-9101.2007.01362.x)
52. Tracy LN, Jamieson IG. 2010 Historic DNA reveals
contemporary population structure results from
anthropogenic effects, not pre-fragmentation
patterns. Conserv. Genet. 12, 517– 526. (doi:10.
1007/s10592-010-0158-9)
53. Draheim HM, Baird P, Hag SM. 2012 Temporal
analysis of mtDNA variation reveals decreased
genetic diversity in least terns. Condor 114,
145 –154. (doi:10.1525/cond.2012.110007)
54. Savary R, Dufresnes C, Champigneulle A, Caudron A,
Dubey S, Perrin N, Fumagalli L. 2017 Stocking
activities for the Arctic charr in Lake Geneva: genetic
effects in space and time. Ecol. Evol. 7, 5201–5211.
(doi:10.1002/ece3.3073)
55. Schmid S. 2016. Spatial and temporal genetic
dynamics of the endangered grasshoper Oedaleus
decorus revealed by museum specimens. Master
thesis, University of Lausanne, Switzerland.
56. Ugelvid LV, Nielsen PS, Boomsma JJ, Nash DR. 2011
Reconstructing eight decades of genetic variation in
an isolated Danish population of the large blue
butterfly Maculinea arion. BMC Evol. Biol. 11, 201.
(doi:10.1186/1471-2148-11-201)
57. Razen N et al. 2016 Long-distance dispersal
connects Dinaric-Balkan and Alpine grey wolf (Canis
lupus) populations. Eur. J. Wildl. Res. 62, 137– 142.
(doi:10.1007/s10344-015-0971-z)
58. Sharma DK, Maldonado JE, Jhala YV, Fleischer RC.
2004 Ancient wolf lineages in India. Proc. R.
Soc. Lond. B 271, S1 –S4. (doi:10.1098/rsbl.
2003.0071)
59. Aggarwal RK, Kivisild T, Ramadevi J, Singh L.
2007 Mitochondrial DNA coding region
sequences support the phylogenetic distinction
of two Indian wolf species. J. Zool. Syst. Evol. Res.
45, 163– 172. (doi:10.1111/j.1439-0469.2006.
00400.x)
60. Taberlet P, Bouvet J. 1994 Mitochondrial DNA
polymorphism, phylogeography, and conservation
genetics of the brown bear (Ursus arctos) in Europe.
Proc. R. Soc. Lond. B 255,195–200. (doi:10.1098/
rspb.1994.0028)
61. Fumagalli L, Taberlet P, Favre L, Hausser J. 1996
Origin and evolution of homologous repeated
sequences in the mitochondrial control region of
shrews. Mol. Biol. Evol. 13, 31 –46. (doi:10.1093/
oxfordjournals.molbev.a025568)
62. Wandeler P, Hoeck PEA, Keller LF. 2007 Back to the
future: museum specimens in population genetics.
Trends Ecol. Evol. 22, 634–642. (doi:10.1016/j.tree.
2007.08.017)
63. Hofreiter M, Jaenicke V, Serre D, von Haeseler A, 9
Pääbo S. 2001 DNA sequences from multiple
rspb.royalsocietypublishing.org
amplifications reveal artifacts induced by cytosine
deamination in ancient DNA. Nucleic Acids Res. 29,
4793– 4799. (doi:10.1093/nar/29.23.4793)
64. Randi E et al. 2014 Multilocus detection of
wolf  dog hybridization in Italy and guidelines
for marker selection. PLoS ONE 9, e86409. (doi:10.
1371/journal.pone.0086409)
65. Jedrzejewski W et al. 2005 Genetic diversity and
relatedness within packs in an intensively hunted
population of wolves Canis lupus. Acta Theriol. 50,
Proc. R. Soc. B 285: 20181148
3–22. (doi:10.1007/BF03192614)
66. Gomercic T, Sindicic M, Galov A, Arbanasic H, Kusak
J, Kocijan I, Gomercic MD, Huber D. 2010 High
genetic variability of the grey wolf (Canis lupus L.)
population from Croatia revealed by mitochondrial
DNA control region sequences. Zool. Stud. 49,
816–823.
67. Hindrikson M, Männil P, Ozolins J, Krzywinski A,
Saarma U. 2012 Bucking the trend in wolf-dog
hybridization: first evidence from Europe of
hybridization between female dogs and male
wolves. PLoS ONE 7, e46465. (doi:10.1371/journal.
pone.0046465)
68. Baltrünaité L, Balciauskas L, Akesson M. 2013 The
genetic structure of the Lithuanian wolf population.
Cent. Eur. J. Biol. 8, 440 –447.
69. Boggiano F, Ciofi C, Boitani L, Formia A, Grottoli L,
Natali C, Ciucci P. 2013 Detection of an East
European wolf haplotype puzzles mitochondrial
DNA monomorphism of the Italian wolf population.
Mamm. Biol. 78, 374 –378. (doi:10.1016/j.mambio.
2013.06.001)
70. Czarnomska SD et al. 2013 Concordant
mitochondrial and microsatellite DNA structuring
between Polish lowland and Carpathian mountain
wolves. Conserv. Genet. 14, 573–588. (doi:10.1007/
s10592-013-0446-2)
71. Fabbri E, Caniglia R, Kusak J, Galov A, Gomercic T,
Arbanasic H, Huber D, Randi E. 2014 Genetic
structure of expanding wolf (Canis lupus)
populations in Italy and Croatia, and the early
steps of the recolonization of the Eastern Alps.
Mamm. Biol. 79, 138 –148. (doi:10.1016/j.mambio.
2013.10.002)
72. Moura AE, Tsingarska E, Dabrowski MJ, Czarnomska
SD, Jedrzejewska B, Pilot M. 2014 Unregulated
hunting and genetic recovery from a severe
population decline: the cautionary case of Bulgarian
wolves. Conserv. Genet. 15, 405–417. (doi:10.1007/
s10592-013-0547-y)
73. Pilot M, Dabrowski MJ, Hayrapetyan V, Yavruyan EG,
Kopaliani N, Tsingarska E, Bujalska B, Kaminski S,
Bogdanowicz W. 2014 Genetic variability of the grey
wolf (Canis lupus) in the Caucasus in comparison
with Europe and the Middle East: distinct or
intermediary population. PLoS ONE 9, e93828.
(doi:10.1371/journal.pone.0093828)
74. Ibis O, Aksöyek E, Ozcan S, Keten A, Yorulmaz T, Tez
C. 2016 Genetic analysis of the Turkish gray wolf
(Canis lupus) based on partial mitochondrial DNA
sequences. Vertebr. Zool. 66, 427–435.
75. Montana L, Caniglia R, Galaverni M, Fabbri E, Randi
E. 2017 A new mitochondrial haplotype confirms
the distinctiveness of the Italian wolf (Canis lupus)
population. Mamm. Biol. 84, 30– 34. (doi:10.1016/
j.mambio.2017.01.005)
76. Pires AE, Amorim IR, Borges C, Simoes F, Teixeira T,
Quaresma A, Petrucci-Fonseca F, Matos J. 2017 New
insights into the genetic composition and
phylogenetic relationship of wolves and dogs in
the Iberian Peninsula. Ecol. Evol. 7, 4404–4418.
(doi:10.1002/ece3.2949)
77. Clement M, Posada D, Crandall KA. 2000 TCS: a computer
program to estimate gene genealogies. Mol. Ecol. 9,
1657–1659. (doi:10.1046/j.1365-294x.2000.01020.x)
78. Guindon S, Gascuel O. 2003 PHYML: a simple, fast
and accurate algorithm to estimate large
phylogenies by maximum-likelihood. Syst. Biol. 52,
696 –704. (doi:10.1080/10635150390235520)
79. Darriba D, Taboada GL, Doallo R, Posada D. 2012 10
jMODELTEST 2: more models, new heuristics and
rspb.royalsocietypublishing.org
parallel computing. Nat. Methods 9, 772. (doi:10.
1038/nmeth.2109)
80. Excoffier L, Foll M, Petit RJ. 2005 Arlequin
ver. 3.0: an integrated software package
for population genetics data analysis.
Evol. Bioinform. 1, 47 –50. (doi:10.1177/
117693430500100003)
Proc. R. Soc. B 285: 20181148