M ar Biol (2007) 151:1275-1286
D O I 10.1007/S00227-006-0556-1
R E S E A R C H A R T IC L E
Population structure and historical demography of the thorny
skate (Amblyraja radiata, Rajidae) in the North Atlantic
Malia Chevolot • Peter H. J. Wolfs • Jónbjöm Pálsson • Adriaan D . Rijnsdorp •
Wytze T. Stam • Jeanine L. Olsen
R eceived: 26 May 2006/ A ccepted: 9 N ovem ber 2006/P u b lish ed online: 19 D ecem ber 2006
© Springer-V erlag 2006
Abstract Population genetic structure of the thorny
skate (Am blyraja radiata) was surveyed in >300 indi­
viduals sam pled from N ew foundland, Iceland, Norway,
the K attegat and the central N orth Sea. A 290-bp
fragm ent of the m t cytochrom e-b gene was first
screened by SSCP. Sequences of SSCP haplotypes re ­
vealed 34 haplotypes, 14 of which w ere unique to
Iceland, 3 to N ew foundland, 1 to N orw ay and 3 to the
K attegat. The global F St was w eak but significant.
R em oval of the two K attegat locations, which were the
m ost differentiated, resulted in no significant genetic
differentiation. H aplotype diversity was high and
evenly distributed across the entire A tlantic (h = 0.8)
w ith the exception of the N o rth Sea (h = 0.48). S ta­
tistical parsimony revealed a star-like genealogy with a
central widespread haplotype. A subsequent nested
clade analysis led to the inference of contiguous
expansion with evidence for long distance dispersal
betw een N ew foundland and Iceland. H istorical
C om m unicated by O. K inne, O ldendorf/L uhe.
M. C hevolot (E l) • P. H. J. W olfs • W. T. Stam ■
J. L. Olsen
D ep artm en t of M arine B enthic Ecology a n d Evolution,
C enter for Ecological and E volutionary Studies, Biological
C entre, U niversity o f G roningen, P O B ox 14, 9750 A A
H a ren , T he N etherlands
e-mail: m.s.c.o.m.chevolot@ rug.nl
J. Pálsson
M arine R esearch Institute, P O B ox 1390, 121 Reykjavik,
Iceland
A . D . Rijnsdorp
W ageningen Institute for M arine R esearch and Ecological
Studies (IM A R E S ), A nim al Sciences G roup, W ageningen
U R , PO B ox 68, 1970 A B IJm uiden, T he N etherlands
dem ographic analysis showed that thorny skates have
undergone exponential population expansion that
started betw een 1.1 million and 690,000 years ago; and
that the L ast Glacial M aximum apparently had little
effect. T hese results strongly differ from those of a
parallel study of the thornback ray (Raja clavata) in
which clear structure and form er réfugiai areas could
be identified. A lthough both species have similar life
history traits and overlapping ranges, the continental
shelf edge apparently does not present a barrier to
m igration in A . radiata, as it does for R. clavata.
Keywords Am blyraja radiata • T horny skate •
R ajidae - Elasm obranchs ■C ytochrom e b •
M itochondrial D N A ■Population structure •
Iceland ■A tlantic
Introduction
T he thorny skate (.Am blyraja radiata; Rajiformes: R a­
jidae) is widely distributed throughout the N orth
A tlantic region from H udson Bay to South Carolina,
U SA in th e west, G reenland, Iceland and Spitzbergen
in the north, and from Norway to the southern N orth
Sea (including the western Baltic) in the east. Thorny
skates live in shallow coastal waters but extend their
habitat to a depth of 1,000 m. T heir life cycle is ovip­
arous, producing 20-80 eggs p er fem ale each year.
T here is no passive pelagic larval state; rather, fully
form ed rays hatch after several m onths (ca. 8-11 cm
disc width) and m ature at 5-6 years (W alker 1998).
T he species seems to be reproductively active all year
round in areas where the reproductive biology has
been studied (i.e. in th e G ulf of M aine Sulikowski et al.
C l Springer
M a r B iol (2007) 151:1275-1286
1276
2005, N o rth Sea W alker 1998). Philopatry of rays and
skates has also been rep o rted and has indicated rela­
tively small hom e ranges. Tagging studies in U K w a­
ters, for example, have shown th at 85% of the
individuals rem ained within a 110 km area (W alker
et al. 1997; Dulvy et al. 2000; H eessen 2004). These
observations, com bined with the life history traits,
suggest th at population genetic differentiation may be
relatively strong in A . radiata, as com pared with fish
characterized by high fecundity, long pelagic larval
stages and large m igratory ranges. G iven th at A . radi­
ata is characterized by low fecundity, slow growth rate,
and late maturity; com bined w ith its vulnerability to
over-exploitation (B rander 1981; H eist 1999), an
assessm ent of population genetic structure over its
range as com pared with o th er skates/rays is o f interest
to m anagem ent and conservation. It is also of interest
in the broader phylogeographic and dem ographic
context of historical population growth.
Distributions of the N o rth A tlantic and northern
E uropean m arine biota have b een drastically affected
by the Pleistocene glacial/interglacial cycles over the
p ast 2.4 Myr. T he last glacial m aximum (L G M ), which
occurred ca. 20,000 years ago, has significantly shaped
contem porary distributions of b oth terrestrial (H ewitt
2000) and shallow-water, m arine organisms (Cunning­
ham and Collins 1998). M arine populations either be­
cam e extinct o r w ere forced to re tre a t (usually
southw ard) into one o r m ore réfugiai areas. A s the ice
retreated , populations expanded and recolonized areas
previously covered by ice. Ice-sheet extension also
drastically affected sea-level and coastlines; and the
Baltic and th e N orth Sea effectively did not exist
(Frenzel et al. 1992). Thus, for boreal species, potential
refugia are postulated to have occurred at about the
latitude of N ew foundland (47°N) in the western
A tlantic and N orthern Portugal/Spain (43°N) in the
eastern A tlantic. In addition, th ere is som e paleoclim atic evidence for a coastal, ice-free zone in southern
Iceland (R undgren and Ingolfsson 1999; Bingham et al.
2003) which is also supported by som e genetic studies
o f benthic m arine invertebrates (W ares and Cunning­
ham 2001; Addison and H a rt 2005; G ovindarajan et al.
2005). The degree to which som e dem ersal skates and
rays were affected represents an interm ediate case
betw een sessile, benthic organism s and widely foraging
pelagic organisms.
M itochondrial D N A (m tD N A ) sequences are
appropriate for assessing genetic population structure,
phylogeography and in m aking inferences about
underlying historical dem ographic processes th at have
shaped present-day structure (Avise 2000). Use of
m tD N A has th e fu rther advantage th a t its effective
£ ) Springer
population size ((Ve) is four tim es sm aller than nuclear
D N A N &due to its haploid nature and generally uni­
parental inheritance (B irky et al. 1989). Thus, the ef­
fect of genetic drift is stronger and a higher level of
population differentiation can be observed with
m tD N A than with nu clear D N A (e.g., H oarau et al.
2004). This can be o f great im portance in populations
that do not reach m igration/genetic drift equilibrium,
as in such areas w here recolonization history is recent.
T h e aim of the p re se n t study was to survey regional
population genetic an d phylogeographic structure in
the thorny skate (A. radiata) in northern A tlantic. W e
focused on four questions: (1) to w hat extent are A.
radiata’s life history traits a good predictor of popula­
tion differentiation; (2) how has recent clim ate history
shaped the regional distribution of A . radiata; (3) to
w hat extent are historical im prints of refugia, recol­
onization and dem ographic expansion detectable; and
(4) how do these results com pare to those for the
thornback ray, Raja clavata (C hevolot e t al. 2006b)?
Materials and methods
Sampling and D N A extraction
A total of 337 rays was sam pled from 13 locations
during various bottom trawl surveys conducted be­
tween 2003 and 2004 (Fig. 1; T able 1). In order to
obtain reasonable sam ple sizes, hauls w ere pooled in
som e cases (See T able 1). Pooling rules w ere guided by
tagging studies which suggested that individuals gen­
erally rem ained w ithin a 110-km area. Using a con­
servative approach, w e pooled hauls th at were close
together, i.e., <50 km apart. A t som e locations, adults
and im m ature individuals were caught together. In
such cases, they w ere separated to avoid tem poral
adm ixture (W aples 1998). These two m aturity stages
w ere distinguished based on reproductive criteria, i.e.
presence of fully differentiated shell glands for females
and fully developed testes and claspers for males
(Stehm ann 1995). M uscle tissue was collected from
each individual and preserved in 70% ethanol. Total
genom ic D N A was extracted using either a modified
CTAB (H oarau e t al. 2002) or a silica-based extraction
protocol (Elphinstone et al. 2003).
SSCP and sequencing
A 290-bp fragm ent of the cytochrom e b (cytb) was
amplified by P C R using the prim ers A rC b-F (5'CACA G A T A A A A T C C C A T T T 3'), fluorescently 5' labeled
with 6-FAM and Cb-R (5'C C G C C C A A TC A C T-
M ar Biol (2007) 151:1275-1286
1277
Fig. 1 Sam pling locations for
A m b lyra ja radiata (See
T able 1 fo r abbreviations)
C A A A C C 3'), fluorescently 5' labeled with H EX . P C R
reactions w ere perform ed in a 10 pi total volum e
containing 1-3 pi of extracted D N A (<1 ng),
1 X reaction B uffer (Prom ega), 0.2 m M of each dN TP,
0.25 U Taq D N A polym erase (Prom ega), 2 mM
M gCl2, 0.5 pM o f A rC b-F and 0.64 pM of Cb-R. P C R
amplifications w ere perform ed with either a PTC100™ therm ocycler (M J R esearch, Inc.) o r M astercycler gradient cycler (Eppendorf). P C R conditions
were: initial dénaturation for 1 min a t 94°C; followed
by 32 cycles of: d énaturation for 30 s a t 94°C, annealing
a t 49°C for 30 s, and extension at 72°C for 1 m in 30,
and follow ed b y a final extension step a t 72°C for
10 min.
Single stran d conform ation polym orphism (SSCP)
(O rita e t al. 1.989; Sunnucks et al. 2000) was used to
screen fo r sequence variation in PC R fragm ents of
equal length. P oint m utations affect conform ation of
the single D N A strand, which can be revealed on n o n ­
denaturing polyacrylam ide gels. SSCP gels w ere run on
an A B I Prism-377 autom atic sequencer (A pplied B io­
systems) as described in Coyer et al. (2002), except
th at we used 0.3 x M D E concentration and added 5%
glycerol to th e gel. Because m utations can affect the
mobility of one or b oth strands differently (Lescasse
1999), separate labeling of each strand increases the
sensitivity to d etect differences.
A ll gels w ere analyzed independently and all unique
haplotypes d etected by SSCP on each gel w ere se­
quenced. W hen m ore than five individuals showed the
sam e SSCP haplotype, a t least two individuals w ere
sequenced. P C R products were cleaned with E xoSapIt
(USB C orporation) enzym e following the providers’
recom m endations. B oth strands w ere sequenced using
th e Big D ye T erm in ato r Kit (A pplied Biosystems) and
run on an A B I Prism-377 autom atic sequencer (A pplied
Biosystems). Sequences w ere edited using BioEdit
v7.0.1 (Haii 1999) and aligned with ClustalW and by eye.
D ata analysis
H aplotype (h) and nucleotide (n) diversities (Nei 1987)
w ere estim ated using D naSP v.3.53 (Rozas and Rozas
1999). Population differentiation was assessed using
W right’s Fsx (W right 1969). G lobal and pairwise FST
w ere estim ated using th e W eir and C ockerham (1984)
6 estim ator with G enetix 4.05 (Belkhir e t al. 2004) and
significance was tested with 3,000 perm utations.
Intraspecific relationships of m tD N A haplotypes
w ere inferred using a statistical parsim ony with the
software tes v.l. 13 (C lem ent et al. 2000). T he resulting
netw ork was used for a N ested Clade Analysis (NCA)
to test for geographical association betw een haplotypes
(or nested clades) and geographical distribution
(Tem pleton et al. 1995). T he idea is to distinguish be­
tw een past (fragm entation, expansion) and contem po­
rary processes as tip clades are expected to be more
recent than interior clades (D onnelly and Tavare
1986). T he first step is to nest the statistical parsimony
netw ork following T em pleton’s nesting rules (Tem­
pleton et al. 1995; T em pleton and Sing 1993): H aplo­
types (“ 0-step clades”) separated by a single m utation
are grouped together into a “ one-step clade” pro­
ceeding from the tips to the interior of the network;
then “ one-step clades” , separated by a single m utation,
are grouped together in “ two-step clades” . T he clus­
tering is continued until the next level of nesting would
encompass the entire tree. A m biguities (intercon­
nected haplotypes forming a closed loop due to mul­
tiple parsim onious interconnections) w ere resolved
Springer
1278
M ar B iol (2007) 151:1275-1286
Table 1 Sam pling locations of A m blyraja radiata
Code L atitude
(deg)
L ongtitude
(deg)
N a M aturity
stageb
Sam pling
d a te
Newfoundland N ew foundland
Nf
46.07
-47.37
30
A (-A )
O ff
B jargtangar
Bj
65.74
-27.4
32
I(-I)
O c to b e r 2004 D F O Fall 10 0.834 0.0070
survey
O c to b e r 2003 ICBTS
9
0.802 0.0055
A
O c to b e r 2003
11
0.749 0.0060
Is
Hu
Skj
-27.56
-27.97
-28.38
-27.98
-27.89
-27.69
-27.50
-27.03
-23.20
-21.08
-17.54
22
Isafjardarjüp
H unafloi
Skjalfandi
65.67
65.42
65.73
65.73
65.79
65.76
65.75
65.84
66.28
65.57
66.14
Ey
Ska
Ox
hi
5
8
9
9
6
10
11
11
0.750
0.738
0.743
0.824
0.784
0.784
0.916
0.916
O ff H öfn
Ho
-18.43
-16.79
-16.28
-13.48
-13.99
-12.53
-12.70
-12.82
-12.60
-12.38
-12.90
2.92
3.14
10.35
A
A
I
A
A
I
I
A
O c to b e r 2003 IC B T S
M arch2004
ICBTS
M a rc h 2004
ICBTS
E yjafjördur
S kagafjördur
O xafjórdur
O ff jaistlfjördur
17
27
25
18
19
19
20
24
17
A
O c to b e r 2003 ICBTS
6
0.806 0.0036
15
I
February2003 IBTS
7
0.771 0.0039
4.81
22
19
11
I
A
I
F e b ru a ry
2003
F eb ru ary
2003
8
6
3
0.789 0.0060
0.719 0.0044
0.473 0.0018
34
0.796 0.0056
Sam pling site
Iceland
E astern
A tlantic
K a tte g at
Ka
66.16
65.94
66.28
66.91
67.06
63.91
63.95
64.02
64.08
64.22
63.96
59.37
58.87
57.50
C entral N orth
Sea
Ns
56.13
O ff H augesund H a
A ll Locations
337
M a rc h 2004
M a rc h 2004
M a rc h 2004
O c to b e r 2003
C ru ise0
ICBTS
ICBTS
ICBTS
ICBTS
IBTS
IBTS
Nh h e
7If
0.0030
0.0047
0.0056
0.0057
0.0038
0.0058
0.0071
0.0071
a Sam pling size
b A A dults, I Im m a tu re
c D FO F a ll survey D e p artm en t o f Fisheries and O ceans, G overnm ent o f C anada, Fall survey, ICBTS IC elandic B ottom T raw l Survey
(M arine In stitu te o f Reykjavik, O ctober and M arch surveys), IB T S In ternational B ottom T raw l survey (International council for the
exploitation o f th e sea)
d N u m b er o f haplotypes
c H ap lo ty p e diversity (N ei 1984)
1 nucleo tid e diversity (N ei 1984)
following T em pleton and Sing’s rules (1993). These
rules a re based o n th ree criteria: th e frequency crite­
rion, in which haplotypes are expected to be connected
to frequent haplotypes rath er than to rare haplotypes;
th e topology criterion, in which haplotypes are ex­
p ected to be connected to interior rath er than tip
haplotypes; and th e geographical criterion, in which
new haplotypes are expected to rem ain in the same
location as th e ancestral haplotype. U sing the nested
design and th e softw are Geodis 2.4 (Posada et al.
2000), th e relationship betw een haplotype or clades
and th eir geographical distribution was assessed. The
program calculates the average clade distances (D c),
S pringer
which is a m easure of the geographical spread of a
clade; the nested clade distances (D n), which is a
m easure of how a clade is distributed in com parison to
o th er clades within th e same nested clade level; and the
interior-tip distances (I-T c and I-T N) which indicate
how w idespread younger clades (tips) are com pared to
their ancestors (interiors), relative to other clades
w ithin the same nesting clade. T he statistical signifi­
cance of the distance m easures was calculated by
com parison with a null distribution after 1,000 p e r­
m utations. Finally, th e biological m eaning of the out­
p u t was interpreted using th e latest T em pleton
inference key (Tem pleton 2004).
M ar Biol (2007) 151:1275-1286
Estim ates of past population expansion were m ade
using the m ism atch distribution of th e cytochrom e b
sequences (frequencies of pairwise differences betw een
haplotypes), T ajim a’s D and F u ’s estim ators and a
generalized skyline plot (coalescence approach). Con­
gruence am ong the th ree approaches provides stronger
inference for population growth or lack thereof. The
m ism atch analysis assum es th at population growth or
decline will reveal a genetic signature (i.e., a unimodal
distribution) different from th at observed with a con­
stant population size (Rogers and H arpending 1992).
T he observed m ism atch distribution was com pared to
an expected distribution under an expansion growth
m odel and param eters of the m odel (x = time of
expansion, 0O = population size prior to expansion,
0i = final population size) were estim ated from the
m ism atch distribution through the least-square proce­
d ure (Schneider and Excoffier 1999). Based on the
estim ated param eters x and the form ula x = 2¡i t, the
tim ing of expansion can be estim ated if the substitution
ra te is known. Substitution rates w ere estim ated at
betw een 0.008 and 0.005 p er M yr (Chevolot et al.
2006b). T ajim a’s D and F u ’s estim ators test for neu­
trality, b u t signature o f a population expansion is also
given by a significantly negative (Tajim a 1989; Fu
1997). T hese values w ere estim ated using A rlequin
v3.0 and significance was tested against 10,000 perm u­
tations.
H istorical dem ography was inferred using a coalescent approach through the generalized skyline plot
based on the phylogeny of the haplotytpes. T he m odel
of variable population size describes th e shape of the
genealogy depending on the dem ographic history of
the population. This approach is particularly effective
fo r non-resolved, star phylogenies and low sequence
variation (Strim m er and Pybus 2001), which is quite
often the case for intraspecific phylogenies. T he first
step in this analysis is to obtain phylogenetic trees from
all m tD N A sequences with branch length proportional
to time. T herefore, we first calculated the m ost likely
m odel of m olecular evolution to explain the data using
M odelTest v3.06 (Posada and Crandall 1998). A hier­
archical test of likelihood was perform ed under 56
m odels and using the likelihood ratio test (LR T), the
H K Y m odel with a gam m a ra te was selected
(/(A ) = 0.24; /(T ) = 0.27; /(C ) = 0.14; /(G ) = 0.35;
transition/transversion ratio = 18.15; gam m a shape =
0.015) (P < 0.0001). T hereafter, m axim um likelihood
trees were estim ated und er the H K Y m odel with
gam m a shape and w ith a m olecular clock assumption
using Paup v4.0bl0 (Swofford 1998). The software
G enie 3.0 (Pybus e t al. 2000) was used to obtain the
generalized skyline plot from the m axim um likelihood
1279
trees w ith a sm oothing algorithm. T he s param eter
governs th e sm oothing algorithm and reduces the noise
due to th e stochasticity of th e coalescence events. This
param eter was chosen using the maximize optimization
option.
Results
G enetic diversity analysis
A m ong th e 337 individuals, 34 haplotypes w ere de­
tected; differences am ong haplotypes w ere due to 26
polym orphic sites, of which 12 were informative. All
differences w ere due to substitutions; no indels were
found. T he overall haplotype and nucleotide diversities
w ere 0.796 and 0.009, respectively. T he highest haplo­
type diversity was found in Ox, Iceland {h = 0.916),
and the lowest in the C entral N orth Sea (h = 0.473). In
general, how ever, haplotype diversity was relatively
hom ogeneous across all o th er locations (Table 1).
Population differentiation
Because the global 0 (0.019) was significant
{P = 0.001), thorny skates w ere initially assumed to be
significantly differentiated across the sampling sites.
Pairw ise 0 com parisons showed th at m ost significant
pairwise comparisons w ere due to the tw o K attegat
locations (K a-I, K a-A ) (Table 2). Rem oval of K attegat
populations resulted in a non-significant global 0
(0.008; P - 0.1). N o significant genetic differentiation
was detected (global 0 = 0.0104, P = 0.12) am ong all of
the Icelandic populations.
G eographic distribution of haplotypes
T he statistical parsim ony netw ork among th e different
haplotypes (Fig. 2) revealed a star-like genealogy with
H21 (found in all sampling locations) as the most
central and th e m ost frequent (54.5%, T able 3; Fig. 2).
T h e second m ost common haplotype was H 8 (also
present in all locations) (Table 3; Fig. 2). Fourteen
haplotypes were exclusive to Iceland, three were un­
ique to N ew foundland (H20, H23, H28), one (H32) to
Norway, and three to the K attegat (H7, H10, H16).
T he nested clade analysis (Fig. 2) revealed 13 one-step
clades, 5 tw o-step clades and the overall netw ork is a
three-step clade. A significant association betw een a
clade and its geographic distribution was found in four
cases (Table 4). T he em erging picture is restricted gene
flow betw een K attegat, Iceland/N ew foundland (clade
1-3, H7); and the possibility of long-distance dispersal
4 6 Springer
1280
M ar Biol (2007) 151:1275-1286
Ha
Ho
io
o ro—
OOÖ
o
?
O iH
ÖO
pi
or isolation by distance betw een Iceland N ew found­
land and K attegat (clad e 1-7, H19). W ithout samples
from G reenland and th e Faroes, we cannot distinguish
betw een these tw o scenarios in th e last case. Contigu­
ous range expansion is the b est explanation for haplotype/clade distribution at th e tw o-step level (clade 2 5, H21) and for the to tal cladogram , which indicates
th at there has been h ig h connectivity among locations
for a long time.
H N H IO
W
OHH
OO O
Ö C^) O Ö
iÖ
«“nÖ H O n>—
n <—
'» r—
■
> t—
'
i f l H H ' O l f l H H N O i n O H H N
o ö o o
I’O
)vO
*1o—
1io N
Otnr
-O
~
feNC
0oO
0V
N
O
N
©
C
m
TV
C
Oo
O
s
o
s
T
—
<
o
o
o
o
o
o
®
©
®Oo o o o ÖÖo C5Oo
O03O
NN
ON
OT
—
1t-~s C
M
oo m
TION
C
N
C
M
C
M
o
p
Oo
S
Oo
o
o
o
o
o
Ö? ? ? i ? c ? ? Ö? o 1
io
I1—
OM
c■
o*TOo
—
1C
tf-N
O
S11C
M
M
-N
- ow
Moo N
1-*N
o1sC
o
O
o
o
o
o
o
o
Ö? ÖÖÖo ÖÖÖ? Ö 1
IO
ON O
Historical dem ography
T he star-like genealogy for th e haplotype netw ork is
consistent with a sudden exponential expansion of A .
radiata populations. T his is supported by th e unim odal
m ism atch distribution in Fig. 3 and the test of goodness
fit of a sudden expansion m odel th a t could not be re ­
jected (P - 0.12). T h e estim ated param eters of the
m odel w ere r = 1.414, 90 = 0.21 and 0! = 1,000, from
which the tim e of expansion for A . radiata was esti­
m ated to have started betw een 690,000 and
1,100,000 years ago and not during the L G M (substi­
tution rate = 0.005 an d 0.008 p er M yr, see M aterials
and m ethods). T ajim a’s D statistics were -1 .6 and the
F u ’s Fs statistics -27.3; both w ere significant (P = 0.026
and P < 0.001, respectively). T h e generalized skyline
plot (Fig. 4) is also consistent w ith past population
growth.
CO C
CO ■'vt' O
cO i—( —i
OHN
OO
OO
T-H o oi-H or—! oCM oO 'D
d d cp d
d
pI
?
t
s
cm
v©cofMONOoot'--ooN
Ox
H C O N n n o n n
o o o o o o o o o
d d d d d d c ^ d d
C M c o c O t'-C M c O jM O
Ska
C M O t- H t - < C M < - < 1 > v -<
o o o o o o o o
d d c j> d d c j) O c ji
N i o
co
Ol ' t
N
ON If) t~~ Os
oo c
omco
o oo
o o o o
Skj-I
Ey
VN
coMoo oh oo oo oo co
o
d d dd d
d
Skj-A
N
o
ot- i
?
Discussion
Population differentiation
N
—
<
C^- O
N—«C
^lI
t—
I t—
iO
1—
o oo o o
dcp^C f
Is
o o o o
d c jid ^
?
Bj-A
Hu
cM v n -3- c o
>-< o
CO T-l
o o
d d
-O
tí
tí
Bj-I
z
o
43
co <
at-<
>O
O
* d
¡3 V
Í3
>*
Nf
Locations
Table 2 Population
differentiation
among the sampling locations based on pairwise multi-locus 9 estimates (Weir and Cockerham
1984).
Ka-I
K a-A
o o o
ö d c^>
« <
Z mm
*£) Springer
'd
d
3 ~ÇN ‘ÇT
_w X
O (0 CÖ (0 m
ffi % % W LO O a S K Sd Ná ¡Z,
T he n e a r absence of genetic differentiation in A. rad­
iata over the N orth A tlantic does not conform to pre­
dictions based on life history characteristics of Rajidae.
A lthough a lack of pow er related to the small sample
size and the use of only one m olecular m arker might
explain this (W aples 1998), the finding of strong and
highly significant structure a t the ocean basic scale in a
parallel study using the same m arker for another R a ­
jidae species Raja clavata (with a 9 30 tim es higher)
(Chevolot et al. 2006b) suggests th at our results are not
an artifact. B oth species have sim ilar life history traits
but different depth ranges such that, the continental
shelf m argins are effective barriers.
Tagging studies (m ark/recapture) conducted for
R ajidae species have indicated traveling distances on
th e o rd er of hundreds of kilom eters, (Tem plem an
1984; W alker et al. 1997; H unter et al. 2005) as com ­
pared to traveling distances on th e order of thousands
of kilom eters for m any bony and cartilaginous fishes
M ar Biol (2007) 151:1275-1286
Fig. 2 Statistical parsim ony
netw ork (heavy lines) and
n ested clade design (light
lined boxes) for m tD N A
haplotypes o f A . radiata. T he
square indicates th e m ost
p robable ancestral haplotype.
O th e r haplotypes are
indicated by circles; and sm all
black dots denote
in term ed iate haplotypes not
p resen t in th e data set. E ach
line in th e netw orks betw een
haplotypes represents a single
m utational change. Internal
am biguities w ere resolved
follow ing T em p leto n and Sing
(1993), and th e less likely
paths betw een haplotypes are
show n by a dashed line. T he
n ested clade level is given in a
hierarchical m anner; 1-n for
1-step clade, 2-n fo r 2-step
clades. T he w hole cladogram
is a 3-step clade
1281
2-5
2-2
1-10
1-11
1-12
H2
O H IO
H26
H15
H33
H13
1-13
H18
H11
H9
2-3
1-14
2-1
1-7
1-8
--y
H17
H14
H30
H22
1-1
H29
H21
H31
H8
H12
H 32
H24
H34
H27
1-2
H19
1-4
H28
Iceland only
O Kattegat only
(S) N orw ay only
Newfoundland only
(M etcalfe and A rnold 1997; Law son and R ose 2000;
K ohler and T urner 2001; Sedberry and L oefer 2001;
Q ueiroz et al. 2005) supporting the hypothesis of re ­
gional p opulation structure in A . radiata. A closer look
within the R ajidae, however, reveals th at A . radiata
travels the farthest—betw een 1.4 and 3 tim es as far as
R. clavata (Tem plem an 1984; W alker e t al. 1997). U ntil
now, little attention has been paid to this difference
because m ism atches betw een tagging and genetic
studies (H o arau et al. 2004; C hevolot e t al. 2006a)
have been so great th at the variance rep o rted within
tagging studies was largely discounted. Tagging studies
typically encom pass a p eriod of a few years (max
4 years), less than a generation (first age at m aturity for
A . radiata 5-6 years) in R ajidae, w hereas genetic
studies integrate processes over m any generations.
Thus, our genetic result on A . radiata, suggests th at the
m igratory range is m uch greater than previously
acknowledged. T he large scale genetic hom ogeneity
m ay lie in the fact th at thorny skates are not restricted
to shallow shelf as they have been caught at depths
down to 1,000 m (Stehm ann and Biirkel 1994). For
exam ple, A . radiata is com m only caught across the
channel separating Iceland and G reenland continental
shelves a t approxim ately 66°N, w here depth is as low
as 600 m and it is as well regularly fished on th e Iceland
F aro e R idge w here depth is around 500 m (J. Palsson,
1-3
H3
H5
H7
2-4
1-6
H25
H20
1-5
H23
Pers. Comm.). This suggests th at its wide depth range
enables A . radiata to m igrate over large distances b e ­
tw een continental shelves using intercontinental ridges.
The strong differentiation observed betw een th e two
K attegat locations am ongst the rest is probably due to
differences in salinity and tem perature that occur in
th e transition from th e N orth Sea into the Baltic, as
significant restricted gene flow across th e transition
zone has been shown in several other groups of m arine
organisms. In the turbot {Scophthalmus m axim us), no
genetic differentiation was found betw een the A tlantic
and the N orth Sea, but highly significant differentiation
was found betw een the Baltic locations and N orth Sea/
A tlantic locations (Nielsen et al. 2004). Similar results
w ere obtained in the E uropean plaice {Pleuronectes
platesssa) with the m itochondrial m arker (H oarau
e t al. 2004) and in the cod (Gadus m orhua) (Nielsen
e t al. 2001). O th er groups of m arine organisms that
showed strong differentiation betw een the N orth Sea
and the Baltic Sea include th e seagrass Zostera marina
(V an O ppen et al. 1995; Reusch et al. 1999; Olsen
e t al. 2004), th e seaweeds Cladophora rupestris (Jo­
hansson et al. 2003) and Fucus serratus (Coyer et al.
2003). Salinity and tem perature play a m ajor role in
shaping population structure in the herring, Clupea
harengus (Jprgensen et aí. 2005). In all of these cases,
local adaptation to environm ental conditions m ay be
*2} Springer
1282
M ar B iol (2007) 151:1275-1286
Table 3 H ap lo ty p e distributions per sam pling sites.
H ap lo ty p e
(A ccession num ber)
H l (D Q 521996)
H 2 (D Q 521997)
H 3 (D Q 521998)
H 4 (D Q 521999)
H 5 (D Q 522000)
H 6 (D Q 522001
H 7 (D Q 522002)
H 8 (D Q 522003)
H 9 (D Q 522004)
H 10 (D Q 522005)
H l l (D Q 522006)
H I 2 (D Q 522007)
H13 (D Q 522008)
H 14 (D Q 522009)
H15 (D Q 522010)
H 16 (D Q 522011)
H 17 (D Q 522012)
H 18 (D Q 522013)
H I 9 (D Q 522014)
H 20 (D Q 522015)
H21 (D Q 522016)
H 22 (D Q 522017)
H23 (D Q 522018)
H 24 (D Q 522019)
H 25 (D Q 522020)
H 26 (D Q 522021)
H 27 (D Q 522022)
H 28 (D Q 522023)
H 29 (D Q 522024)
H 30 (D Q 522025)
H31 (D Q 522026)
H 32 (D Q 522027)
H 33 (D Q 522028)
H 34 (D Q 522029)
T o ta l
Nf
Bj-I
B j-A
Is
2
1
2
1
Hu
Skj-I
Skj-A
1
1
1
1
Ska
Ox
1
1
1
1
1
1
2
2
pi
1
2
1
3
Ho
1
Ha
K a-I
6
6
3
2
1
5
9
1
5
1
7
3
1
1
4
3
1
K a-A
Ns
1
6
2
1
4
4
3
1
1
1
3
1
1
1
1
2
3
1
4
1
2
1
1
1
1
1
1
3
1
1
1
1
1
5
1
1
1
10
13
1
12
1
2
1
1
1
1
4
11
6
1
1
1
7
11
12
4
1
8
1
9
7
7
11
8
22
11
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
2
1
1
2
30
32
3
1
22
17
1
27
25
18
im p o rtan t and may lead to stronger genetic differen­
tiation than large geographic distance or depth.
Phylogeographic patterns
F o r m uch of th e N o rth A tlantic biota, h ab itat loss
during th e L G M resulted in local extinctions and in a
generally southw ard range modifications into periglacial refugia. A s ice sheets retreated, populations ex­
panded generally following the leading edge
hypothesis (Ibrahim et al. 1996). In this m odel, theory
predicts th at th e leading edges will be less genetically
diverse as founders m ove away from the geneticallym ore-diverse form er refugia (and under th e further
assum ption of large réfugiai population sizes). The
degree to which this m odel holds is also expected to
differ betw een shallow benthic and pelagic organisms,
th e form er suffering the most. Rays and skates might
be predicted to occupy an interm ediate position as they
*£) S pringer
Ey
19
19
1
20
24
17
15
19
T otal
4
6
1
8
1
12
9
61
18
1
5
1
6
7
3
1
6
2
7
1
140
4
1
4
3
3
3
2
1
4
2
1
8
1
337
are free to move b u t are still dependent on bottom
substrate for feeding and egg laying.
Low haplotype diversity in the N orth Sea is consis­
te n t with the leading edge m odel since is has only ex­
isted for the past 8,000 years (Zagwijn 1992; D inter
2001). However, th e broad distribution of haplotypes
across the entire A tlantic, com bined with the star-like
haplotype netw ork and the contiguous range expansion
does not conform to the usual refugium m odel but to a
high gene flow species with long-term connectivity
betw een the eastern and western A tlantic (Avise et al.
1987; Avise 2000), as it has been found in som e teleosts
species (e.g. B rem er et al. 2005; Ely et al. 2005). Thus,
an East-W est A tlantic connection has existed for a
long time.
T h e large num ber of haplotypes found in Iceland is
m ost likely due to th e adm ixture betw een w estern and
eastern A tlantic populations as a consequence of its
central location ra th er than as a potential glacial
M ar Biol (2007) 151:1275-1286
1283
Table 4 Phylogeography inference from th e N ested clade analysis follow ing T em p leto n ’s key (2004)
C la d e
D o m in an t haplotype/geographical distribution
Inference
C lade 1-3
C lade 1-7
H 4/Iceland, N ew foundland and K attegat
H 19/Iceland, N ew foundland and K attegat
C lade 2-5
T otal Cladogram
H 21/all sites
H 21/ all sites
R estricted g e n e flow
T oo few sam ples to distinguish betw een
Long d istan ce dispersal and isolation by distance
Contiguous ra n g e expansion
Contiguous ra n g e expansion
O nly clade w ith significant association betw een haplotype/clades and geographical distribution are shown. See Fig. 2 for the nested
design
refugium . This is supported by the presence of haplo­
types found in N ew foundland and Iceland (H2, H19,
H 33), and a second se t of haplotypes found only in
Iceland and E urope (H 7, H 9, H U , H14, H17, H30).
A lthough our sam pling effort was proportionally larger
th an in the rest of th e N orth A tlantic, m ost of the
Icelandic locations have haplotype diversities within
the range of those found in Canada, N orw ay and the
K attegat.
0,35
0,30 0,25 -
u.
0,10 -
H istorical dem ography
0,05 0,00
0
4
2
8
6
10
Number of pairwise differences
Fig. 3 M ism atch distribution fo r A . radiata. T he line represents
th e expected distribution u n d e r a sudden expansion m odel
e=0.0009
2 0,01
0,002
0,004
500,000 y rs
0,006
0,008
0,010
1000,000 yrs
T im e ( s u b s titu tio n p e r s ite )
Fig. 4 G en eralized skyline p lo t of A . radiata. T he x-axis
represents th e tim e since th e expansion in substitutions p e r site
and in thousands o f years; th e y-axis th e estim ated effective
population size scaled to th e substitution rate. T he e-param eter
governing th e sm oothing algorithm was selected from the
A kaike Inform ation criterion (A IC ). T he last glacial m axim um
p eriod is re p re se n ted by the dashed area
Like R. clavata, A . radiata population expansion defi­
nitely predated th e end of the LG M and was estim ated
at 0.6-1.1 Myrs, which corresponds to th e B avelian and
C r omeri an complexes, both being successions of cold
and w arm periods (Zagwijn 1992). In the N orth
A tlantic, m arine species (so far investigated) seem to
follow a pre-L G M expansion m odel, e.g., th e common
goby (Gysels e t al. 2004), the A tlantic swordfish
(B rem er et al. 2005), th e A tlantic bluefin tu n a (Brem er
et al. 2005), A tlantic bigeye tuna (M artinez et al.
2006), th e red alga Palmeria palmata (Provan et al.
2005), th e brown alga Fucus serratus (H oarau et al.
subm itted), the bivalve M acom a balthica (Luttikhuizen
et al. 2003) and the estuarine fish, Ethm alosa fimbriata
(D urand et al. 2005); w here the date of expansion was
estim ated a t betw een 536,000 (for the com m on goby)
and 128,000 years (for the red alga, Palmeria palmata).
Thus, it is likely that highly m obile species and/or those
able to shift in th e subtidal fared better in the many
glacial-interglacial periods. F o r A. radiata, a general
southerly displacem ent of th e distribution probably
occurred; population sizes w ere probably little af­
fected.
T o conclude, although A. radiata and R. clavata
share sim ilar life-history traits, different phylogeo­
graphic and population genetic structure patterns were
found: no significant population differentiation for A.
radiata in th e N orth A tlantic; and strong population
differentiation for R. clavata in E u ro p ean waters
(Chevolot et al. 2006b). Clearly, life-history traits in
C ) Springer
1284
the R ajidae are poor predictors o f the population dif­
ferentiation.
Acknowledgments W e thank H e n k H eessen from th e R IV O
(IBTS survey, T h e N etherlands), D avid W . K ulka, T odd Inkpen
and Jo e Firth from th e D ep artm en t o f Fisheries and O ceans of
th e C anadian governm ent (D F O , Fall survey), and crew m em ­
bers from the different research vessels for their help in the
sampling; and J.A . C oyer and G. H o a rau for their useful com ­
m ents on previous versions of this m anuscript. T his research was
supported by N W O -P R IO R IT E IT program m a SU SU SE, P ro ­
jec t N r. 885-10-311.
References
A ddison JA , H a rt M W (2005) C olonization, dispersal, and
hybridization influence phylogeography of N orth A tlantic
sea urchins (Strongylocentrotus droebachiensis). E volution
59:532-543
A vise JC (2000) Phylogeography: the history and form ation of
species, 1st edn. H a rv ard U niversity Press, Cam bridge
A vise JC , A rnold J , B all R M , B erm ingham E , Lam b T , Neigel
JE , R eeb C A , S aunders N C (1987) Intraspecific phylogeog­
raphy: the m itochondrial D N A b ridge betw een population
genetics and system atics. A n n R ev E col Syst 18:489-522
B elkhir K, B orsa P , C hikhi L, R aufaste N, B onhom m e F (2004)
G enetix, logiciel sous windows T M p o u r la génétique des
populations. L ab o rato ire G énom e, Populations, Interactions,
CNRS U M R 5000, U niversité de M ontpellier II, M ontpellier
Bingham R G , H ulto n N R J, D ugm ore A J (2003) M odelling the
southern extent o f th e last Icelandic ice-sheet. J Q u at Sei
18:169-181
Birky CW , F u erst P, M aruyam a T (1989) O rganelle gene
diversity u n d e r m igration, m utation, and drift-equilibrium
expectations, a pproach to equilibrium , effects o f heteroplasm ic cells, and com parison to nuclear genes. G enetics
121:613-627
B rander K (1981) D isap p earan ce o f com m on skate Raia batis
from Irish Sea. N a tu re 290:48-49
B rem er JR A , V inas J, M ejuto J, E ly B , P ia C (2005) C om par­
ative phylogeography o f A tlantic bluefin tim a and swordfish:
the com bined effects of vicariance, secondary contact,
introgression, and p opulation expansion on the regional
phylogenies o f two highly m igratory pelagic fishes. Mol
Phylogenet E vol 36:169-187
C hevolot M, E llis JR , H o a rau G , R ijnsdorp A D , Stam W T,
O lsen JL (2006a) P o pulation stru ctu re o f thom back ray
(Raja clavata L .) in B ritish w aters. J Sea R es 56:305-316
C hevolot M, H o a rau G , R ijnsdorp A D , Stam W T, O lsen JL
(2006b) P hylogeography and pop u latio n structure o f th o m ­
back rays (Raja clavata L., R ajidae). M ol E col 15:3693-3705
C lem ent M, P osada D , C randall K (2000) TCS: a com puter
program to estim ate gene genealogies. M ol E col 9:16571659
Coyer JA , P eters A F , H o a ra u G , Stam W T, O lsen JL (2002)
H ybridization of the m arine seaw eeds, F ucus serratus and
Fucus evanescens (H etero k o n to p h y ta: Phaeophyceae) in a
100-year-old zone of secondary contact. Proc R Soc L ond B
Biol Sei 269:1829-1834
Coyer JA , P e ters A F , Stam W T, O lsen JL (2003) Post-ice age
recolonization and differentiation o f Fucus serratus L.
(Phaeophyceae; F ucaceae) populations in N o rth ern E urope.
M ol E col 12:1817-1829
“ê J Springer
M ar B iol (2007) 151:1275-1286
C unningham C, Collins T (1998) B eyond areas relationships:
E xtinction and re co lo n iz a tio n in m olecular m arine bioge­
ography. In: S c h ierw ater B, S treit B , W agener G, D e Salle R
(eds) M olecular e co lo g y and evolution: approaches and
applications. B irk h a u s e r V erlag, B asel, pp. 297-321
D in ter W P (2001) B io g eo g ra p h y o f th e O S P A R m aritim e area.
A synopsis and synthesis o f biogeographical patterns
described for the N o rth -E a st A tlantic. F ederal A gency of
N a tu re C o n serv atio n , B onn
D onnelly P, T avare S (1986) T h e ages o f alleles and a coalescent.
A d v A p p P robab 18:1-19
D ulvy N K , M etcalfe J D , G lanville J, Paw son M G , R eynolds JD
(2000)
Fishery sta b ility , Local extinctions, and shifts in
com m unity stru c tu re in Skates. C onserv Biol 14:283-293
D u ran d JD , T ine M , P anfili J , Thiaw O T , L ae R (2005) Im pact of
glaciations and geo g rap h ic distance on th e genetic structure
o f a tropical e stu a rin e fish, Ethm alosa fim briata (C lupeidae,
S. Bow dich, 1825). M o l Phylogenet E vol 36:277-287
E lphinstone M S, H in te n G N , A nderson M J, N ock CJ (2003) A n
inexpensive and high -th ro u g h p u t p rocedure to ex tract and
purify to tal genom ic D N A for po p u latio n studies. M ol Ecol
N o tes 3:317-320
Ely B, V inas J, B re m e r JR A , Black D , Lucas L, C ovello K,
L ab rie A V , T h elen E (2005) C onsequences o f th e historical
dem ography on th e global p opulation structure o f two
highly m igratory cosm opolitan m arin e fishes: th e yellowfin
tu n a ( Thunnus albacares) and th e skipjack tu n a (Katsuw­
onus pelam is). B M C E vol Biol 5:19
Frenzel B, Pécsi M, V e lic h k o A A (1992) A tlas of paleoclim ates
and p aleoenvironm ents of the N o rth ern hem isphere: L ate
P leistocene-H olocene. G eographic research institute, H u n ­
garian A cadem y o f Sciences, B udapest
Fu Y X (1997) Statistical tests o f neutrality of m utations against
po p u latio n grow th, hitchhiking a n d background selection.
G enetics 147:915-925
G ovindarajan A F , H alan y ch KK, C unningham CW (2005)
M itochondrial e v olution and phylogeography in th e hydrozoan O belia geniculata (Cnidaria). M ar Biol 146:213-222
Gysels E S, H ellem ans B , Pam poulie C, V olckaert F A M (2004)
Phylogeography o f th e com m on goby, Pom atoschistus m i­
crops, w ith p a rticu la r em phasis on th e colonization of the
M ed iterran ean a n d th e N orth Sea. M ol E col 13:403-417
H all T (1999) B ioedit: a user-friendly biological sequence
alignm ent and analysis program fo r windows 95/98/NT.
N ucleic A cids Sym p S er 41:95-98
H eessen H J L E (2004). D evelopm ent o f elasm obranchs assess­
m ent, D E L A SS . D G fish study co n tract 99/055. Final rep o rt
605 p
H eist E J (1999) A review o f population genetics in Sharks. A m
Fish Soc Symp 23:161-168
H ew itt G M (2000) T h e genetic legacy o f th e q u aternary ice ages.
N a tu re 405:907-913
H o a rau G , R ijnsdorp A D , V an der V eer H W , Stam W T, Olsen
J L (2002) Population structure o f plaice (Pleuronectes
platessa L .) in n o rth ern E urope: M icrosatellites revealed
large-scale spatial and tem poral hom ogeneity. M ol Ecol
11:1165-1176
H o a ra u G, P iquet A M T , V an der V e e r H W , R ijnsdorp A D ,
S tam W , O lsen J L (2004) Population structure o f plaice
(Pleuronectes platessa L .) in n orthern E urope: a com parison
o f resolving pow er betw een m icrosateflites and m itochon­
drial D N A data. J Sea R es 51:183-190
H u n te r E , B uckley A A , Stew art C, M etcalfe JD (2005) M igra­
to ry behavior o f the thom back ray, Raja clavata in the
S o u th ern N orth Sea. J M ar Biol A ssoc U K 85:1095-1105
M ar Biol (2007) 151:1275-1286
Ibrahim KM , Nichols R A , H ew itt G M (1996) Spatial p atterns of
genetic variation generated by different form s o f dispersal
during range expansion. H eredity 77:282-291
Jprgensen H B H , H ansen MM, B ekkevold D , R uzzante D E ,
L oeschcke V (2005) M arine landscapes and population
genetic structure o f herring (Clupea harengus L.) in the
Baltic Sea. M ol E col 14:3219-3234
Johansson G , Sosa P A , Snoeijs P (2003) G enetic variability and
level o f differentiation in N o rth Sea and B altic Sea
populations of th e green alga Cladophora rupestris. M ar
Biol 142:1019-1027
K o h ler N E , T u rn er P A (2001) Shark tagging: a review of
conventional m ethods and studies. E n v iro n Biol Fishes
60:191-223
Law son G L , R ose G A (2000) Seasonal distribution and m ove­
m ents o f coastal cod (Gadus m orhua L .) in Placentia Bay,
N ew foundland. Fish R es 49:61-75
Lescasse R. (1999) R echerche de variants génétiques dans le
gène KCN Q1 codant un canal potassique, dans le cadre
d ’une é tu d e épidém iologique: M ise au p o in t d e la SSCP
fluorescente su r séquenceur. P hD T hesis, D E A thesis,
U niversity of P ie rre and M arie C urie, Paris
L uttikhuizen PC, D re n t J, B ak er A J (2003) D isjunct distribution
o f highly diverged m itochondrial lineage clade and popula­
tion subdivision in a m arine bivalve w ith pelagic larval
dispersai. Mol E col 12:2215-2229
M artinez P , G onzalez E G , C astilho R, Z ard o y a R (2006)
G enetic diversity and historical dem ography o f A tlantic
bigeye tuna ( Thunnus obesus). M ol Phylogenet Evol 39:404416
M etcalfe JD , A rnold GP (1997) T racking fish w ith electronic
tags. N atu re 387:665-666
N ei M (1987) M olecular evolutionary genetics. C olum bia U n i­
versity Press, N ew Y ork
N ielsen E E , H ansen M M , Schm idt C, M eldrup D , G ro n k jaer P
(2001) Fisheries: p opulation o f origin o f A tlantic cod.
N atu re 413:272
N ielsen E E , Nielsen P H , M eldrup D , H an sen M M (2004)
G enetic population structure o f tu rb o t (Scophthalm us m a x­
im us L.) supports th e presence o f m ultiple hybrid zones for
m arine fishes in th e transition zone betw een th e Baltic Sea
and the N orth Sea. M ol E col 13:585-595
O lsen JL, Stam W T, C oyer JA , R eusch T B H , B illingham M,
B ostrom C, C alvert E , C hristie H , G ra n g er S, L a Lum iere
R , M ilchakova N , O udot-L e Secq M P, Procaccini G, Sanjabi
B, Serrao E , V eldsink J, W iddicom be S, W yllie-E cheverria
S (2004) N orth A tlantic phylogeography a n d large-scale
population differentiation o f the seagrass Zostera m arina L.
M ol E col 13:1923-1941
O rita M, Iw ahana H , K anazaw a H , H ayashi K , Sekiya T (1989)
D etection of polym orphism s o f hum an D N A by gel
electrophoresis as single stran d conform ation polym or­
phisms. Proc N atl A cad Sei U S A 86:2766-2770
P osada D, C randall K (1998) M O D E L T E S T : testing th e m odel
o f D N A substitution. B ioinform atics 14:817-818
P osada D , C randall K A , T em p leto n A R (2000) Geodis: a
program for th e cladistic nested analysis o f th e geographical
distribution of genetic haplotypes. M ol E col 9:487-488
P rovan J, W attier R A , Maggs C A (2005) Phylogeographic
analysis of the red seaw eed Palm aria palm ata reveals a
Pleistocene m arine glacial refugium in th e E nglish Channel.
M ol E col 14:793-803
Pybus O G , R am baut A , H arvey P H (2000) A n integrated
fram ew ork for th e inference o f viral p opulation history from
reconstructed genealogies. G enetics 155:1429-1437
1285
Q ueiroz N, L im a F P , M a ia A, R ibeiro P A , C orreia JP , Santos
A A (2005) M o v e m en t o f blue shark, Prionace glauca, in the
north-east A tlantic b a se d on m ark-recapture data. J M ar
Biol A ssoc U K 85:1107-1112
R eusch T B H , Stam W T, O lsen JL (1999) M icrosatellite loci in
eelgrass Zostera m arina reveal m arked polym orphism with­
in a n d am ong populations. M ol E col 8:317-321
R ogers A R , H arp en d in g H (1992) Population grow th m akes
waves in th e distribution of pairw ise genetic differences.
M ol B iol E v o l 9:552-569
R ozas J, R ozas R (1999) D n aS P version 3: an integrated program
for m o lecular po p u latio n genetics and m olecular evolution
analysis. B ioinform atics 15:174-175
R undgren M, Ingolfsson O (1999) Plant survival in Iceland
during periods o f glaciation? J B iogeogr 26:387-396
Schneider S, E xcoffier L (1999) E stim ation o f past dem ographic
param eters from th e distribution o f pairw ise differences
when the m u ta tio n ra te s very am ong sites: application to
hum an m itochondrial D N A . G enetics 152:1079-1089
Sedberry G R , L o efe r JK (2001) Satellite telem etry tracking of
swordfish, X ip h ia s gladius off th e eastern U nited States. M ar
Biol 139:355-360
Stehm ann M (1995). M a tu ratio n guide for Elasm obranchs.
R e p o rt o f th e IC E S study group on E lasm obranch fishes.
CM 1995/G:3 85 p
Stehm ann M, B ürkel D L (1994) R ajidae. In: W hitehead PJP,
B auchot M L, H u re a u J-C, N ielsen J, and T ortonese E (eds)
Fishes of th e N o rth -eastern A tlantic and M editerranean
V ol. I. pp. 163-196, U N E S C O , Paris
Strim m er K, Pybus O G (2001) E xploring the dem ographic
history o f D N A sequences using th e generalized skyline
plot. M ol B iol E vol 18: 2298-2305
Sulikowski JA , K n eeb o n e J, E lzey S, Ju re k J, D anley PD,
H ow ell W H , T sang P C W (2005) T h e reproductive cycle of
th e thorny sk a te (A m blyraja radiata) in the w estern G ulf of
M aine. Fish B ull 103:536-543
Sunnucks P , W ilson A , B eheregaray L , Z enger K, F ren ch J,
T aylor A (2000) SSCP is not so difficult: th e application and
utility of single-stranded conform ation polym orphism in
evolutionary biology and m olecular ecology. M ol E col
9:1699-1710
Swofford D (1998) PA U P*4blO . Phylogenetic analysis using
parsim ony (*and o th e r m ethods). Sinauer A ssociates, Sun­
derland, M assachusetts
T ajim a F (1989) Statistical m eth o d for testing th e neutral
m utation hypothesis by D N A polym orphism . G enetics
123:585-595
T em plem an W (1984) M igrations o f thorny skate, Raja radiata,
tagged in N ew foundland area. J N o rth A tl Fish Sei 5:55-63
T em pleton A R (2004) Statistical phylogeography: m ethods of
evaluating and m inim izing inference errors. M ol Ecol
13:789-809
T em pleton A R , Sing CF (1993) A Cladistic analysis o f pheno­
typic associations with haplotypes inferred from restriction
endonuclease m apping. 4. N ested analyses with cladogram
uncertainty and recom bination. G enetics 134:659-669
T em pleton A R , R o u tm an E , Phillips C A (1995) Separating
population stru ctu re from p opulation history: a cladistic
analysis of th e geographical distribution of m itochondrial
D N A haplotypes in the tiger salam ander, A m bystom a
tigrinum. G enetics 140:767-782
V an O ppen M JH , O lsen JL , Stam W T (1995) G enetic variation
w ithin and am ong N orth A tlantic and B altic populations of
th e b enthic alga Phycodrys rubens (R hodophyta). E u r J
Phycol 30:251-260
4Î) Springer
1286
W alker P A . (1998) D ynam ics of N orth Sea ray populations. PhD
T hesis, U niversity of A m sterdam , T h e N etherlands
W alker P A , H ow lett G, M illner R (1997) D istribution, m ove­
m en t and stock structure of th ree ray species in th e N orth
Sea and E astern English C hannel. IC E S J M ar Sei 54:797808
W aples R S (1998) S eparating th e w heat from th e chaff: p atterns
o f genetic differentiation in high gene flow species. J H e red
89:438-450
W ares JP , C unningham CW (2001) Phylogeography and histor­
ical ecology o f th e N orth A tlantic Intertidal. E volution
55:2455-2469
<£) S p rin g er
M ar B iol (2007) 151:1275-1286
W eir BS, Cockerham C C (1984) E stim atin g F-statistics for the
analysis of p o p u la tio n structure. E v o lu tio n 38:1358-1370
W right S (1969) E v o lu tio n a n d th e genetics o f population. V ol 2.
T he theory of g e n e frequencies. C hicago U niversity Press,
Chicago
Zagw ijn W H (1992) T h e beginning o f th e lee A ge in E u ro p e and
its m ajor subdivisions. Q u a t Sei R e v 11:583-591