Introduction
De Candolle (1820) was the first author to propose that
the current geographical distribution of living organisms
depends upon both ecological and historical parameters.
The influence of historical parameters can be assessed via
historical biogeographic studies (Morrone & Crisci 1995;
Wiley 1988b). Most studies of this type have compared the
geographical distribution of taxa at or above the species
level. With the development of molecular methods, it is
now possible to investigate the geographical variation
using molecular markers, and to deduce intraspecific
phylogeographic structures (Avise et al. 1987). In this
context, it is tempting to apply the methods of historical
biogeography below the species level, and to compare the
intraspecific phylogeographic patterns of several taxa over
the same area. Up to now, very few studies have been published
in this field. The main goal of these studies has been
to search for concordant geographical distribution among
lineages within different species, which would indicate the
influence of a common historical factor.
Thus, in south-eastern United States, a comparison of
the intraspecific phylogeographies of four freshwater
fishes (Bermingham & Avise 1986), and of 15 additional
species including a bird, a reptile, a crustacean, a
mollusc, and other marine and freshwater fishes (Avise
1992), has been carried out based on mitochondrial DNA
(mtDNA) polymorphism. The most interesting result
was that almost all the assayed species exibited a deep
genetic break between populations from the Atlantic
Coast and populations from the Gulf of Mexico. The
same kind of mtDNA concordance, with a deep genetic
break at the intraspecific level, has also been recorded in
Molecular Ecology (1998) 7, 453–464
© 1998 Blackwell Science Ltd
Comparative phylogeography and postglacial colonization
routes in Europe
PIERRE TABERLET,* LUCA FUMAGALLI,† ANNE-GABRIELLE WUST-SAUCY,‡
JEAN-FRANÇOIS COSSON§
*Laboratoire de Biologie des Populations d’Altitude, CNRS UMR 5553, Université Joseph Fourier, BP 53, F-38041 Grenoble Cedex
9, France, †Department of Zoology, University of Queensland, Brisbane, Qld 4072, Australia, ‡Institut de Zoologie et d’Ecologie
Animale, Université de Lausanne, Bâtiment de Biologie, CH-1015 Lausanne, Switzerland, §Laboratoire Génome et Populations,
CNRS UPR 9060, Université de Montpellier II, Place Eugène Bataillon, C.C.63, F-34095 Montpellier Cedex 5, France
Abstract
The Quaternary cold periods in Europe are thought to have heavily influenced the
amount and distribution of intraspecific genetic variation in both animals and plants. The
phylogeographies of 10 taxa, including mammals (Ursus arctos, Sorex spp., Crocidura
suaveolens, Arvicola spp.), amphibians (Triturus spp.), arthropods (Chorthippus parallelus),
and plants (Abies alba, Picea abies, Fagus sylvatica, Quercus spp.), were analysed to
elucidate general trends across Europe. Only a small degree of congruence was found
amongst the phylogeographies of the 10 taxa, but the likely postglacial colonization
routes exhibit some similarities. A Brooks parsimony analysis produced an unrooted area
phylogram, showing that: (i) the northern regions were colonized generally from the
Iberic and Balkanic refugia; and (ii) the Italian lineages were often isolated due to the
presence of the Alpine barrier. The comparison of colonization routes highlighted four
main suture-zones where lineages from the different refugia meet. Some of the intraspecific
genetic distances among lineages indicated a prequaternary divergence that
cannot be connected to any particular cold period, but are probably related mainly to the
date of arrival of each taxon in the European continent. As a consequence, molecular
genetics so far appears to be of limited use in dating Quaternary events.
Keywords: chloroplast DNA, glaciation, mitochondrial DNA, Quaternary, refugium, suture-zone
Correspondence: P. Taberlet. Tel. +33-476-51-46-00 ext. 3705;
Fax. +33-476-51-42-79; E-mail: Pierre.Taberlet@ujf-grenoble.fr
the wet tropics of Australia for four species out of six
tested (five birds, one reptile) (Joseph et al. 1995). Only
one study so far has examined concordance amongst
species at the continental level. The mtDNA intraspecific
phylogeographic structure of five bird species has been
compared throughout North America (Zink 1996). A lack
of a congruent phylogeographic pattern was found,
suggesting that the five codistributed species have not
had a long history of coassociation.
Before discussing comparative phylogeography in
Europe, it is important to: (i) briefly describe the
Quaternary cold periods which appear to have had a
dramatic influence on most organisms in temperate
regions (Hewitt 1996); and (ii) mention the geographical
particularities of this continent. Twenty years ago, it was
recognized that changes in the earth’s orbit are the fundamental
causes of the Quaternary climatic oscillations
(Hays et al. 1976). The ice sheets of the Northern
Hemisphere began to grow about 2.5 Myr ago, and the
major climatic oscillations occurred during the last 700 ky,
with a dominant 100-ky cycle (Webb & Bartlein 1992). It
has been demonstrated more recently that short-term and
high-amplitude climatic variation also occurred during
the late Pleistocene (130 ky to 10 ky), both within the
glacial and the interglacial periods (Roy et al. 1996). A
decrease of 10–14 °C in average temperature (above
Greenland) may have occurred within only 10–20 y, and
lasted for periods of 70–5000 y (Dansgaard et al. 1993;
GRIP Members 1993). Such temperature shifts could be
linked to enormous discharge of icebergs into the North
Atlantic (Bond et al. 1993). In this context, the relative
climatic stability recorded during the last 8000 y seems to
be the exception rather than the rule. The consequence is
that the European biota has experienced some dramatic
climatic changes during the last few million years, with
extensive oscillations during the last 700 ky. Figure 1
presents the maximum extension of the ice sheet during
the last cooling, about 20–18 ky ago.
From a biogeographic point of view, Europe has some
distinctive features. It is a large peninsula connected to
Asia, with an east–west orientation. The Mediterranean
sea in the south constitutes a strong barrier, and has
limited the possibility of southern displacement of biota
during cold periods. Furthermore, the east–west orientation
of the main mountain ranges of the Alps and the
Pyrenees acted as a barrier to northward expansion of
species during warm periods. The effects of the ice ages
on European species has been examined in detail by
Hewitt (1996): during the Quaternary, each species went
through many contractions/expansions of range, characterized
by extinctions of northern populations when the
temperature decreased, and a northward expansion from
refugia involving spreading from the leading edge. Such a
colonization process implies successive bottlenecks that
may lead to a loss of genetic diversity in the northern
populations, with the exception of cold-tolerant taxa. The
refugia were localized in the mountains of southern
Europe and, for the long-term presence of a species in
Europe, survival in these refugia may have been as important
during warm periods as during cold periods (Bennett
et al. 1991). The disappearance of a species in southern
refugia during warm periods could lead to the complete
extinction of that taxon during the next cooling. The last
northward expansion of European trees, which occurred
from 13 ky BP, is well documented from the pollen record
(Huntley & Birks 1983), and exibited a rate of spread up to
2000 m/y (Bennett 1986).
However, the influence of the Quaternary cold periods
on the geographical distribution of genetic diversity in
Europe is still unclear. Do repeated cycles of isolation in
southern refugia promote divergence and speciation?
How is intraspecific biodiversity distributed? Are there
common patterns of divergence and colonization? In this
study, we present a comparative phylogeographic study
of plants and animals at the European level. Our first goal
is to elucidate general patterns in answer to the above
questions. Our second goal is, assuming the existence of a
molecular clock, to test if the depth of a concordant
454 P. TABERLET ET AL.
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
Fig. 1 Maximum extension of ice sheets in Europe during the last
cold period, 20 000–18 000 y ago (redrawn from Frenzel et al.
1992; Lundqvist & Saarnisto 1995). R1, R2, and R3 indicate the
three main potential refugia in Portugal–Spain, in Italy, and in the
Balkans, respectively. The southern limit of the permafrost is
indicated by the scaled line. Lowered sea shore is shown by a
thinner line at the 100 m submarine contour.
genetic break observed within different species could be
directly related to a particular historical factor such as a
Quaternary cold period.
Materials and methods
The more taxonomically and ecologically diverse the
species compared in this type of study, the more generally
applicable the results will be (Zink 1996). Therefore, in
order to extend the relevance of our results, the taxa compared
were not limited by their taxonomic or ecological
group, but were chosen only according to the availability
of intraspecific phylogeographic studies over the geographical
range considered. Studies where no clear phylogeographic
pattern has been found (e.g. Estoup et al. 1996;
Thomaz et al. 1996) or covering a restricted geographical
range (e.g. Bernatchez et al. 1992; Garnery et al. 1995;
Hardy et al. 1995; Jaarola & Tegelström 1995; Smith et al.
1991) were not taken into account for direct comparison,
but are introduced where necessary. To have the opportunity
to analyse as many data sets as possible, we considered
phylogeographic patterns deduced from different
molecular markers: mtDNA for animals, chloroplast
DNA (cpDNA) or isozymes for plants. Due to clonal
inheritance, cpDNA may be considered to be the plant
counterpart for animal mtDNA. Isozymes are encoded by
nuclear genes, which are characterized by biparental
transmission and by pollen (haploid) or seed (diploid)
dispersal, making phylogeographic inference more
complex. Nevertheless, in plants, isozyme data can be
useful to identify populations which exhibit a common
recent history.
A more restricted data set was used to test if the depth
of genetic breaks observed within different animal species
could be related to any historical factor. The taxa chosen
for this purpose were those for which the mitochondrial
DNA sequence divergence among lineages is known, and
for which a time of divergence can be estimated assuming
a molecular clock. The data used were composed essentially
of intraspecific variation, but as the species boundaries
are sometimes difficult to assess and can be
questionable, data between closely related species were
also considered when showing parapatric distributions.
We simplified the original mtDNA phylogenetic trees by
grouping the more closely related lineages to emphasize
the main features of each taxon.
In order to highlight general patterns that are not
obvious when the phylogeographic relationships are
interpreted separately, a Brooks parsimony analysis
(Brooks 1985, 1990; Wiley 1988a,b) was carried out. It
corresponds to a Wagner parsimony analysis of an
(area × taxon) matrix, where, in our case, each taxon is
represented by a lineage. All the available data from the
comparative study above were used, taking into
account phylogenetic information when available, but
excluding lineages present in all areas. The DELTRAN
optimization routine of the PAUP program (Swofford
1993) was used to find the most parsimonious solution.
COMPARATIVE PHYLOGEOGRAPHY IN EUROPE 455
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
Table 1 The 10 taxa included in the comparative phylogeographic study
Scientific name Common name Taxonomic position Techniques used Reference
Ursus arctos Brown bear Mammalia, Carnivora mtDNAsequencing Taberlet & Bouvet (1994)
Sorex araneus Common shrew Mammalia, Insectivora mtDNAsequencing Taberlet et al. (1994)
Sorex granarius Iberian shrew
Sorex coronatus Millet’s shrew
Sorex samniticus Apennine shrew
Crocidura suaveolens Lesser white-toothed shrew Mammalia, Insectivora mtDNAsequencing Cosson et al. (unpublished)
Arvicola terrestris Northern water vole Mammalia, Rodentia mtDNA sequencing Wust-Saucy et al. (unpublished)
Arvicola sapidus Southwestern water vole
Triturus cristatus Crested newt Amphibia, Urodeles mtDNARFLP Wallis & Arntzen (1989)
Triturus carnifex
Triturus karekini
Triturus marmoratus Marbled newt
Triturus pygmaeus
Chorthippus parallelus Grasshopper Insecta, Orthoptera nuclear DNAsequencing Cooper et al. (1995)
Abies alba Silver fir Gymnospermae, Pinaceae protein electrophoresis Konnert & Bergmann (1995)
Picea abies Norway spruce Gymnospermae, Pinaceae protein electrophoresis Lagercrantz & Ryman (1990)
Fagus sylvatica Common beech Angiospermae, Fagaceae cpDNAPCR RFLP Demesure et al. (1996)
Quercus robur White oaks Angiospermae, Fagaceae cpDNAPCR RFLP Dumolin-Lapègue et al. (1997)
Quercus petraea
Quercus pubescens
and related species
The results of the Brooks parsimony analysis were
visualized by drawing an unrooted area phylogram,
with the different areas being placed to fit as much as
possible to the geography.
Results
Table 1 gives details of the 10 studies considered in our
comparative analysis. The original papers should be consulted
to obtain details of the different phylogeographies.
The simplified phylogeographies or postglacial colonization
routes are presented in Figs 2, 3 and 4.
Ursus arctos (Taberlet & Bouvet 1994)
The brown bear exhibits two distinct mtDNA lineages in
Europe. The eastern lineage is represented mainly by the
large populations of Russia and Romania, whereas the
western lineage, which includes the other European populations,
appears to be organized into two clades corresponding
to two different Quaternary refugia (Fig. 2).
Contact zones between these two main lineages are localized
in Scandinavia (Taberlet et al. 1995) and in Central
Europe (Kohn et al. 1995). Sequence divergences in the
cytochrome b gene amongst these lineages (Fig. 2) have
been estimated by comparing results from different
studies (Randi et al. 1994; Taberlet & Bouvet 1992), and are
relatively low compared to intraspecific divergences in
other taxa (Fig. 2).
Sorex spp. (Taberlet et al. 1994)
The shrews considered in our analysis were Sorex
samniticus and the West European representatives of the S.
araneus group. S. granarius, restricted to Spain, is closely
related to a lineage of S. araneus found in western, northern
and Central Europe. Another lineage of S. araneus
found in Italy and the southern slopes of the Alps is
slightly more divergent. S. coronatus, found in northern
Spain and in France, and S. samniticus, restricted to Italy,
differ from the other lineages by 3.7% and 9.1% in
cytochrome b gene sequences, respectively (Fig. 2).
Crocidura suaveolens (Cosson et al., unpublished)
The lesser white-toothed shrew is represented in Europe
by two main lineages which differ from each other by a
mean pairwise genetic distance of approx. 6.5% in the
cytochrome b gene. The western lineage includes fragmented
populations in western France and northern
Spain. The eastern lineage is represented by the large populations
in Central Europe, Italy, the Balkans, and Turkey
(Fig. 2).
Arvicola spp. (Wust-Saucy et al., unpublished)
The water voles exhibit four main lineages in Europe
differing from each other by 3.8–7.6% in cytochrome b
sequences: one lineage represented by Arvicola sapidus in
Spain and western France, and three lineages of A.
terrestris (Fig. 2). Each of the three lineages of A. terrestris
were composed of populations belonging to the same
clade and named according to a subspecies considered as
representative of that clade: A. terrestris italicus in Italy, the
aquatic form A. terrestris terrestris in northern and Central
Europe, and the fossorial form A. terrestris sherman in
northern Spain, Pyrenees, Massif Central, and the Alps.
Triturus spp. (Wallis & Arntzen 1989)
Five lineages of newts have been included: Triturus
pygmaeus (southern Spain), T. marmoratus (Spain and
France), T. cristatus (north-east France, Germany, Poland
and Russia), T. carnifex (Italy), and T. karelini (the Balkans
and Turkey). The entire mtDNA sequence divergence
estimated from RFLPs amongst these lineages ranges
from 4% to 9% (Fig. 2). The intralineage variation has not
been considered, although it is substantial in T. carnifex
and T. karelini. T. dobrigicus, only present in the Danube
valley, has not been taken into account.
Chorthippus parallelus (Cooper et al. 1995)
The analysis of the variation at a noncoding nuclear marker
in this species revealed geographical subdivision and an
unambiguous pattern of postglacial expansion (Fig. 3).
Most of the European populations of this grasshopper are
homogeneous, originating after a range expansion from a
Balkan refugium, and meet populations from Italian and
Spanish refugia in the Alps and the Pyrenees, respectively.
Abies alba (Konnert & Bergmann 1995)
Silver fir occurs in mountainous regions of the southern
part of Europe. The geographical distribution of allele
frequencies at isozyme loci indicates five refugia during
the last glaciation. Most of the current populations originate
predominantly from two refugia in the Balkans and
in Central Italy, with extensive mixing where these two
lineages meet (Fig. 4). Another hypothetical refugium in
south-eastern France could also have contributed to the
isozyme polymorphism. The isolated populations in the
Pyrenees and in Calabria correspond to two additional
refugia, but did not expand.
Picea abies (Lagercrantz & Ryman 1990)
Norway spruce presents a phylogeographic pattern completely
different from the other species examined here,
456 P. TABERLET ET AL.
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
because it prefers cold climate and was not present in usual
southern refugia (R1, R2 and R3 in Fig. 1). Isozyme polymorphism
and fossil pollen analyses suggest postglacial
westward colonization routes from three refugia in the present-day
area of Moscow, in the Carpathians, and in the
Dinaric Alps (Fig. 4) (Huntley & Birks 1983). Population
bottlenecks in the two last refugia could account for the very
low level of heterozygosity observed in Central Europe.
Fagus sylvatica (Demesure et al. 1996)
Common beech exhibits an outstanding homogeneity in
chloroplast DNA amongst populations from northern
Spain, France, Germany, Poland, and the Balkans. Only
populations from Italy are distinct. The interpretation of
these results together with fossil pollen data (Huntley &
Birks 1983), which clearly localized the refugia in southern
COMPARATIVE PHYLOGEOGRAPHY IN EUROPE 457
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
Fig. 2 Simplified phylogeographies of
Ursus arctos, Sorex spp., Crocidura
suaveolens, Arvicola spp., and Triturus spp.
in Europe deduced from Taberlet &
Bouvet (1994), Taberlet et al. (1994),
Cosson et al. (unpublished), Wust-Saucy
et al. (unpublished), and Wallis & Arntzen
(1989), respectively. The lineages taken
into account in the Brooks parsimony
analysis are indicated in square brackets.
Italy and in the Balkans, allows us to infer the postglacial
colonization routes (Fig. 4): all of the current geographical
range except Italy was colonized from a refugium located
in the Balkans (Carpathes), and the populations originating
from a refugium in Calabria were confined to Italy.
Quercus spp. (Dumolin-Lapègue et al. 1997)
Study of chloroplast DNA variation in eight species of
white oaks, which hybridize and share the same set of
haplotypes, suggests some postglacial colonization routes
originating from the three potential refugia in Spain, Italy,
and the Balkans (Fig. 4). An extensive mixing of different
lineages has been observed in the northern populations,
while some haplotypes did not expand outside of these
refugia.
Results of the Brooks parsimony analysis
Five areas were taken into account for the Brooks parsimony
analysis. Three of them correspond to the three
potential refugia indicated in Fig. 1: Portugal–Spain, Italy,
and the Balkans. Only two areas have been defined in the
remaining regions and have been called France, and
Germany–Poland, although they can include some other
bordering areas. Due to the scarcity of available data, we
were not able to consider a larger number of distinct
areas, such as the British Isles, Fennoscandia, and the
European portion of Russia. Furthermore, the lack of geographical
accuracy of the available data set prevented us
from defining more distinct areas in south-western and
Central Europe. The lineages considered are all quoted in
square brackets on Figs 2, 3, and 4. The construction of the
data matrix (Table 2) involved some inevitable simplifications
according to the chosen areas. For example, the lineage
[T3] (Triturus cristatus) was not considered as present
in France, despite its presence in the north-east portion of
this area, and the lineage [F2] (Fagus sylvatica) was not
considered as present in Italy, despite its limited occurrence
in the northern part of the peninsula.
The parsimony analysis gave a single most parsimonious
tree 39-steps long (consistency index: 0.821). The
results are shown on an unrooted area phylogram where
assigned branch lengths and bootstrap values (10 000 replicates)
are indicated (Fig. 5). The grouping of
Portugal–Spain/France vs. Germany–Poland/Italy/
Balkans is supported by a bootstrap value of 97%. Another
feature of this area phylogram is that Italy is connected via
a very long branch length, whereas Portugal–Spain and the
Balkans are connected via shorter branch lengths.
Discussion
Lack of congruence among phylogeographic patterns
Ten taxa, including mammals, amphibians, insects, and
plants, have been compared for phylogeographic patterns
458 P. TABERLET ET AL.
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
Fig. 3 Post-glacial colonization routes for Chorthippus parallelus
(redrawn from Cooper et al. 1995). The lineages taken into
account in the Brooks parsimony analysis are indicated in square
brackets.
Table 2 Data matrix of the Brooks parsimony analysis. The names of lineages or populations correspond to that indicated in square
brackets in Figs 2, 3, and 4.
Lineages or populations
Areas U1 U2 U3 S1 S2 S3 S4 S5 Sx C1 C2 A1 A2 A3 A4 Ax T1 T2 T3 T4 T5 Tx Ty Tz G1 G2 G3 F1 F2 Q1 Q2 Q3
Portugal–Spain 1 0 0 1 0 0 1 0 1 1 0 0 1 0 1 1 0 0 0 1 1 0 0 1 1 0 0 0 1 1 0 0
Italy 0 1 0 0 0 1 0 1 0 0 1 0 0 1 0 0 0 1 0 0 0 1 1 0 0 1 0 1 0 0 1 0
Balkans 0 1 1 0 1 0 0 0 1 0 1 1 0 0 0 1 0 0 1 0 0 0 1 0 0 0 1 0 1 0 0 1
France 1 0 0 0 1 0 1 0 1 1 0 0 1 0 1 1 0 0 0 1 0 0 0 1 0 0 1 0 1 1 1 1
Germany–Poland 1 0 1 0 1 0 0 0 1 0 1 1 0 0 0 1 1 0 0 0 0 1 1 0 0 0 1 0 1 1 1 1
across Europe. The most striking result is the considerable
dissimilarity among European-wide phylogeographic
patterns, although the divergences amongst populations
from the three refugia could be considered as a certain
degree of concordance. It seems that each taxon has
responded independently to Quaternary cold periods,
and therefore is largely a unique case with its own history.
For example, if we compare lineages present in Italy and
in the Iberic peninsula, they are closely related in Ursus
(less than 1% of sequence divergence in the cytochrome b
gene), but much more distantly related in Crocidura
(6.4%), in Arvicola (7.6%), and in Triturus (8.5%), while the
Sorex species considered here exhibit two lineages in each
of these two refugia. Populations occurring in France
come either from a refugium in the Iberic peninsula (e.g.
Arvicola sapidus, Triturus marmoratus), or from a refugium
in the Balkans (e.g. Chorthippus parallelus, Fagus sylvati-
cus).
Although some spatial congruence has previously been
found elsewhere in some restricted areas (Avise 1992;
Bermingham & Avise 1986; Joseph et al. 1995), the results
obtained in Europe and in North America (Zink 1996)
suggest that congruence is the exception at the continental
scale. The consequence of an independent history for each
taxon is that assemblages of plants and animals comprising
particular communities are not stable over time, an
observation consistent with previous findings based
mainly on fossil pollen data (Bennett 1990).
In the context of repeated contraction/expansion of geographical
ranges during Quaternary climatic fluctuations,
the differing rates of colonization amongst taxa probably
played an important role and could be at least partially
responsible for the incongruences. It is likely that the first
population colonizing a particular region can prevent the
arrival of another one because the space is already occupied,
and introgression or elimination by competition is
always much slower than colonization of free areas.
Extinction of a population in a potential refugium during a
cold period, followed by a recolonization by another one,
could also explain some incongruences.
COMPARATIVE PHYLOGEOGRAPHY IN EUROPE 459
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
Fig. 4 Post-glacial colonization routes for
four tree species: Picea abies, Abies alba, Fagus
sylvatica, Quercus spp., deduced or redrawn
from Lagercrantz & Ryman (1990), Konnert
& Bergmann (1995), Demesure et al. (1996),
and Dumoulin-Lapègue et al. (1997). The
lineages or populations taken into account
in the Brooks parsimony analysis are
indicated in square brackets.
The depth of intraspecific phylogenies
In our data set, we were able to estimate the depth of
intraspecific phylogenies between populations from two
refugia in Portugal–Spain and in the Balkans, for three
taxa. These depths in mtDNA sequence divergence for
Ursus arctos, Crocidura suaveolens, and Arvicola sp. are 2%,
6.4%, and 7.6%, respectively (Fig. 2). Assuming approximately
equal rates of mtDNA sequence divergence as a
working hypothesis (2% per Myr; Wilson et al. 1985), then
if the depth in phylogenies was related to a common
event, we may expect divergence times dating from the
beginning of the strong climatic oscillations, i.e. 0.7 Mya.
Our results do not reveal any evidence of a common time
of divergence within these three taxa, except maybe for C.
suaveolens and Arvicola spp. A divergence of 6.4% or 7.6%
in the cytochrome b gene sequence for these two taxa even
indicates a pre-Pleistocene population split, showing that
Quaternary cold periods are unlikely to be a cause of the
divergence. Fortunately, palaeontological data are available
for these three species, indicating that U. arctos
arrived in Europe 0.91–0.98 Mya (Mazza & Rustioni 1994),
Mymomys davakosi, the ancestor of Arvicola, was present
4.0–3.4 Mya (Chaline & Graf 1988; Chaline et al. 1993),
while C. suaveolens was known on the continent from the
Plio-Pleistocene (Reumer 1983), i.e. approx. 2 Mya. One
can imagine that populations of each species began to
diverge among the three (subsequently refugium) areas,
even as early as the date of first arrival of the species in
Europe. The divergence may have occurred simply due to
isolation by distance among Portugal–Spain, Italy, and the
Balkans. The subsequent range contractions and expansions
due to climatic oscillations may have eliminated only
haplotypes present in northern areas, without affecting
those present in the refugia. Assuming no retention of
ancestral polymorphism from the date of arrival, this scenario
suggests that the depth of the deepest intraspecific
divergences are related simply to the date of arrival of each
taxon in the European continent, rather than to any subsequent
historical event. This could also explain why no congruence
was found in the dates of divergence even in cases
of congruent phylogeographic patterns. Indeed, several
authors (Avise 1992; Bermingham & Avise 1986; Joseph
et al. 1995) pointed out the discrepancy in divergence date
among lineages, and explained this either by a considerable
heterogeneity in mtDNA evolutionary rates, or by the
fact that the species reacted differently to the separate
episodes of climatic variations (i.e. the more ancient the
split, the more ancient the historical event involved in the
divergence). We argue instead that, in the case of phylogeographies
that are congruent due to the same historical
event, the divergence times amongst lineages are expected
to be different, and lie somewhere between the date of colonization
and the date of the historical vicariant event.
Consequently, the deepest divergence times between congruent
intraspecific phylogeographies are unlikely to
accurately date the historical event responsible for that
congruence, although they may provide a minimum age.
Recently, several studies have emphasized that speciation
events occurred mainly during the Pliocene
(Bermingham et al. 1992; Zink & Slowinski 1995). The
present study shows that this observation can be
extended to the intraspecific level, and that many lineages
within the same species can also stem from the
Pliocene. In our data set, only the genus Sorex could
have speciated during the Quaternary, but this could be
related more to chromosomal incompatibilities rather
than to genetic differentiation via gradual molecular
divergence (e.g. Hausser 1994).
Refugia, colonization routes, and suture zones
Almost all the considered taxa used the three potential
refuge areas in the south during Quaternary cold periods:
the Iberic peninsula, Italy, and the Balkans. Only the
Norway spruce and the common beech failed to follow
this rule. The Norway spruce, a tree from a boreal climate,
did not survive in the southernmost regions, but only in
suitable habitats in the Dinaric Alps, in the Carpathes, and
in the present-day area of Moscow (Huntley & Birks 1983;
Lagercrantz & Ryman 1990). The common beech, a tree
from a temperate climate, found refuge in the south as did
other taxa, but it seems that it was not able to survive in
the Iberic refugium; as a consequence, all the western part
of the geographical distribution was colonized from the
Balkans (Demesure et al. 1996). As predicted by the expansion/contraction
model, a much higher intraspecific
diversity in the southern areas than in recently colonized
460 P. TABERLET ET AL.
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
Fig. 5 Unrooted area phylogram illustrating the results of the
Brooks parsimony analysis for 32 different European lineages or
populations. Assigned branch lengths are given above each
branch with the minimum and maximum possible lengths in
square brackets. Bootstrap values are indicated in brackets.
northern regions has been documented for Triturus
carnifex in Italy and T. karelini in the Balkans (Wallis &
Arntzen 1989), for Chorthippus parallelus in the Balkans
(Cooper et al. 1995), and for Quercus spp. and Fagus sylvatica
in Italy (Demesure et al. 1996; Dumolin-Lapègue et al.
1997).
Although no evidence of common phylogeographic
histories across Europe have been found, postglacial colonization
routes exhibit some interesting concordances.
The colonization route out of an Iberic refugium towards
the south of Scandinavia seems to be identical for the
brown bear and the white oaks, and the spread from a
Balkanic refugium towards the southeast of France is similar
for the grasshopper and the common beech (Figs 2, 3,
and 4). Furthermore, the barrier of the Alps prevented the
northward expansion of populations isolated in the
Italian refugium for at least four taxa in our data set.
Even if some concordances in colonization routes are
apparent, the dissimilarity of the global phylogeographic
patterns makes the elaboration of a synthetic view difficult.
In this regard, the Brooks parsimony analysis is valuable
in that it integrates all the available data to produce
the single picture presented in Fig. 5. The long branch
length to Italy indicates how the scale of the barrier of the
Alps, and the isolation of the Italian peninsula.
Consequently, many lineages at the European level are
present only in this area (Figs 2, 3, and 4). Italy and the
Balkans form a clade, due probably to their geographical
proximity, and to the connection across the Adriatic sea
between these two refugia when the sea level was lower
during cold periods (see Fig. 1). Shorter branch lengths to
Portugal–Spain and to the Balkans convey the fact that
northern areas were colonized generally from these two
refugia. Although the Brooks parsimony analysis has
been criticized in the field of historical biogeography (see
Morrone & Crisci 1995), we feel that this approach is
useful in our case by giving a consistent overview of a
complex data set. Furthermore, using the parsimony
approach, it is possible to produce not only a branching
pattern, but also branch lengths which provide a valuable
information concerning the degree of isolation of a particular
region.
The term suture-zone has been introduced by
Remington (1968) to describe ‘a band, whether narrow or
broad, of geographical overlap between major biotic
assemblages, including some pairs of species or semispecies
which hybridize in the zone’. The emergence of this
term is due to the observation that hybrid zones are often
concentrated within limited areas. This concept could easily
be extended to the intraspecific level (see Jaarola &
Tegelström 1995) to characterize areas where different
populations meet after a postglacial expansion. In Europe
four main suture-zones may be recognized (Fig. 6). The
most obvious one corresponds to the Alpine barrier which
separates lineages occurring in Italy from lineages distributed
in the west and in the north of the Alps. Many
examples can be cited to illustrate this discontinuity: Sorex
araneus (Taberlet et al. 1994), Arvicola terrestris (WustSaucy
et al. 1997), Triturus sp. (Wallis & Arntzen 1989),
Salmo trutta (Bernatchez et al. 1992), C. parallelus (Cooper
et al. 1995), and Apis mellifera (Garnery et al. 1992).
However, some Italian lineages of the white oaks were
able to cross the Alps and to spread north-, east-, and
westward (Dumolin-Lapègue et al. 1997). The second
suture-zone corresponds to the junction between populations
of the Iberic refugium with populations of eastern
origin (the Balkanic refugium, or another easternmost
refugium) somewhere around the border between France
and Germany. As this region is relatively flat, without any
substantial topographic barrier, this suture zone is not
localized exactly in the same place for different organisms:
it occurs in the northeast of France between Arvicola
sapidus/A. terrestris, and Triturus marmoratus/T. cristatus,
and in a more eastern position for Sorex coronatus/S. araneus.
Previous morphological studies highlighted a suture
zone in the same area, for instance for the ringed snake,
Natrix natrix (Thorpe 1979). The third suture-zone is represented
by the Pyrenees, and could be considered as a
particular case of the second. In our dataset, only C. parallelus
illustrates this Pyrenean suture-zone, but genetic
data for the warbler Phylloscopus collybita (Helbig et al.
1993) also support it. The last suture-zone is located in
COMPARATIVE PHYLOGEOGRAPHY IN EUROPE 461
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
Fig. 6 Main postglacial colonization routes and subsequent
suture zones in Europe.
Scandinavia, and indicates that this area may have been
colonized from the north and from the south by different
populations originating from different refugia. So far, four
mammalian species have been shown to support this
suture zone: Sorex araneus (Fredga & Nawrin 1977),
Clethrionomys glareolus (Tegelström 1987), Microtus agrestis
(Jaarola & Tegelström 1995), and Ursus arctos (Taberlet
et al. 1995). As a consequence, the genetic diversity in
Scandinavia may be higher than expected.
Conclusions and perspectives
Previous studies have shown that the Quaternary in
Europe was characterized by many cycles of contraction/expansion
of geographical ranges according to climatic
fluctuations: contraction of ranges to southern
regions during cold periods, and expansion from the leading
edges during subsequent warmings (Hewitt 1996).
Using this model, one can predict that: (i) the highest
diversity should be found amongst the southern regions;
and (ii) the distribution of intraspecific polymorphism in
northern regions should be dictated by the colonization
routes used from the refugia. Clearly, the available phylogeographic
data confirm both predictions and strongly
support the contraction/expansion model.
Although this study revealed no evidence of common
phylogeographies amongst the 10 taxa compared, some
general trends can be drawn. The current distribution of
intraspecific polymorphism in Europe can be explained
by the persistence of each taxon in three refugia (Fig. 1)
and the subsequent colonization routes used (Fig. 6). The
northern part of Europe has been colonized primarily
from the Iberic and the Balkan refugia, whilst populations
evolving in Italy were not usually able to spread northward
due to the Alpine Barrier. The comparison of colonization
routes suggested four main suture-zones where
populations from different refugia meet (Fig. 6). However,
it is noteworthy that the few data currently available for
eastern Europe and Fennoscandia may underestimate the
contribution of potential easternmost refugia localized in
Europe and/or Asia.
These general trends have implications for conservation
genetics, highlighting areas where conservation
efforts should be concentrated. First, Italy has many
endemic lineages. Second, the southern regions in general,
where refugia are localized, are of particular interest:
they support most of the current genetic variation, and for
long-term conservation the preservation of genetic diversity
in these areas seems desirable.
Traditionally, the history of species and communities
over the last few million years was studied by using
fossils and pollen data. As a result, the influence of
Quaternary climatic oscillation has been estimated, and a
general picture concerning the changes in faunas and
floras has been put forward. Molecular genetics can
describe intraspecific geographical structure by identifying
lineages, and consequently can reveal postglacial
colonization routes provided that the location of the
refugia are known. Palaeontology and palynology are not
able to discern intraspecific phylogeographic structure
and, in that, the molecular approach clearly provides an
advance. However, molecular genetics based on polymorphism
present in extant organisms cannot by itself localize
refugia with precision. Furthermore, and contrary to
widespread current belief, molecular genetic studies may
be of limited use in dating a vicariant event that occurred
during the Quaternary. Indeed, the intraspecific divergence
amongst mtDNA lineages appears to often largely
predate the Quaternary cold periods.
This study is a first attempt to synthesize our current
knowledge, based on the limited number of available data
sets. Although incomplete, it will help to define the
appropriate scope for future phylogeographic studies in
this region, and to design more pertinent sampling strategies.
Advances in the field of comparative phylogeography
will clearly come from a better integration of several
complementary approaches including genetics, ecology,
palaeontology, palynology, and climatology.
Acknowledgements
We wish to thank E. Bermingham, S. Lavery, C. Moritz, C.
Schneider, L. Waits, and two anonymous reviewers for valuable
comments on this work. We also thank R. Petit, B. Demesure, and
S. Dumolin-Lapègue for kindly giving us submitted and in-press
manuscripts. This study was partly funded by the Université
Joseph Fourier (Grenoble, France), the Centre National de la
Recherche Scientifique, the Institut National de la Recherche
Agronomique, and was undertaken as part of an agreement in
research collaboration between the Université Joseph Fourier
and the Université de Lausanne. L.F. was supported by a postdoctoral
fellowship of the Swiss National Science Foundation
and the Agassiz Foundation. A.-G.W.-S. was supported by a
Heim–Vögtlin grant of the Swiss National Science Foundation.
References
Avise JC (1992) Molecular population structure and the biogeographic
history of a regional fauna: a case history with lessons
for conservation biology. Oïkos, 63, 62–76.
Avise JC, Arnold J, Ball RM, et al. (1987) Intraspecific phylogeography:
the mitochondrial DNA bridge between population
genetics and systematics. Annual Review of Ecology and
Systematics, 18, 489–522.
Bennett KD (1986) The rate of spread and population increase of
forest trees during the postglacial. Philosophical Transactions of
the Royal Society of London Series B, 314, 523–531.
Bennett KD (1990) Milankovitch cycles and their effects on
species in ecological and evolutionary time. Paleobiology, 16,
11–21.
Bennett KD, Tzedakis PC, Willis KJ (1991) Quaternary refugia of
462 P. TABERLET ET AL.
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
north European trees. Journal of Biogeography, 18, 103–115.
Bermingham E, Avise JC (1986) Molecular zoogeography of
freshwater fishes in the southeastern United States. Genetics,
113, 939–965.
Bermingham E, Rohwer S, Freeman S, Wood C (1992) Vicariance
biogeography in the Pleistocene and speciation in North
American wood warblers: a test of Mengel’s model.
Proceedings of the National Academy of Sciences of the United
States of America, 89, 6624–6628.
Bernatchez L, Guymard R, Bonhomme F (1992) DNA sequence
variation of the mitochondrial control region among geographically
and morphologically remote European brown
trout Salmo trutta populations. Molecular Ecology, 1, 161–173.
Bond G, Broecker W, Johnsen S, et al. (1993) Correlations between
climate records from North Atlantic sediments and
Greenland ice. Nature, 365, 143–147.
Brooks DR (1985) Historical ecology: a new approach to studying
the evolution of ecological associations. Annals of the Missouri
Botanical Garden, 72, 660–680.
Brooks DR (1990) Parsimony analysis in historical biogeography
and coevolution: methodological and theoretical update.
Systematic Zoology, 39, 14–30.
Chaline J, Graf J-D (1988) Phylogeny of the Arvicolidae
(Rodentia): biochemical and paleontological evidence. Journal
of Mammalogy, 69, 22–33.
Chaline J, Laurin B, Brunet-Lecomte P, Viriot L (1993)
Morphological trends and rates of evolution in Arvicolids
(Arvicolidae, Rodentia): towards a punctated equilibria/disequilibria
model. Quaternary International, 19, 27–39.
Cooper SJ, Ibrahim KM, Hewitt GM (1995) Postglacial expansion
and genome subdivision in the European grasshopper
Chorthippus parallelus. Molecular Ecology, 4, 49–60.
Dansgaard W, Johnsen SJ, Clausen HB, et al. (1993) Evidence for
general instability of past climate from a 250-kyr ice-core
record. Nature, 364, 218–220.
De Candolle AP (1820) Géographie botanique. In: Dictionnaire des
Sciences Naturelles, pp. 359–422.
Demesure B, Comps B, Petit RJ (1996) Chloroplast DNA phylogeography
of the common beech (Fagus sylvatica L. ) in
Europe. Evolution, 50, 2515–2520.
Dumolin-Lapègue S, Demesure B, Fineschi S, Le Corre V, Petit RJ
(1997) Phylogeographic structure of white oaks throughout
the European continent. Genetics 146, 1475–1487.
Estoup A, Solignac M, Cornuet J-M, Goudet J, Scholl A (1996)
Genetic differentiation of continental and island populations
of Bombus terrestris (Hymenoptera: Apidae) in Europe.
Molecular Ecology, 5, 19–31.
Fredga K, Nawrin J (1977) Karyotype variability in Sorex araneus
L. (Insectivora, Mammalia). In: Chromosome Today (eds
DeLaChapelle A, Sorsa M), pp. 153–161. Elsevier/North
Holland Biomedical Press, Amsterdam.
Frenzel B, Pécsi M, Velichko AA (1992) Atlas of Paleoclimates And
Paleoenvironments Of The Northern Hemisphere. Late Pleistocene–
Holocene. Geographical Research Institute, Hungarian
Academy of Sciences. Gustav Fischer Verlag, Budapest,
Stuttgard.
Garnery L, Cornuet J-M, Solignac M (1992) Evolutionary history
of the honey bee Apis mellifera inferred from mitochondrial
DNA analysis. Molecular Ecology, 1, 145–154.
Garnery L, Mosshine EH, Oldroyd BP, Cornuet J-M (1995)
Mitochondrial DNA variation in Moroccan and Spanish
honey bee populations. Molecular Ecology, 4, 465–471.
GRIP Members (1993) Climate instability during the last interglacial
period recorded in the GRIP ice core. Nature, 364,
203–207.
Hardy C, Callou C, Vigne J-D, et al. (1995) Rabbit mitochondrial
DNA diversity from prehistoric to modern times. Journal of
Molecular Evolution, 40, 227–237.
Hausser J (1994) The Sorex of the araneus-arcticus group
(Mammalia: Soricidae): do they actually speciate? In:
Advances in the Biology Of Shrews (eds Merritt JF, Kirkland GLJ,
Rose RK), pp. 295–306. Carnegie Museum of Natural History,
Special publication 18, Pittsburgh.
Hays JD, Imbrie J, Shackleton NJ (1976) Variations in the Earth’s
orbit: pacemaker of the ice ages. Science, 194, 1121–1132.
Helbig J, Salomon M, Wink M, Bried J (1993) Absence de flux
génique mitochondrial entre les Pouillots ‘véloces’ médioeuropéen
et ibérique (Aves: Phylloscopus collybita collybita, P.
(c.) brehmii); implications taxinomiques. Résultats tirés de la
PCR et du séquençage d’ADN. Comptes Rendus de l’Académie
des Sciences Série III, 316, 205–210.
Hewitt GM (1996) Some genetic consequences of ice ages, and
their role in divergence and speciation. Biological Journal of the
Linnean Society, 58, 247–276.
Huntley B, Birks HJB (1983) An Atlas Of Past And Present Pollen
Maps For Europe: 0–13000 Years Ago. Cambridge University
Press, Cambridge.
Jaarola M, Tegelström H (1995) Colonization history of north
European field voles (Microtus agrestis) revealed by mitochondrial
DNA. Molecular Ecology, 4, 299–310.
Joseph L, Moritz C, Hugall A (1995) Molecular support for vicariance
as a source of diversity in rainforest. Proceedings of the
Royal Society of London Series B, 260, 177–182.
Kohn M, Knauer F, Stoffella A, Schröder W, Pääbo S (1995)
Conservation genetics of the European brown bear – a study
using excremental PCR of nuclear and mitochondrial
sequences. Molecular Ecology, 4, 95–103.
Konnert M, Bergmann F (1995) The geographical distribution of
genetic variation of silver fir (Abies alba, Pinaceae) in relation to
its migration history. Plant Systematics and Evolution, 196,
19–30.
Lagercrantz U, Ryman N (1990) Genetic structure of Norway
spruce (Picea abies): concordance of morphological and
allozymic variation. Evolution, 44, 38–53.
Lundqvist J, Saarnisto M (1995) Summary of project IGCP-253.
Quaternary International, 28, 9–18.
Mazza P, Rustioni M (1994) On the phylogeny of Eurasian bears.
Palaeontographica Abstracts A, 230, 1–38.
Morrone JJ, Crisci JV (1995) Historical biogeography: introduction
to methods. Annual Review of Ecology and Systematics, 26,
373–401.
Randi E, Gentile L, Boscagli G, Huber D, Roth HU (1994)
Mitochondrial DNA sequence divergence among some west
European brown bear (Ursus arctos L. ) populations. Lessons
for conservation. Heredity, 73, 480–489.
Remington CL (1968) Suture-zones of hybrid interaction between
recently joined biotas. Evolutionary Biology, 2, 321–428.
Reumer JWF (1983) Ruscinian and early Pleistocene Soricidae
(Insectivora, Mammalia) from Tegele (The Netherlands) and
Hungary. PhD thesis, University of Leiden.
Roy K, Valentine JW, Jablonski D, Kidwell SM (1996) Scales of climatic
variability and time averaging in Pleistocene biotas:
implications for ecology and evolution. Trends in Ecology and
Evolution, 11, 458–453.
COMPARATIVE PHYLOGEOGRAPHY IN EUROPE 463
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464
Smith DR, Palopoli MF, Taylor BR, et al. (1991) Geographical
overlap of two mitochondrial genomes in Spanish honeybees
(Apis mellifera iberica). Journal of Heredity, 82, 96–100.
Swofford DL (1993) Phylogenetic Analysis Using Parsimony Version
3.1.1. Illinois Natural History Survey, Champaign, Illinois.
Taberlet P, Bouvet J (1992) Génétique de l’Ours brun des Pyrénées
(Ursus arctos): premiers résultats. Comptes Rendus de
l’Académie des Sciences Série III, 314, 15–21.
Taberlet P, Bouvet J (1994) Mitochondrial DNA polymorphism,
phylogeography, and conservation genetics of the brown bear
(Ursus arctos) in Europe. Proceedings of the Royal Society of
London Series B, 255, 195–200.
Taberlet P, Fumagalli L, Hausser J (1994) Chromosomal versus
mitochondrial DNA evolution: tracking the evolutionary history
of the southwestern European populations of the Sorex
araneus group (Mammalia, Insectivora). Evolution, 48,
623–636.
Taberlet P, Swenson JE, Sandegren F, Bjärvall A (1995)
Localization of a contact zone between two highly divergent
mitochondrial DNA lineages of the brown bear (Ursus arctos)
in Scandinavia. Conservation Biology, 9, 1255–1261.
Tegelström H (1987) Transfer of mitochondrial DNA from the
northern red-backed vole (Clethrionomys rutilus) to the bank
vole (C. glareolus). Journal of Molecular Evolution, 24, 218–227.
Thomaz D, Guiller A, Clarke B (1996) Extreme divergence of
mitochondrial DNA within species of pulmonate land snails.
Proceedings of the Royal Society of London Series B, 263, 363–368.
Thorpe RS (1979) Multivariate analysis of the population systematics
of the ringed snake, Natrix natrix (L. ). Proceedings of the
Royal Society of Edinburgh, 78B, 1–62.
Wallis GP, Arntzen JW (1989) Mitochondrial–DNA variation in
the crested newt superspecies: limited cytoplasmic gene flow
among species. Evolution, 43, 88–104.
Webb T, Bartlein PJ (1992) Global changes during the last 3
million years: climatic controls and biotic response. Annual
Review of Ecology and Systematics, 23, 141–173.
Wiley EO (1988a) Parsimony analysis and vicariance biogeography.
Systematic Zoology, 37, 271–290.
Wiley EO (1988b) Vicariance biogeography. Annual Review of
Ecology and Systematics, 19, 513–542.
Wilson AC, Cann RL, Carr SM, et al. (1985) Mitochondrial DNA
and two perspectives on evolutionary genetics. Biological
Journal of the Linnean Society, 26, 375–400.
Zink RM (1996) Comparative phylogeography in North
American birds. Evolution, 50, 308–317.
Zink RM, Slowinski JB (1995) Evidence from molecular systematics
for decreased avian diversification in the Pleistocene
Epoch. Proceedings of the National Academy of Sciences of the
United States of America, 92, 5832–5835.
This synthesis forms part of a continuing programme on
intraspecific phylogeographic studies of animals and plants carried
out in the Laboratoire de Biologie des Populations
d’Altitude (Grenoble, France). The four authors are currently
working in four different laboratories, but share a common interest
in studying intraspecific phylogeographies in Europe. Pierre
Taberlet is also involved in conservation genetics and in developing
new molecular techniques for population studies; Luca
Fumagalli is working in the fields of conservation genetics and
molecular evolution; Anne-Gabrielle Wust-Saucy is studying the
phylogeography of the genus Arvicola; Jean-François Cosson
recently joined the Laboratoire ‘Génome and Populations’
(Montpellier, France) and is interested in the phylogeography of
the genus Crocidura.
464 P. TABERLET ET AL.
© 1998 Blackwell Science Ltd, Molecular Ecology, 7, 453–464