phylogeography and historical ecology of the north atlantic intertidal

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors nonprofit publishers academic institutions researchlibraries and research funders in the common goal of maximizing access to critical research

PHYLOGEOGRAPHY AND HISTORICAL ECOLOGY OF THE NORTHATLANTIC INTERTIDALAuthor(s) John P Wares and Clifford W CunninghamSource Evolution 55(12)2455-2469 2001Published By The Society for the Study of EvolutionDOI httpdxdoiorg1015540014-3820(2001)055[2455PAHEOT]20CO2URL httpwwwbiooneorgdoifull1015540014-3820282001290555B24553APAHEOT5D20CO3B2

BioOne (wwwbiooneorg) is a nonprofit online aggregation of core research in the biological ecological andenvironmental sciences BioOne provides a sustainable online platform for over 170 journals and books publishedby nonprofit societies associations museums institutions and presses

Your use of this PDF the BioOne Web site and all posted and associated content indicates your acceptance ofBioOnersquos Terms of Use available at wwwbiooneorgpageterms_of_use

Usage of BioOne content is strictly limited to personal educational and non-commercial use Commercial inquiriesor rights and permissions requests should be directed to the individual publisher as copyright holder

2455

q 2001 The Society for the Study of Evolution All rights reserved

Evolution 55(12) 2001 pp 2455ndash2469

PHYLOGEOGRAPHY AND HISTORICAL ECOLOGY OF THENORTH ATLANTIC INTERTIDAL

JOHN P WARES12 AND CLIFFORD W CUNNINGHAM1

1Duke University Department of Biology Box 90325Durham North Carolina 27708-0325

Abstract Recent glaciation covered the full extent of rocky intertidal habitat along the coasts of New England andthe Canadian Maritimes To test whether this glaciation in fact caused wholesale extinction of obligate rocky intertidalinvertebrates and thus required a recolonization from Europe we compared American and European populations usingallelic diversity and techniques adapted from coalescent theory Mitochondrial DNA sequences were collected fromamphi-Atlantic populations of three cold-temperate obligate rocky intertidal species (a barnacle Semibalanus bal-anoides and two gastropods Nucella lapillus and Littorina obtusata) and three cold-temperate habitat generalist species(a seastar Asterias rubens a mussel Mytilus edulis and an isopod Idotea balthica) For many of these species wewere able to estimate the lineage-specific mutation rate based on trans-Arctic divergences between Pacific and Atlantictaxa These data indicate that some obligate rocky intertidal taxa have colonized New England from European pop-ulations However the patterns of persistence in North America indicate that other life-history traits including mech-anisms of dispersal may be more important for surviving dramatic environmental and climatic change

Key words Coalescent glaciation North Atlantic phylogeography range expansion

Received January 17 2001 Accepted August 29 2001

A primary goal of historical ecology is to determine howcommunities and biological interactions develop (Ricklefsand Schluter 1993 Stone et al 1996) Interactions amongspecies may vary geographically based on the ecological con-text and such variation may arise in very recent species as-sociations (Vermeij 1992 Travis 1996) For example theisopod Idotea balthica is a more dominant herbivore and isfound in a broader range of algal and seagrass habitats (J PWares pers obs) in New England populations than in Eu-ropean populations In European populations this speciestends to be ecologically marginalized through competitionwith congeneric grazers (Franke and Janke 1998) What eco-logical variables play a determinant role in producing thisvariation in community composition and interactions Onlyby knowing both the historical and ecological contexts ofthese interactions can we decide (Endler 1982 Brooks andMcLennan 1991 Vermeij 1992) and studies of comparativephylogeography represent a solid approach to this problem

In North Atlantic intertidal marine communities the lastglacial maximum (around 20000 years ago Barash et al1992 Holder et al 1999) is thought to have been especiallydifficult for obligate rocky intertidal species on the NorthAmerican coast Many such species may have been forcedfrom their typical habitat by Pleistocene glaciation becausethere is currently little hard substrate beyond the southern-most extent of the glaciers at Long Island Sound (Knott andHoskins 1968 Ingolfsson 1992 Riggs et al 1996) If gla-ciation caused localized extinction in New England rockyintertidal species then much of the extant rocky intertidalcommunity may have recently immigrated from Europewhere hard substrata were more available and Pleistoceneclimatic changes were not as severe (Vermeij 1991 van Op-pen et al 1995) Distributional evidence suggests this is alikely scenario because the list of obligate rocky intertidal

2 Present address Department of Biology University of NewMexico Castetter Hall Albuquerque New Mexico 87131-1091E-mail jpwaresunmedu

species in North America is a subset of the same group ofspecies found in Iceland which in turn represents a subsetof these species found in northern Europe (Ingolfsson 1992)Although many marine species may have been able to shiftor contract their geographic range into southern refugia alongthe North American coast species requiring hard substratamay not have been able to find suitable habitat

Analysis of data from multiple species with varied life-history traits (including larval dispersal trophic level andecological traits such as substrate requirements) allows us tofocus on species or groups of species that did not survivethe most recent glacial maximum in North America Popu-lations of these species in New England should be recogniz-able based on two basic genetic patterns First recentlyfounded populations are expected to represent a subset of thegenetic diversity in the source population This leads to theprediction that recently founded populations should have asignificantly lower genetic diversity and have a high fre-quency of alleles that are identical to or descended fromalleles in the founding population (Hewitt 1996 2000 Aus-terlitz et al 1997 Johnson et al 2000 Zink et al 2000)Second recently founded populations should show the ge-netic signature of a rapidly expanding population (eg Neeet al 1995 Kuhner et al 1998) in part because they areunlikely to have reached an equilibrium between allelic di-versity (mutation) and demographic changes (genetic drift)

In this paper we use genetic markers to describe the post-glacial reassembly of the intertidal marine community in NewEngland and the Canadian Maritimes We will focus on thehypothesis that substratum requirements have played a majorrole in North Atlantic historical biogeography Because thereare recognizable patterns that we expect in recently foundedpopulations we test these predictions using both standardpopulation genetic measures as well as recently developedcoalescent methods Coalescent methods (for review seeHudson 1990) are expected to more accurately estimate thesepatterns than methods using pairwise sequence comparisons

2456 J P WARES AND C W CUNNINGHAM

TABLE 1 Species collected for this study with geographic range and collection sites indicated Dispersal type for each species is summarizedin the first column and described further in the Discussion Sister groups used to ensure proper identification and for estimate of m are alsolisted Collection sites include Beaufort NC (NC) Beavertail State Park RI (RI) Woods Hole MA (WH) Darling Marine Center DamariscottaME (M) Antigonish Nova Scotia (NS) St Johnrsquos Newfoundland (W) Reykjavik Iceland (Ic) Trondheim Norway (No) Galway Ireland(Ir) and Roscoff France (Fr) Species considered to be obligate rocky intertidal taxa are marked with an asterisk Outgroup species used formutation rate calibration are marked with the symbolDagger

Species (n)dispersal type Geographic range Collection sites Congeners identified

Data collected(Genbank accession)

Asterias rubens (48) Echino-dermata Asteroideabroad planktonic dispersal

N Europe Iceland N Ameri-ca to Long Island Soundsubmerged to mid-Atlantic

M NS W Ic NoIr Fr

DaggerA amurensis (Pacific) A for-besii (Atlantic N Americanendemic)

627 bpAF240022ndash240081

Idotea balthica (34) Arthropo-da Isopoda

low dispersal brooder

Mediterranean W Europe NAmerica from Cape Hatter-as north

NC IR M NSW Ic No Ir Fr

I metallica I emarginata Igranulosa (Atlantic)

516 bpAF241889ndash241935

Mytilus edulis (91) MolluscaBivalviabroad planktonic dispersal

Mediterranean W Europe NAmerica from Cap Hatterasnorth

NC RI M NS WIc No Ir Fr

M galloprovincialis (Atlanticincluded) DaggerM trossulus (Pa-cific Atlantic populations)

463 bpAF241936ndash142035

Littorina obtusata (44) Mol-lusca Gastropodalow dispersal crawl-away

N Europe Iceland N Ameri-ca to Long Island Sound

RI M NS W IcNo Ir Fr

L fabalis (Atlantic) DaggerL saxa-tilis (Atlantic)

537 bpAF242067ndash242117

Semibalanus balanoides (69)Arthropoda Cirripediabroad planktonic dispersal

N Europe Iceland N Ameri-ca to Long Island SoundAlaska


S cariosus (Pacific) DaggerS balan-oides (Alaskan populations)

615 bpAF242660ndash242728

Nucella lapillus (61) Mollus-ca Gastropodalow dispersal crawl-away



DaggerN freyceneti (Pacific) 589 bpAF242118ndash242178

(eg Slatkin and Hudson 1991 Rogers and Harpending1992) which do not make full use of the genealogical struc-ture of the data (Felsenstein 1992) Furthermore because wewere able to estimate lineage-specific mutation rates for thespecies in this study we use these coalescent methods tocompare the population sizes and ages of North Americanand European populations Although there are uncertaintiesin using such estimates they nevertheless allow us to estab-lish a quantitative as well as qualitative test of the hypoth-esis that New England rocky intertidal populations werefounded since the most recent glaciation

Here we apply these methods to a series of empirical da-tasets that allow us to compare the idiosyncratic history ofeach species with expectations based on the coalescent mod-el We have studied mitochondrial DNA (mtDNA) sequencesfrom amphi-Atlantic populations of three cold-temperate ob-ligate rocky intertidal species (a barnacle Semibalanus bal-anoides and two gastropods Nucella lapillus and Littorinaobtusata) and three cold-temperate habitat generalist species(a seastar Asterias rubens an isopod I balthica and theblue mussel Mytilus edulis) These data indicate that someobligate rocky intertidal taxa have colonized New Englandfrom European populations however the patterns of persis-tence in North America indicate that other life-history traitsincluding mechanisms of dispersal may be more importantfor surviving dramatic environmental and climatic change

MATERIALS AND METHODS

Specimen Collection and Preservation

Specimens were collected from intertidal sites listed inTable 1 and tissues were then placed in 95 ethanol or aDMSO-based buffer (Seutin et al 1991) immediately Withtwo exceptions species were identified based on morpho-

logical characters listed in Hayward and Ryland (1995) andGosner (1978)

For M edulis the phenomenon of doubly uniparental mi-tochondrial inheritance (reviewed in Quesada et al 1996)required that female lineages be identified by comparison toidentified male and female lineages in Hoeh et al (1996)only females were incorporated into our analyses Mytilusedulis were distinguished from the European endemic speciesM galloprovincialis using both the nuclear Glu-59 polymer-ase chain reaction (PCR) marker (Rawson et al 1996) andcomparisons with sequence data from the mitochondrial cy-tochrome oxidase III locus (Quesada et al 1995) Individualsidentified as M galloprovincialis were not excluded if theirmitochondrial type was that of M edulis

Littorina obtusata is difficult to distinguish morphologi-cally from its congener L fabalis We identified a clade ofL fabalis individuals with our cytochrome oxidase I (COI)data and confirmed their identity by comparing sequence datafrom the cytochrome oxidase b gene to published L fabalissequences (Reid et al 1996) these individuals were removedfrom further analysis

DNA Extraction and Amplification

Tissues were phenol-extracted and amplified with the PCRprotocol listed in Wares (2001a) and the mitochondrial COIgene primers LCO1490HCO2198 (Folmer et al 1994) Forsome taxa species-specific primer pairs were developed frominitial sequence data (available from authors) All individualswere sequenced from both directions using PCR primersSequences were obtained in both directions and edited usingSequencher 30 (Genecodes Corp Ann Arbor MI) No indelswere found but missing or ambiguous end regions weretrimmed so that all individuals in a species had the samesequence length

2457PHYLOGEOGRAPHY AND NORTH ATLANTIC GLACIATION

Phylogenetic Analysis

Parsimony networks were developed for each species usingthe methods of Templeton et al (1992) The network for eachspecies was compared with the topology of the maximum-likelihood (ML) phylogeny to ensure concordance the best-fit ML model (Goldman 1993 Cunningham et al 1998) foreach dataset was determined using ModelTest (Posada andCrandall 1998) Searches were performed with stepwise ad-dition (simple addition sequence) and TBR branch swappingwith zero-length branches collapsed

Isolation by Distance and Haplotype Diversity

Pairwise estimates of Nm (where N is the effective pop-ulation size and m is the rate of migration) between sampledlocations for each species were calculated based on an islandmodel using Slatkinrsquos linearized FST-values (Slatkin 19911995) in the program Arlequin 2001 (Schneider et al 1997)along with AMOVA tests (Excoffier et al 1992) for eachspecies Geographic distances (km) between locations werecalculated using the great circle distance between each siteA Mantel test (1000 permutations using Mantel 20 availablevia wwwsciquteduauNRSmantelhtm) compared the cor-relation between log(Nm) and log(km) distance matrices tothe null hypothesis that they are not linearly related (Slatkin1993 Hellberg 1994) Additionally haplotype diversity (HNei 1987 eq 84) and its sampling variance were calculatedfor each species andor continental population using Arlequinvers 2001

Estimation of Ancestral Population Parameters

The full geographic range of each species was divided intoNorth American and EuropeanIcelandic regions Estimatesof Q (5 2Nm where m is the mutation rate for mtDNA) weremade for each region as well as the North Atlantic as a wholeThe ML estimate of the parameters Q and g (the exponentialgrowth parameter in units of m21) were made using Fluctuate(Kuhner et al 1998) Seeds for all analyses were generatedrandomly The appropriate transitiontransversion ratio (es-timated by maximum likelihood in PAUP 40 Swofford1998) was input for each dataset Analyses were repeated fivetimes per species andor population to ensure stability ofparameter estimates the mean values are reported Final anal-ysis of each dataset employed 10 short Monte Carlo chainsof 4000 steps each and five long chains of length 20000with a sampling increment of 20 Fluctuate generated a ran-dom topology for initial searching These estimates of ex-ponential growth are used to generate the historical size ofeach population

2(gm)tN 5 Qe t (1)

where Nt is the effective population size at any time t in thepast (Kuhner et al 1998)

RESULTS


Haplotype networks representing the complete datasets forall six species are shown in Figure 1 Because outgroup root-

ing is not reliable for intraspecific genealogies (Castelloe andTempleton 1994) the most likely root haplotypes are indi-cated on each network For datasets that are compatible withthe infinite-sites model (Watterson 1975) the likelihood ofthe genealogy given each possible root haplotype was cal-culated using Genetree (Griffiths and Tavare 1994) The datafor M edulis violate this model so the heuristic method ofCastelloe and Templeton (1994) was used This method alsoobtained the same haplotype for each of the other species asthe ML methods described above


There were no patterns of isolation by distance in theseNorth Atlantic species that were more than marginally sig-nificant (J P Wares unpubl data) In each case the signalof population structure could be traced to the European pop-ulations trans-Atlantic isolation by distance does not seemto be an important mechanism Haplotype diversities (H) ofAmerican and European populations are shown in Table 2In all cases but M edulis North American populationsshowed significantly lower haplotype diversity than Europeanpopulations Complete AMOVA results indicating how ge-netic variation is partitioned in each species are reported inAppendix 2 Only in two species (A rubens and L obtusata)was there any indication of subdivision between North Amer-ican and European populations In Littorina there is a distinctclade in the European population (Fig 1) that contributes tothis statistical partitioning whereas in Asterias all Americanhaplotypes are shared with Europe suggesting that the AMO-VA result is artifactual (see genealogical description below)This result is likely an artifact of pairwise FST analysis gen-erated by the significant difference in genetic diversity be-tween these two populations (see Whitlock and McCauley1999)

Lineage-Specific Estimates of Mutation Rate

Fossil evidence indicates that many of the lineages in thisstudy invaded the North Atlantic from the Pacific during thetrans-Arctic interchange 35 million years ago (Worley andFranz 1983 Vermeij 1991 Collins et al 1996 Reid et al1996) Although earlier estimates have been made for theopening of the Bering Strait (Marincovich and Gladenkov1999) the first fossil evidence of large-scale invasion fromthe Pacific was at 35 million years ago (Vermeij 1991) Be-cause this invasion took place just before the onset of North-ern Hemisphere cooling and glaciation (Berggren and Hol-lister 1974) the cold-temperate species in this study wereable to invade the North Atlantic but have had little oppor-tunity for recent gene flow across the Arctic For each species(except I balthica) we identified the Pacific sister taxon andused AtlanticPacific sequence divergence as the basis for ourlineage-specific mutation rates with an estimated date of di-vergence of 35 million years ago

To account for polymorphism in ancestral species lineagesthe ML estimate of the internal branch length separating thesister taxapopulations from the Pacific and Atlantic was usedto estimate the appropriate amount of divergence per site (seeEdwards and Beerli 2000) This measure represents the netnucleotide divergence d (Nei and Li 1979) and allows a


FIG 1 Parsimony networks for each species in this study Dark circles represent alleles found in Europe and white circles representalleles found in North America Each branch represents a single nucleotide substitution with missing intermediates shown as smallempty circles Bold lines indicate firstsecond position substitutions Shared alleles are shown as half-shaded ovals Sample locationsare designated as in Table 1 The maximum-likelihood root presented (arrows) is always at least an order of magnitude greater likelihoodthan other possible root alleles in Mytilus the maximum-likelihood root could not be calculated using an infinite-sites model The Castelloeand Templeton (1994) method indicates that the root probability of the shared allele combined with the alleles marked lsquolsquo9Frrsquorsquo lsquolsquoEurrsquorsquoand lsquolsquoNAmrsquorsquo is 0662 Additional information on allele sampling and composition of shared alleles for each species is available inAppendix 1


TABLE 2 Comparisons of haplotype diversity in North American and European populations Haplotype diversity (H) and sampling variancewere calculated for each species in Arlequin 2001 (Schneider et al 1997) using equation (84) in Nei (1987) In each species except for Mytilusedulis haplotype diversity is significantly higher (P 005) in European populations than in North American populations

Species North American (H) Europe (H)

Asterias rubensSemibalanus balanoidesNorth American clades separated by maximum-likelihood rootMytilus edulisLittorina obtusataNucella lapillusIdotea balthica

0597 6 00630771 6 0052

0275 6 0015 0662 6 01160893 6 00320186 6 01100145 6 00890257 6 0142

0893 6 00380936 6 0033

0923 6 00250933 6 00250434 6 01010942 6 0038

TABLE 3 Estimated mutation rates for species in this study (substi-tutions per site per generation) Estimates are based on the internalbranch length separating populations of ingroup and outgroup taxa(see Materials and Methods) using the best-fit model (F84) for thirdposition characters only Estimates of third-position mutation rateswere also made using datasets for species separated by the Isthmus ofPanama including the crab genus Sesarma (S rhizophorae vs tran-sisthmian sister group Schubart et al 1998) the shrimp Alpheus (Achacei vs A penultimum Knowlton and Weigt 1998) and the barnacleEuraphia (Wares 2001a)

Species Mutation rate (SE)

Asterias rubensMytilus edulisLittorina obtusataNucella lapillusSemibalanus balanoides

484 3 1028 (877 3 1029)951 3 1028 (196 3 1028)249 3 1028 (13 3 1028)443 3 1028 (107 3 1028)276 3 1028 (632 3 1029)

Idotea balthicaSesarma spp (Panama)Alpheus spp (Panama)Euraphia spp (Panama)

36 3 1028 (see text)21 3 1028

19 3 1028

38 3 1028

calculation of the mutation rate as m 5 (12)d(35 3 106

years) using the trans-Arctic divergence estimate discussedabove These ML estimates were calculated using the F84model (separate rates for transitions and transversions)which is the only model applied in Fluctuate Only the Medulis data rejected the assumption of a molecular clock (x2

distributed with [n 2 2] degrees of freedom P 005 Gold-man 1993)

Estimates of m were based on divergence between Atlanticpopulations and the Pacific sister groups For two speciesdirect estimates of AtlanticPacific divergence were not avail-able For L obtusata we were not able to obtain specimensof the Pacific outgroup L sitkana (Reid et al 1996) Ourestimate of m in Littorina was based on comparisons betweenL obtusata L saxatilis and L sitkana for cytochrome b (Reidet al 1996) and our COI divergence between L obtusata andL saxatilis A direct estimate for I balthica is not availablebecause the trans-Arctic history and interspecific phylogenyof this genus are both unknown Subsequent analysis of Ibalthica used the mean estimate of m from other species inthis study a rate very similar to other crustacean rate esti-mates for this fragment of COI as shown in Table 3

We chose to restrict our estimates of mutation rate to thirdpositions for two reasons First substitutions at the thirdposition are much less likely to cause amino-acid substitu-tions and are more likely to obey the assumption of neutralityrequired by our coalescent methods even within-species

comparisons often contain more nonneutral polymorphismsthan would be expected under the neutral model (Hasegawaet al 1998) Second to include all three codon positions anestimate of among-site rate variation is necessary Becauseour Pacific and Atlantic divergences are very recent we be-lieve that the appropriate parameter for the gamma distri-bution cannot be accurately estimated based on the low se-quence variation in these samples Using the discrete gammadistribution the mean substitution rate across site classesmust be one (Yang 1994) yet in these datasets one class ofsubstitutions is always dramatically more frequent than oth-ers meaning its rate is proportional to the number of classeschosen for the discrete gamma In fact Pacific and Atlanticdivergences differed by up to an order of magnitude de-pending on how many discrete rate categories were estimatedfor the gamma distribution (C Cunningham pers obs) Mu-tation rates based on third positions are presented in Table 3

Estimates of Population Growth

The parameters Q and g were jointly estimated for thirdposition datasets for all six species using Fluctuate (Table 4Kuhner et al 1998) under two conditions an unconstrainedexponential growth parameter and an assumption of constantN (g 5 0) Allowing Fluctuate to estimate Q and g jointlyproduced estimates of N that were two to three times largerthan when Q was estimated with no growth These jointlyestimated parameters were used to generate plots of thechange in relative size of N over time For example the timeat which the population was half its current size is given asln(05)mg Table 4 presents confidence intervals indicatingthe size of each population 20000 years ago relative to itscurrent size The plots of relative N through time are givenin Figure 2 For the barnacle S balanoides the genealogicalstructure and root likelihood suggested that the analysisshould be performed assuming multiple colonization events(see below) If population size is assumed constant eachspecies except for M edulis has significantly lower estimatesof Q and H in North American populations than in Europeanpopulations (Tables 2 4) The approximate age of each pop-ulation is presented in the Results and Discussion as the ageat which N drops below 1 of the current estimated size

Genealogical Descriptions for Each Species

Asterias rubens

The North American population of this seastar containedno unique alleles relative to the European population which


TA

BL

E4

Est

imat

esof

Q(t

heco

mpo

und

para

met

erre

pres

enti

ngth

eef

fect

ive

popu

lati

onsi

zean

dsu

bsti

tuti

onra

te)

and

g(t

heex

pone

ntia

lgr

owth

para

met

er)

for

spec

ies

inth

isst

udy

usin

gF

luct

uate

(Kuh

ner

etal

19

98)

Eac

hsp

ecie

sis

divi

ded

into

Nor

thA

mer

ican

and

Eur

opea

npo

pula

tion

sto

calc

ulat

eth

era

nge

ofpr

ojec

ted

ages

for

each

regi

on

The

Flu

ctua

tees

tim

ates

(the

com

posi

teof

five

repl

icat

ean

alys

es)

indi

cate

the

max

imum

-lik

elih

ood

esti

mat

eof

the

join

tpr

obab

ilit

yof

Qan

dg

The

esti

mat

esof

Ndu

ring

the

mos

tre

cent

glac

ial

max

imum

(20

000

year

sag

o)re

lati

veto

pres

ent-

day

esti

mat

esof

Nar

eba

sed

ong

and

m

The

stan

dard

devi

atio

nis

pres

ente

daf

ter

each

esti

mat

or

Est

imat

edpa

ram

eter

sfo

rId

otea

()

base

don

resu

lts

usin

gal

lco

don

posi

tion

sas

the

Nor

thA

mer

ican

popu

lati

onis

inva

rian

tat

thir

dpo

siti

onH

owev

er

the

max

imum

-lik

elih

ood

estim

ate

ofg

ism

axim

alus

ing

eith

eral

lco

don

posi

tion

sor

ada

tase

tsi

mul

ated

usin

gth

esa

me

mut

atio

nal

para

met

ers

asth

eId

otea

thir

dpo

siti

onda

ta

and

this

esti

mat

eis

conc

orda

ntw

ith

high

esti

mat

esof

gin

othe

rsp

ecie

sw

ith

very

low

Nor

thA

mer

ican

alle

lic

dive

rsit

y(L

itto

rina

N

ucel

la)

Spe

cies

Pop

ulat

ion

Q(n

ogr

owth

)Q

(joi

ntly

esti

mat

ed)

gT

ime

to1

rela

tive

NR

elat

ive

N20

000

year

sag

o

Ast

eria

sru

bens

Idot

eaba

lthi

ca

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

000

429

026

63

000

398

005

004

000

566

000

2

041

906

030

1

034

046

032

80

1883

60

012

875

36

437

4

962

16

314

2

100

00

278

36

186

104

800

(59

500

ndash25

390

0)95

400

(61

200

ndash17

160

0)13

400

482

700

017

ndash0

99

022

ndash0

72

25

310

25ndash

001

10

81ndash

085

Myt

ilus

edul

is

Lit

tori

naob

tusa

ta

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

002

965

003

264

000

248

003

137

012

596

004

5

031

806

009

53

001

346

001

15

011

706

002

28

109

46

494

356

66

553

100

006

(19

46)

493

66

393

184

310

6

(11

2ndash

389

310

6 )5

663

105

(43

2ndash

774

310

5 )18

500

(11

047

ndash38

389

)3

743

105

(24

7ndash

680

310

5 )

091

ndash0

99

081

ndash0

89

68

310

23ndash

68

310

23

075

ndash0

82

Sem

ibal

anus

bala

noid

esN

orth

Am

eric

a

Sou

thcl

ade

Nor

thcl

ade

na

000

309

002

341

003

256

000

23

715

316

392

3

044

566

059

94

308

65

9

2326

36

460

2

100

006

(11

38)

514

310

6

(35

4ndash

814

310

6 )68

000

(46

685

ndash10

840

1)15

800

(11

679

ndash22

826

)

098

ndash0

99

015

ndash0

44

29

310

23ndash

29

310

23

Nuc

ella

lapi

llus

Eur

ope

Nor

thA

mer

ica

Eur

ope

006

347

000

434

000

534

018

266

001

73

009

666

007

49

053

796

053

42

176

06

263

100

006

(11

61)

4391

46

396

7

899

310

5

(64

23

105ndash

135

310

6 )10

400

(747

9ndash

154

96)

236

00(1

740

0ndash

343

00)

087

ndash0

93

14

310

23ndash

14

310

23

001

ndash0

04


FIG 2 Patterns of growth in each species (with 95 confidence intervals) as reconstructed from estimates of the parameter g estimatedjointly with Q using Fluctuate (Kuhner et al 1998) The vertical axis in each graph represents ancestral Ne relative to current populationsize horizontal axis is time measured in thousands of years The most recent glacial maximum is indicated at 20000 years on eachchart Populations that appear to have less than 1 of current N at 20000 years (see Table 4) are likely to be recent colonists Thereconstruction of Idotea is based on the same growth parameter being obtained for both analysis of full dataset and third position onlywith random substitutions It is also the same growth parameter obtained for Nucella and Littorina with nearly identical North Americanallelic patterns The pattern shown for Asterias is illustrative of the effect of ancestral polymorphism on these reconstructions as it issimilar to the pattern obtained for European populations (data not shown) despite significantly lower haplotype diversity in North America(see Discussion)

is consistent with a recent founding event from Europe (Fig1) This is supported by a significantly lower haplotype di-versity in North America (Table 2) However parameter es-timates (Q and g) are very similar in North American andEuropean populations and the North American populationappears to date back beyond the recent glacial maximum(Table 4 Fig 2) Such a contradiction indicates that thisanalysis was confounded by ancestral polymorphism Whena population is founded by more than one haplotype ancestral

polymorphism from the founding population will inflate theestimated size and age of the newly founded population Thisissue is further described in the Discussion

Semibalanus balanoides

With very large diversity in Europe and two shared hap-lotypes with North America the genealogical pattern in Sbalanoides appears consistent with multiple colonization


events from Europe to North America (Fig 1) When allNorth American individuals are included our Fluctuate anal-ysis suggests an unrealistic time of origin for North Americanpopulations an age that actually predates the trans-Arcticinterchange (Table 4) As with A rubens this result can beeasily explained by ancestral polymorphism because the twomajor clades in North America (Fig 1) are distantly relatedto one another Each clade shares a common haplotype withEuropean populations and taken individually each clade issignificantly less diverse than the European population (Table2) This hypothesis of multiple colonization events is sup-ported by the separation of the two North American cladesby the ML root in Europe (Fig 1) Separate analysis of eachNorth American clade indicates that whereas one of theseevents may be quite recent (time to most recent commonancestor [TMRCA] 5 33 3 104 years population less than1 current size less than 20000 years ago) the other cladeappears to predate the last glacial maximum (Table 4) ahypothesis also suggested by Brown et al (2001)

Mytilus edulis

The mussel M edulis appears to have long histories onboth coasts (Table 4) with a tremendous amount of endemicdiversity present in each region The only shared allele be-tween North America and Europe was also identified as theroot haplotype note that the presence of high allelic diversityin North America cannot be easily explained by a very largerecent founding population because there is only a singleshared haplotype Inspection of the network in Figure 1 showsseveral unique clades to each continent including two NorthAmerican lineages that are distantly related to an isolatedEuropean clade and the shared North AmericanndashEuropeanhaplotype This pattern of missing intermediate haplotypesmay indicate incomplete lineage sorting or gene flow fromNorth America to Europe When these genetically isolatedlineages are removed our conclusion of a long history forM edulis in North America is not affected similarly exclu-sion of the haplotypes collected from M galloprovincialisindividuals (see Materials and Methods) did not affect ourconclusion of a long history in Europe

Littorina obtusata and Nucella lapillus

Littorina obtusata and N lapillus share remarkably similarpatterns in the North Atlantic (Fig 1) Both have a high-frequency haplotype in North America that is shared withEurope and each species has two unique North Americanhaplotypes Both have significantly lower haplotype diversityin North America than in Europe and both have populationages that are almost identical these results are consistentwith postglacial colonization from Europe For both speciestranslation of the ML estimates of g indicate that the NorthAmerican populations of these species were less than 1 ofthe current size during the most recent glacial maximum (Ta-ble 4 Fig 2) The apparent migration event from NorthAmerica to Europe by L obtusata (Fig 1) seems to be theresult of a homoplastic amino-acid substitution that unites aNorth American and European haplotype

Idotea balthica

The isopod I balthica has significantly lower diversity inNorth America than in Europe (Tables 2 4) Unlike all otherspecies in this study there are no shared haplotypes betweenNorth America and Europe Moreover Genetree identifiesthe common North American haplotype as the root of thenetwork and therefore the oldest allele in the North AtlanticBecause all seven European lineages are descended from theroot haplotype it is parsimonious to conclude that this hap-lotype was once present in the European population Ourfailure to find it in our European sample can be explainedby its loss due to lineage sorting or simply by our limitedsampling effort The very large diversity of the Europeanpopulation confirms a long history in Europe We concludethat this pattern is consistent with a recent colonization ofNorth America from Europe The alternative hypothesismdashthat the root allele originated in North Americamdashwould re-quire either seven independent invasions of Europe or sevenunique mutation events following a single invasion followedby extinction of the invading haplotype(s) These hypotheseshave been statistically rejected (J P Wares unpubl data)

Although I balthica has three haplotypes in North Amer-ica the two singleton haplotypes are characterized by firstsecond-position substitutions and are therefore not present inour third position datasets This means that at third positionsthe North American population is monotypic making an es-timate of age from third positions by Fluctuate impossibleDating the age of the North American population of I bal-thica is also complicated because it is the only species forwhich we lack lineage-specific estimates of mutation rateThat being said the allelic diversity of the North Americanpopulation is comparable to other species whose ages appearto date since the last glacial maximum (L obtusata and Nlapillus Table 4 Fig 2) and analysis of either the full datasetor a third position dataset with two randomly introducedsubstitutions generates a maximal estimate of populationgrowth and is reflected in the curve drawn in Figure 2

DISCUSSION

Glaciation and North Atlantic Populations

Our central focus has been to test the hypothesis that glacialevents had a greater impact on the rocky intertidal of NorthAmerica than in Europe and species requiring rocky substratawould be more likely to go extinct in North American pop-ulations This hypothesis makes two specific predictionsFirst all species that currently have an amphi-Atlantic dis-tribution will show a genetic signature of a long populationhistory in Europe Second some or all amphi-Atlantic specieswill show a genetic signature of postglacial range expansionfrom Europe to North America By distinguishing betweenspecies with North American populations that did or did notsurvive the last glacial maximum we can begin to understandthe organismal characteristics necessary to survive this ex-treme climatic event

Of these two predictions it is easiest to show that all Eu-ropean populations of amphi-Atlantic species survived thelast glacial maximum An older population is more likely tobe at an equilibrium level of genetic diversity and is char-


acterized by high allelic diversity with many unique haplo-types All of our European populations fit this prediction andin each case except for N lapillus the estimated age of theEuropean population appears to be much older than the lastglacial maximum (Table 4) In the case of Nucella high hap-lotype diversity (Table 2) and the fact that French populationsare distinguished from other European populations by a fixedsubstitution (Fig 1) substantiate the claim for the old age ofEuropean populations Additionally there are high levels ofboth allozyme and karyotypic polymorphism in Europe (Dayet al 1993 Collins et al 1996) This evidence of geographicsubdivision in European N lapillus populations is evidenceof an older population than is estimated using coalescentmethods

The prediction that some or all amphi-Atlantic species willhave recently colonized North America from Europe is moredifficult to test definitively but makes the following threepredictions (1) haplotype diversity will be significantly low-er in North American populations (2) the estimated age(based on the time to the most recent common ancestor) ofNorth American populations will postdate the last glacialmaximum and (3) all North American haplotypes will eitherbe shared with Europe or be descended from European hap-lotypes that have participated in range expansion

Of our six species five show evidence of colonization fromEurope including the barnacle S balanoides the gastropodsN lapillus and L obtusata the seastar A rubens and theisopod I balthica (see Results) Of these all but S balanoidesshow patterns consistent with postglacial colonization fromEurope (Fig 2 Table 4) One of two American clades of Sbalanoides appears to have survived the last glacial maxi-mum a result supported by an independent study of thisspecies (Brown et al 2001) Although the actual times sincecolonization of some of these species such as Asterias areconfounded by ancestral polymorphism or cannot be explic-itly determined using the coalescent methods we employed(ie Idotea) each dataset is consistent with postglacial rangeexpansion from Europe This conclusion follows the datashowing consistently lower allelic diversity in North Americaand the signal of population growth obtained from coalescentanalysis of each speciesrsquo gene tree

Several methods of inference were used in this study be-cause it is difficult to accurately test hypotheses about theage of recently founded populations due to ancestral poly-morphism Ancestral polymorphism is known to confoundestimates of population divergence time in many typical phy-logeographic studies (Nielsen et al 1998 Edwards and Beerli2000 Hewitt 2000) Estimating the time to the most recentcommon ancestor in a population will be biased when thefounding population itself was genetically diverse as in stud-ies such as this one where populations from North Americaand Europe are not divergent but are largely composed ofthe same common alleles We have taken ancestral poly-morphism into account by considering how alternative col-onization histories could affect our estimates In S balano-ides it seems apparent that there are two independent col-onization events to be analyzed separately However datingthe invasion of species such as A rubens is still complicatedbecause it is unclear how much diversity in North Americais due to multiple events or to founding population diversity

We realize that an alternative hypothesis exists that is con-sistent with some but not all of our observations If NorthAmerican populations did not go extinct but existed in aglacial refugium of restricted size we could expect signifi-cantly lower North American haplotype diversity Our esti-mates of recent population age from our Fluctuate analysesdo not rule out this hypothesis because Fluctuate was onlyused to generate a growth curve that suggests the time to themost recent common ancestor of the alleles in a population(the minimum population age assuming no effects due toancestral polymorphism)

Our best chance of distinguishing the refugium bottleneckand colonization hypothesis comes from our repeated obser-vation that North American haplotypes are either shared withor descend from European haplotypes (Fig 1) The repeatedobservation of descent from European haplotypes is unlikelyunder the refugium bottleneck hypothesis and is specificallypredicted by the colonization hypothesis Given the lengthof time that hypothetical North American refugia and Eu-ropean refugia would have been separated due to unfavorableclimatic conditions during glacial maxima simulations in-dicate that the likelihood of observing shared alleles betweenthese populations today is vanishingly small if gene flow wasnot maintained (Johnson et al 2000 J P Wares unpublms) Simply put separate refugia in North America and Eu-rope would tend to generate greater divergence among allelesfrom each population

Even within our data there is a good example of the dif-ference expected between the North American refugium andrecolonization hypotheses All I balthica individuals col-lected from Iceland form a clade with low haplotype diver-sity and this population shares no alleles with other Atlanticpopulations The Icelandic clade has three fixed substitutionsrelative to all other North Atlantic haplotypes Applying aNei and Li (1979) correction for the polymorphism extant inthe ancestral population the divergence between Iceland andthe rest of Europe is greater than 200000 years ago (con-servatively based on the most rapid estimate of m Table 3)This is consistent with the hypothesis that southwestern Ice-land was unglaciated during the last glacial maximum (seeHolder et al 1999) and that I balthica survived in this re-fugium Other species including the ocean quahog Arcticaislandica also show evidence of high endemic diversity inIceland (Dahlgren et al 2000) supporting a refugium hy-pothesis over predictions of recent gene flow

Of the species that do not appear to have survived the lastglacial maximum both N lapillus and L obtusata have beenfound as fossil remains on the North American coast priorto Pleistocene glaciation (Vermeij 1991) and archaeologicalevidence places these species again in New England prior tohuman colonization from Europe (Jenkins et al 1997) Aliteral reading of the fossil record in these cases could bemisleading because our data strongly suggest that they havenot been continually present in North America and yet theyhave not been introduced by humans either Although thepossibility of historical introduction by humans should al-ways be considered (Carlton and Geller 1993) the amountof unique genetic diversity in North American populationsof each of these species (see Hewitt 2000 Johnson et al2000 J P Wares D Goldwater and C W Cunningham


unpubl ms) and the fact that they are easily recognizedmacroinvertebrates suggests human-mediated introduction isunlikely

Persistence and Range Expansion Associations withLife-History Characteristics

What characteristics allowed species to survive the lastglacial maximum in North America We have presentedstrong evidence that a population of the obligate rocky in-tertidal barnacle S balanoides survived the most recent gla-cial maximum (see also Brown et al 2001) and other obligaterocky intertidal species in this study were apparently dis-tributed through New England long before those rocky shoreswere defined by glacial scouring (Pratt and Schlee 1969Vermeij 1991) It may be more appropriate to assume thatthere has long been suitable substrata for a variety of speciesin this region and glacial maxima simply increased theamount of rocky shore available north of Cape Cod Someobligate rocky intertidal species were able to find suitablehabitat during glacial maxima whereas others were notTherefore substrate requirements do not adequately explainthe patterns of persistence and range expansion seen in thesespecies

Persistence and colonization may require a mechanism ofdispersal that allows species to find suitable habitat quicklyduring environmental change Modes of dispersal are prob-ably not indicative of colonization ability (Johannesson1988) L obtusata and N lapillus two species that appearto have recently colonized North America have crawl-awaylarvae or very short planktonic dispersal (Crothers 1985 Reidet al 1996) suggesting that lack of a pelagic phase does notprevent trans-Atlantic colonization However dispersalmodes may explain the persistence of taxa in North Americathrough glacial maxima The two species with the strongestsignatures of preglacial persistence in North America Medulis and S balanoides each have long-dispersing plank-totrophic larvae (Gosling 1992 Holm and Bourget 1994)

To the list of glacial survivors in North America we canadd several other temperate species Although A rubens hasrecently colonized North America from Europe its sister spe-cies A forbesi is endemic to North America has plankto-trophic larvae (Clark and Downey 1992) and clearly survivedglacial maxima (Worley and Franz 1983 Wares 2001c) Ourdata show that whereas most haplotypes of I balthica sharea recent ancestor with European colonists two endemic NorthAmerican cryptic species have also been identified (Wares2001b) Idotea broods its offspring but is frequently foundon algal rafts that play a large role in its dispersal (Lockeand Corey 1989 Ingolfsson 1995) suggesting again that dis-persal may allow persistence through climatic changes Sev-eral other North American endemic species with broadly dis-persing larvae are reciprocally monophyletic with respect tosister species in Europe (eg the hermit crab Pagurus aca-dianus Cunningham et al 1992 C W Cunningham unpubldata and the lobster Homarus americanus Kit and Kornfield1998) suggesting long-term persistence in North America

These results suggest that persistence through glacial max-ima in North America may be facilitated by a capability forincreased dispersal whether by pelagic larvae or by other

mechanisms (ie algal rafting in isopods) This might beexplained by opportunistic colonization of ephemeral habitatsduring glaciation The hypothesis that high dispersal abilitymay help species resist extinction was proposed by Jablonskiand Lutz (1983) and Valentine and Jablonski (1986) basedon paleontological findings These fossil data illustrate thatplanktonic species tend to survive longer in the fossil recordand survive major extinction events with a higher frequencythan expected This life-history character may be crucial tounderstanding the patterns of diversity found in regions thatwere previously altered by glaciation or other climatic andenvironmental changes However it is also of great interestto community ecologists to recognize the differences amongspecies in terms of community assembly and colonizationdynamics Although planktonic larval dispersal may play arole in the long-term survival of species populations it isalso clear that other ecological and physiological dynamicsplay a role in this system

Methodological Considerations

Inference of population histories has become a popular andcontroversial enterprise In this paper we focus on estimatingrates of population growth using coalescent methods (Fluc-tuate Kuhner et al 1998) Here we focus on three aspectsof our approach that deserve attention our reliance onmtDNA our estimation of taxon-specific substitution ratesand the idiosyncrasies of coalescent methods for estimatingancestral population parameters

Corroboration from additional loci

It is well known that estimates of population historiesbased on a single locus will have high variances (Kuhner etal 1998 Wollenberg and Avise 1998 Fu and Li 1999) Al-though mtDNA is especially useful for inferring populationhistory with its advantages of a high mutation rate lack ofrecombination and small effective population size (Avise etal 1987) we must take care to recognize its shortcomingsIn this study we are mainly concerned with identifying spe-cies that survived the last glacial maximum in North AmericaHow might our mtDNA data be misleading in this regardIf mtDNA shows a signature of long population history onboth coasts this is unlikely to be contradicted by the additionof more loci This is confirmed for M edulis which has asecond locus available in its paternally inherited mtDNA Infact the male mtDNA shows even stronger evidence than thefemale mtDNA for long histories on both coasts Based onlimited sampling Quesada et al (1998) found reciprocalmonophyly between Europe and North America in the malemtDNA a result that is confirmed by additional sampling inour laboratory (C Riginos pers comm)

However our inference of recent immigration from Europefor several species can only be regarded as tentative withoutadditional loci What we interpret as a recent colonizationfrom Europe might simply reflect a recent introgression ofEuropean haplotypes into a long-standing North Americanpopulation If this were the case nuclear loci should showevidence of long divergence between Europe and NorthAmerica We tested this hypothesis with preliminary studiesusing sequence data from the ribosomal internal transcribed


spacer (ITS) region for the seastar A rubens (Wares 2001c)the barnacle S balanoides and the gastropod N lapillus (JP Wares unpubl data) For Asterias we sampled the Pacificoutgroup and both Atlantic species (A amurensis A forbesiand A rubens) Based on the Pacific and Atlantic divergencesthe ITS showed a mutation rate that was comparable to theCOI third position rate (m 5 36 3 1028) but found nosubstitutions at all between five North American and fiveEuropean A rubens individuals This is entirely consistentwith the recent colonization event from Europe inferred fromthe mtDNA data and inconsistent with a hypothesis of long-term residence on both coasts Very limited ITS analyses inthe other two species also indicated no divergence betweenEuropean and North American populations

Estimating mutation rates

As with any estimation of mutation rate our estimates mustbe considered provisional A potential source of error is ouruse of a date of 35 million years ago for the divergencebetween North Pacific and North Atlantic cold-temperatepopulations (see Materials and Methods) There have beenreports of the Bering Strait opening as early as 5 millionyears ago (Collins et al 1996 Marincovich and Gladenkov1999) although the major fossil evidence of invasion did notappear until 35 million years ago (Vermeij 1991) Using theolder date would increase our population age estimates byabout 40 which does not dramatically change our inter-pretations of postglacial histories

It is encouraging that our estimates of mtDNA mutationrate are similar to those in other taxa Our estimates for COIthird positions range from m 5 25ndash48 3 1028 (Table 2)excluding Mytilus which is known to have an elevated mu-tation rate (Hoeh et al 1996) These are within the rangenoted for COI third position data in other taxa estimated fromeither side of the Isthmus of Panama using our methodology(Table 3) Finally our estimate of mutation rate in Littorinausing the date of the Bering Strait opening (35 million yearsago m 5 25 3 1028) is close to the estimate obtained if weuse a fossil date (m 5 19 3 1028) based on fossil-dataestimates of divergence between L obtusata and L saxatilis(20 million years ago Reid et al 1996)

Coalescent methods and coalescent problems

Our inferences are also clearly affected by methods ofestimating Q and the exponential growth parameter g Whenan estimate of the growth parameter g is incorporated esti-mates of Q are much higher than under a model of constantpopulation size Thus although the use of an exponentialgrowth model may or may not be biologically accurate (seeKuhner et al 1998) it is again conservative in that our es-timates of historical population size assuming growth arelarger than when we do not

However this highlights potentially serious biases thatarise during joint estimation of parameters from genealogicaldata (Wakeley 2000) There appears to be a strong bias inthe estimation of growth parameters that may cause over-estimation of population size and age this bias may be aninherent property of estimating growth from genealogicaldata (Kuhner et al 1998) Thus strong covariation among

parameters estimated from genealogical data makes definitiveestimates of historical demographics elusive (Markovtsovaet al 2000)

The strongest confirmation of the methods we advocate istheir success when applied to other empirical systems Hell-berg et al (2001) propose that the gastropod Acanthinucellaspirata expanded its range north of Point Conception in Cal-ifornia since the last glacial maximum As with our two gas-tropods with crawling larvae (N lapillus and L obtusata)newly colonized northern populations are predominantlycomposed of a common haplotype shared with the south withseveral unique haplotypes descended from the putativefounding haplotype When we applied our approach usingFluctuate to A spirata third positions using our mutation ratecalculated for N lapillus the population growth curve strong-ly suggested that northern populations were founded in thelast 20000 years (CW Cunningham unpubl data) In gen-eral it appears that estimates of population age are affectedmore by the problem of ancestral polymorphism (ie ourAsterias data) than the problem of coestimation of parameterssuch as Q and g

Finally it should be noted that although nested clade anal-ysis (reviewed in Templeton 1998) is potentially useful insuch studies it can only infer range expansion when theranges of younger allelic lineages are more broadly distrib-uted than older lineages In our datasets in which range ex-pansion is too recent to be associated with much geneticdiversity there is not enough allelic or geographic resolutionto make a statistical inference using this method For ourtaxa the putatively invading haplotypes are the most deeplynested haplotype in the European (putative source) popula-tion (Fig 1) This is to be expected because deeply nestedancestral haplotypes are often the most common (Castelloeand Templeton 1994) therefore they have a higher proba-bility of participating in long-distance dispersal events

Conclusions Historical Biogeography and its Role inComparative Ecology

Reconstruction of historical change in a complex intertidalcommunity is perhaps inseparable from the idiosyncrasiesintroduced by the population genetics and life-history char-acteristics of the component species (Ricklefs and Schluter1993 Cunningham and Collins 1998) In many ways it isthis variation among species in their response to climaticchange and community disruption that is most interesting forthe study of historical ecology and community reassemblyDespite some broad similarities among groups of species inthis study there are also patterns of community history andcomposition that cannot be fully explained in terms of thetraits discussed here

For example the two species of Asterias in North Americaare distinguished not only by historical patterns but also byphysiological and geographic range differences (Worley andFranz 1983) and there are hypotheses remaining to be testedabout the historical associations among endemic North Amer-ican lineages of Idotea and their ecological milieu (Wares2001b) Although there are apparent relationships betweenmodes of larval dispersal and genetic patterns in the NorthAtlantic comparative work in two polychaete species (Nean-


thes virens and Hediste diversicolor) also points to adaptationto estuarine environments as another mechanism that couldisolate populations and prevent trans-Atlantic migrationevents (S Breton P Blier F Dufresne and G Desrosiersunpubl ms)

Nevertheless all of these species have experienced the samehistory of climatic change and this common cause shouldlead to some common patterns In the North Atlantic Pleis-tocene glaciation not only scoured the rocky intertidal of NewEngland but also caused sea level to drop up to 200 m thisprobably exposed suitable habitat for organisms displaced byglaciers and accompanying climatic change (Emery and Gar-rison 1967 Cronin 1988 Riggs et al 1996) although thetrue shoreline and its composition is as yet unknown forglacial periods (Pielou 1991) Here we show albeit with onlya handful of species that mechanisms of dispersal in inter-tidal organisms may be an important factor in the survivalof North American populations because these proposed off-shore reefs and mounts may have been widely scattered andonly temporarily available

Some important patterns are described with these data thatdo not directly test the proposed hypotheses of this study butthat illustrate the utility of comparative phylogeography tostudies of the genetics and ecology of North Atlantic inver-tebrates The growing body of literature on the populationgenetics of species such as M edulis and N lapillus indicatethat there are many elements of mitochondrial heredity andpatterns of gene flow that may only be resolved by a betterunderstanding of the underlying life-history traits an under-standing gained through comparisons with other species suchas in this study Additional ecological experimentation with-in the context of these phylogeographic patterns may be ableto discern the historical influences on the diversity of thiscommunity It is likely to be difficult to tease apart all theinfluences as studies show that predicting even simple com-munity responses to climatic change are quite complex (Bert-ness et al 1999 Sullivan et al 2000)

ACKNOWLEDGMENTS

Many thanks go to D S Goldwater C Torres D GarbaryD M Rand A Ingolfsson R Osmundsson B OrsquoConnorS Zaklan and G Manchenko for assistance in the field andin the laboratory M J Hickerson C Riginos S PalumbiD M Rand T F Turner M E Hellberg T H Oakley KZigler C Henzler and two anonymous reviewers also pro-vided valuable feedback and criticism on the initial appli-cation of these methods and this manuscript This researchwas funded by the National Science Foundation (DEB-96808267 to CWC) and a National Science Foundation Dis-sertation Improvement Grant (NSF-DEB-99-72707) and Sig-ma Xi Grant-in-Aid of Research to JPW

LITERATURE CITED

Austerlitz F B Jung-Muller B Godelle and P-H Gouyon 1997Evolution of coalescence times genetic diversity and structureduring colonization Theor Popul Biol 51148ndash164

Avise J C J Arnold R M Ball Jr E Bermingham T LambJ E Neigel C A Reeb and N C Saunders 1987 Intraspecificphylogeography the mitochondrial DNA bridge between pop-

ulation genetics and systematics Annu Rev Ecol Syst 18489ndash522

Barash M S and 61 additional authors 1992 Atlas of paleocli-mates and paleoenvironments of the Northern Hemisphere BFrenzel M Pecsi and A A Velichko eds Geographical Re-search Institute Hungarian Academy of Science Budapest

Berggren W A and C D Hollister 1974 Paleogeography pa-leobiogeography and the history of circulation in the AtlanticOcean Society of Economic Paleontologists and Mineralogistsspecial publication 20126ndash186

Bertness M D G H Leonard J M Levine and J F Bruno1999 Climate-driven interactions among rocky intertidal or-ganisms caught between a rock and a hot place Oecologia 120446ndash450

Brooks D R and D A McLennan 1991 Phylogeny ecologyand behavior a research program in comparative biology Univof Chicago Press Chicago IL

Brown A F L M Kann and D M Rand 2001 Gene flow versuslocal adaptation in the northern acorn barnacle Semibalanusbalanoides insights from mtDNA control regions polymor-phisms Evolution 551972ndash1979

Carlton J T and J B Geller 1993 Ecological roulette the globaltransport of nonindigenous marine organisms Science 26178ndash82

Castelloe J and A R Templeton 1994 Root probabilities forintraspecific gene trees under neutral coalescent theory MolPhyl Evol 3102ndash113

Clark A M and M E Downey 1992 Seastars of the AtlanticNatural History Museum Publications Vol 3 Chapman andHall London

Collins T M K Frazer A R Palmer G J Vermeij and W MBrown 1996 Evolutionary history of Northern Hemisphere Nu-cella (Gastropoda Muricidae) molecular morphological eco-logical and paleontological evidence Evolution 502287ndash2304

Cronin T M 1988 Evolution of marine climates of the US At-lantic coast during the past four million years Proc R SocLond B 318661ndash678

Crothers J H 1985 Dog-whelks an introduction to the biologyof Nucella lapillus (L) Field Stud 6291ndash360

Cunningham C W and T M Collins 1998 Beyond area rela-tionships extinction and recolonization in molecular marine bio-geography Pp 297ndash321 in B Schierwater B Streit G Wagnerand R DeSalle eds Molecular ecology and evolution ap-proaches and applications Birkhauser Verlag Basel Switzer-land

Cunningham C W N W Blackstone and L W Buss 1992Evolution of king crabs from hermit crab ancestors Nature 355539ndash542

Cunningham C W H Zhu and D M Hillis 1998 Best-fit max-imum-likelihood models for phylogenetic inference empiricaltests with known phylogenies Evolution 52978ndash987

Dahlgren T G J R Weinberg and K M Halanych 2000 Phy-logeography of the ocean quahog (Arctica islandica) influencesof paleoclimate on genetic diversity and species range MarBiol 137487ndash495

Day A J H P Leinas and M Anstersrud 1993 Allozyme dif-ferentiation of populations of the dogwhelk Nucella lapillus (L)the relative effects of geographic distance and variation in chro-mosome number Biol J Linn Soc 51257ndash277

Edwards S V and P Beerli 2000 Perspective gene divergencepopulation divergence and the variance in coalescence time inphylogeographic studies Evolution 541839ndash1854

Emery K O and L E Garrison 1967 Sea levels 7000 to 20000years ago Science 157684ndash687

Endler J A 1982 Problems in distinguishing historical from eco-logical factors in biogeography Am Zool 22441ndash452

Excoffier L P E Smouse and J Quattro 1992 Analysis of mo-lecular variance inferred from metric distances among DNA hap-lotypes application to human mitochondrial DNA restrictiondata Genetics 131479ndash491

Felsenstein J 1992 Estimating effective population size from sam-ples of sequences inefficiency of pairwise and segregating sitesas compared to phylogenetic estimates Genet Res 59139ndash147


Folmer O M Black W Hoeh R Lutz and R Vrijenhoek 1994DNA primers for amplification of mitochondrial cytochrome coxidase subunit I from diverse metazoan invertebrates MolMar Biol Biotech 3294ndash299

Franke H-D and M Janke 1998 Mechanisms and consequencesof intra- and interspecific interference competition in Idotea bal-thica (Pallas) and Idotea emarginata (Fabricius) (Crustacea Is-opoda) a laboratory study of possible proximate causes of hab-itat segregation J Exp Mar Biol Ecol 2271ndash21

Fu Y and W-H Li 1999 Coalescing into the 21st century anoverview and prospects of coalescent theory Theor Popul Biol561ndash10

Goldman N 1993 Statistical tests of models of DNA substitutionJ Mol Evol 36182ndash198

Gosling E M 1992 Genetics of Mytilus Pp 1ndash20 in E Goslinged The mussel Mytilus ecology physiology genetics and cul-ture Elsevier Press New York

Gosner K L 1978 A field guide to the Atlantic seashore from theBay of Fundy to Cape Hatteras Houghton Mifflin Boston MA

Griffiths R C and S Tavare 1994 Sampling theory for neutralalleles in a varying environment Philos Trans R Soc LondB 344403ndash410

Hasegawa M Y Cao and Z Yang 1998 Preponderance of slight-ly deleterious polymorphism in mitochondrial DNA nonsynon-ymoussynonymous rate ratio is much higher within species thanbetween species Mol Biol Evol 151499ndash1505

Hayward P J and J S Ryland 1995 Handbook of the marinefauna of North-West Europe Oxford Univ Press Oxford UK

Hellberg M E 1994 Relationships between inferred levels of geneflow and geographic distance in a philopatric coral Balano-phyllia elegans Evolution 481829ndash1854

Hellberg M E D P Balch and K Roy 2001 Climate drivenrange-expansion and morphological evolution in a marine gas-tropod Science 2921707ndash1710

Hewitt G M 1996 Some genetic consequences of ice ages andtheir role in divergence and speciation Biol J Linn Soc 58247ndash276

mdashmdashmdash 2000 The genetic legacy of the Quaternary ice ages Nature405907ndash913

Hoeh W R D T Stewart G W Sutherland and E Zouros 1996Multiple origins of gender-associated mitochondrial DNA lin-eages in bivalves (Mollusca Bivalvia) Evolution 502276ndash2286

Holder K R Montgomerie and V L Friesen 1999 A test of theglacial refugium hypothesis using patterns of mitochondrial andnuclear DNA sequence variation in rock ptarmigan (Lagopusmutus) Evolution 531936ndash1950

Holm E R and E Bourget 1994 Settlement behavior and mi-crogeographic population structure in the barnacle Semibalanusbalanoides Pp 154ndash158 in A R Beaumont ed Genetics andevolution of aquatic organisms Chapman and Hall London

Hudson R R 1990 Gene genealogies and the coalescent processPp 1ndash44 in D Futuyma and J Antonovics eds Oxford surveysin evolutionary biology Vol 7 Oxford Univ Press OxfordUK

Ingolfsson A 1992 The origin of the rocky shore fauna of Icelandand the Canadian maritimes J Biogeogr 19705ndash712

mdashmdashmdash 1995 Floating clumps of seaweed around Iceland naturalmicrocosms and a means of dispersal for shore fauna Mar Biol12213ndash21

Jablonski D and R A Lutz 1983 Larval ecology of marinebenthic invertebrates paleobiological implications Biol Rev5821ndash89

Jenkins J B A Morrison and C L MacKenzie 1997 The mol-luscan fisheries of the Canadian Maritimes NOAA Tech RepNMFS 12715ndash44

Johannesson K 1988 The paradox of Rockall Why is a broodinggastropod (Littorina saxatilis) more widespread than one havinga planktonic larval dispersal stage (L littorea) Mar Biol 99507ndash513

Johnson K P F R Adler and J L Cherry 2000 Genetic andphylogenetic consequences of island biogeography Evolution54387ndash396

Kit Y and I Kornfield 1998 Phylogenetic relationships of clawedlobster genera (Decapoda Nephropidae) based on mitochondrial16S rRNA gene sequences J Crustacean Biol 18138ndash146

Knott S T and H Hoskins 1968 Evidence of Pleistocene eventsin the structure of the continental shelf off the northeastern Unit-ed States Mar Geol 65ndash43

Knowlton N and L A Weigt 1998 New dates and new rates fordivergence across the Isthmus of Panama Proc R Soc LondB 2652257ndash2263

Kuhner M K J Yamato and J Felsenstein 1998 Maximumlikelihood estimation of population growth rates based on thecoalescent Genetics 149429ndash434

Locke A and S Corey 1989 Amphipods isopods and surfacecurrents a case for passive dispersal in the Bay of Fundy Can-ada J Plankton Res 11419ndash430

Marincovich L and A Y Gladenkov 1999 Evidence for an earlyopening of the Bering Strait Nature 397149ndash151

Markovtsova L P Marjoram and S Tavare 2000 The effects ofrate variation on ancestral inference in the coalescent Genetics1561427ndash1436

Nee S E C Holmes A Rambaut and P H Harvey 1995 In-ferring population history from molecular phylogenies PhilosTrans R Soc Lond B 34925ndash31

Nei M 1987 Molecular evolutionary genetics Columbia UnivPress New York

Nei M and W-H Li 1979 Mathematical model for studyinggenetic variation in terms of restriction endonucleases ProcNatl Acad Sci USA 765269ndash5273

Nielsen R J L Mountain J P Huelsenbeck and M Slatkin1998 Maximum-likelihood estimation of population divergencetimes and population phylogeny in models without mutationEvolution 52669ndash677

Pielou E C 1991 After the Ice Age the return of life to glaciatedNorth America Univ of Chicago Press Chicago IL

Posada D and K A Crandall 1998 Modeltest testing the modelof DNA substitution Bioinformatics 14817ndash818

Pratt R M and J Schlee 1969 Glaciation on the continentalmargin off New England Bull Geol Soc Am 802335ndash2342

Quesada H R Wenne and D O F Skibinski 1995 Differentialintrogression of mitochondrial DNA across species boundarieswithin the marine mussel genus Mytilus Proc R Soc Lond B26251ndash56

Quesada H D A G Skibinski and D O F Skibinski 1996 Sex-biased heteroplasmy and mitochondrial DNA inheritance in themussel Mytilus galloprovincialis Lmk Curr Genet 29423ndash426

Quesada H M Warren and D O F Skibinski 1998 Nonneutralevolution and differential mutation rate of gender-associated mi-tochondrial DNA lineages in the marine mussel Mytilus Genetics1491511ndash1526

Rawson P D K L Joyner K Meetze and T J Hilbish 1996Evidence for intragenic recombination within a novel geneticmarker that distinguishes mussels in the Mytilus edulis speciescomplex Heredity 77599ndash607

Reid D G E Rumbak and R H Thomas 1996 DNA mor-phology and fossils phylogeny and evolutionary rates of thegastropod genus Littorina Philos Trans R Soc Lond B 351877ndash895

Ricklefs R E and D Schluter 1993 Species diversity regionaland historical influences Pp 350ndash363 in R E Ricklefs and DSchluter eds Species diversity in ecological communities Univof Chicago Press Chicago IL

Riggs S R S W Snyder A C Hine and D L Mearns 1996Hardbottom morphology and relationship to the geologic frame-work mid-Atlantic continental shelf J Sediment Res 66830ndash846

Rogers A R and H Harpending 1992 Population growth makeswaves in the distribution of pairwise genetic differences MolBiol Evol 9552ndash569

Schneider S J Kueffer D Roessli and L Excoffier 1997 Ar-lequin ver 11 a software for population genetic data analysisGenetics and Biometry Laboratory Univ of Geneva Switzer-land


Schubart C D R Diesel and S B Hedges 1998 Rapid evolutionto terrestrial life in Jamaican crabs Nature 393363ndash365

Seutin G B N White and P T Boag 1991 Preservation of avianblood and tissue samples for DNA analyses Can J Zool 6982ndash90

Slatkin M 1991 Inbreeding coefficients and coalescence timesGenet Res 58167ndash175

mdashmdashmdash 1993 Isolation by distance in equilibrium and nonequilib-rium populations Evolution 47264ndash279

mdashmdashmdash 1995 A measure of population subdivision based on mi-crosatellite allele frequencies Genetics 139457ndash462

Slatkin M and R R Hudson 1991 Pairwise comparisons of mi-tochondrial sequences in stable and exponentially growing pop-ulations Genetics 129555ndash562

Stone L T Dayan and D Simberloff 1996 Community-wideassembly patterns unmasked the importance of species differinggeographical ranges Am Nat 148997ndash1015

Sullivan J E Arellano and D S Rogers 2000 Comparativephylogeography of Mesoamerican highland rodents concertedversus independent response to past climatic fluctuations AmNat 155755ndash768

Swofford D 1998 PAUP phylogenetic analysis using parsimonySinauer Associates Sunderland MA

Templeton A R 1998 Nested clade analyses of phylogeographicdata testing hypotheses about gene flow and population historyMol Ecol 7381ndash397

Templeton A R K A Crandall and C F Sing 1992 A cladisticanalysis of phenotypic associations with haplotypes inferredfrom restriction endonuclease mapping and DNA sequence dataIII Cladogram estimation Genetics 132619ndash633

Travis J 1996 The significance of geographical variation in speciesinteractions Am Nat 148S1ndashS8

Valentine J W and D Jablonski 1986 Mass extinctions sen-sitivity of marine larval types Proc Natl Acad Sci USA 836912ndash6914

van Oppen M J H S G A Draisma J L Olsen and W TStam 1995 Multiple trans-Arctic passages in the red alga Phy-codrys rubens evidence from nuclear rDNA ITS sequences MarBiol 123179ndash188

Vermeij G 1991 Anatomy of an invasion the trans-Arctic inter-change Paleobiology 17281ndash307

mdashmdashmdash 1992 Time of origin and biogeographical history of spe-cialized relationships between northern marine plants and her-bivorous molluscs Evolution 46657ndash664

Wakeley J 2000 The effects of subdivision on the genetic diver-gence of populations and species Evolution 541092ndash1101

Wares J P 2001a Patterns of speciation inferred from mitochon-drial DNA in North American Chthamalus (Cirripedia Balan-omorpha Chthamaloidea) Mol Phylogenet Evol 18104ndash116

mdashmdashmdash 2001b Intraspecific variation and geographic isolation inIdotea balthica (Isopoda Valvifera) J Crustacean Biol 211007ndash1013

mdashmdashmdash 2001c Biogeography of Asterias North Atlantic climatechange and speciation Biol Bull 20195ndash103

Watterson G A 1975 On the number of segregating sites in ge-netical models without recombination Theor Popul Biol 7256ndash276

Whitlock M C and D E McCauley 1999 Indirect measures ofgene flow and migration Fst plusmn 1(4Nm 1 1) Heredity 82117ndash125

Wollenberg K and J C Avise 1998 Sampling properties of ge-nealogical pathways underlying population pedigrees Evolution52957ndash966

Worley E K and D R Franz 1983 A comparative study ofselected skeletal structures in the seastars Asterias forbesi Avulgaris and A rubens with a discussion of possible relation-ships Proc Biol Soc Wash 96524ndash527

Yang Z 1994 Maximum likelihood phylogenetic estimation fromDNA sequences with variable rates over sites approximatemethods J Mol Evol 39306ndash314

Zink R M R C Blackwell-Rago and F Ronquist 2000 Theshifting roles of dispersal and vicariance in biogeography ProcR Soc Lond B 267497ndash503

Corresponding Editor R Burton

APPENDIX 1Populations Represented in Common or Shared Alleles in Figure 1Asterias rubensHaplotype A individuals from Maine (n 5 8) Nova Scotia (n 52) Newfoundland (n 5 2) Iceland (n 5 1) Norway (n 5 2) andIreland (n 5 1)Haplotype B individuals from Maine (n 5 2) Nova Scotia (n 54) Newfoundland (n 5 3) Iceland (n 5 1) Ireland (n 5 2) andFrance (n 5 2)Haplotype C individuals from Maine (n 5 1) Norway (n 5 3)Ireland (n 5 2) and France (n 5 1)Haplotype D individuals from Maine (n 5 1) and Ireland (n 5 1)

Idotea balthicaNorth American common haplotype represents 13 individualsMaine (n 5 8) Rhode Island (n 5 1) and Nova Scotia (n 5 5)

Mytilus edulisHaplotype NA individuals from North Carolina (n 5 1) Maine (n5 5) and Nova Scotia (n 5 1)Haplotype A individuals from North Carolina (n 5 3) Rhode Island(n 5 3) Maine (n 5 6) Iceland (n 5 3) Norway (n 5 1) andFrance (n 5 2)Haplotype Eur individuals from Iceland (n 5 2) Norway (n 5 1)Ireland (n 5 1) and France (n 5 3)Haplotype 9Fr individuals sampled at Roscoff France

Littorina obtusataShared Haplotype individuals from Maine (n 5 7) Nova Scotia(n 5 8) Newfoundland (n 5 4) Iceland (n 5 3) and Norway (n5 1)

Nucella lapillusShared Haplotype individuals from Maine (n 5 14) Nova Scotia(n 5 7) Rhode Island (n 5 3) Iceland (n 5 8) Norway (n 5 7)Ireland (n5 5 7) and Newfoundland (n 5 5)

Semibalanus balanoidesHaplotype A individuals from Maine (n 5 4) Nova Scotia (n 55) Rhode Island (n 5 2) Iceland (n 5 7) and Newfoundland (n5 1)Haplotype B individuals from Rhode Island (n 5 2) Maine (n 57) Nova Scotia (n 5 2) Newfoundland (n 5 2) and Iceland (n 5 4)


APPENDIX 2

AMOVA results for each species being compared

Species Source of variation (df )Percentage of

variation Fixation indices

Asterias rubens North America vs EuropeWithin continental groupsWithin populations

(1)(5)

(41)

911123

8966

FCT 5 0091FSC 5 0013FST 5 0103

Mytilus edulis North America vs EuropeWithin continental groupsWithin populations

(1)(8)

(78)

61621907194

FCT 5 0062FSC 5 0233FST 5 0281

Idotea balthica North America vs EuropeWithin continental groupsWithin populations

(1)(3)

(28)

2143766914747

FCT 5 20144FSC 5 0585FST 5 0525

Nucella lapillus North America vs EuropeWithin continental groupsWithin populations

(1)(5)

(51)

2117765114666

FCT 5 20117FSC 5 0582FST 5 0533

Littorina obtusata North America vs EuropeWithin continental groupsWithin populations

(1)(5)

(38)

4189387

5424

FCT 5 0418FSC 5 0066FST 5 0457

Semibalanus balanoides Pacific vs North America vs EuropeWithin continental groupsWithin populations(Atlantic clades only)Between Atlantic cladesWithin Atlantic cladesWithin populations

(2)(6)

(59)

(1)(2)

(60)

6023684

3293

7322606

2072

FCT 5 0602FSC 5 0172FST 5 0671

FCT 5 0732FSC 5 0226FST 5 0792

P 005

2455

q 2001 The Society for the Study of Evolution All rights reserved

Evolution 55(12) 2001 pp 2455ndash2469

PHYLOGEOGRAPHY AND HISTORICAL ECOLOGY OF THENORTH ATLANTIC INTERTIDAL

JOHN P WARES12 AND CLIFFORD W CUNNINGHAM1

1Duke University Department of Biology Box 90325Durham North Carolina 27708-0325

Abstract Recent glaciation covered the full extent of rocky intertidal habitat along the coasts of New England andthe Canadian Maritimes To test whether this glaciation in fact caused wholesale extinction of obligate rocky intertidalinvertebrates and thus required a recolonization from Europe we compared American and European populations usingallelic diversity and techniques adapted from coalescent theory Mitochondrial DNA sequences were collected fromamphi-Atlantic populations of three cold-temperate obligate rocky intertidal species (a barnacle Semibalanus bal-anoides and two gastropods Nucella lapillus and Littorina obtusata) and three cold-temperate habitat generalist species(a seastar Asterias rubens a mussel Mytilus edulis and an isopod Idotea balthica) For many of these species wewere able to estimate the lineage-specific mutation rate based on trans-Arctic divergences between Pacific and Atlantictaxa These data indicate that some obligate rocky intertidal taxa have colonized New England from European pop-ulations However the patterns of persistence in North America indicate that other life-history traits including mech-anisms of dispersal may be more important for surviving dramatic environmental and climatic change

Key words Coalescent glaciation North Atlantic phylogeography range expansion

Received January 17 2001 Accepted August 29 2001

A primary goal of historical ecology is to determine howcommunities and biological interactions develop (Ricklefsand Schluter 1993 Stone et al 1996) Interactions amongspecies may vary geographically based on the ecological con-text and such variation may arise in very recent species as-sociations (Vermeij 1992 Travis 1996) For example theisopod Idotea balthica is a more dominant herbivore and isfound in a broader range of algal and seagrass habitats (J PWares pers obs) in New England populations than in Eu-ropean populations In European populations this speciestends to be ecologically marginalized through competitionwith congeneric grazers (Franke and Janke 1998) What eco-logical variables play a determinant role in producing thisvariation in community composition and interactions Onlyby knowing both the historical and ecological contexts ofthese interactions can we decide (Endler 1982 Brooks andMcLennan 1991 Vermeij 1992) and studies of comparativephylogeography represent a solid approach to this problem

In North Atlantic intertidal marine communities the lastglacial maximum (around 20000 years ago Barash et al1992 Holder et al 1999) is thought to have been especiallydifficult for obligate rocky intertidal species on the NorthAmerican coast Many such species may have been forcedfrom their typical habitat by Pleistocene glaciation becausethere is currently little hard substrate beyond the southern-most extent of the glaciers at Long Island Sound (Knott andHoskins 1968 Ingolfsson 1992 Riggs et al 1996) If gla-ciation caused localized extinction in New England rockyintertidal species then much of the extant rocky intertidalcommunity may have recently immigrated from Europewhere hard substrata were more available and Pleistoceneclimatic changes were not as severe (Vermeij 1991 van Op-pen et al 1995) Distributional evidence suggests this is alikely scenario because the list of obligate rocky intertidal

2 Present address Department of Biology University of NewMexico Castetter Hall Albuquerque New Mexico 87131-1091E-mail jpwaresunmedu

species in North America is a subset of the same group ofspecies found in Iceland which in turn represents a subsetof these species found in northern Europe (Ingolfsson 1992)Although many marine species may have been able to shiftor contract their geographic range into southern refugia alongthe North American coast species requiring hard substratamay not have been able to find suitable habitat

Analysis of data from multiple species with varied life-history traits (including larval dispersal trophic level andecological traits such as substrate requirements) allows us tofocus on species or groups of species that did not survivethe most recent glacial maximum in North America Popu-lations of these species in New England should be recogniz-able based on two basic genetic patterns First recentlyfounded populations are expected to represent a subset of thegenetic diversity in the source population This leads to theprediction that recently founded populations should have asignificantly lower genetic diversity and have a high fre-quency of alleles that are identical to or descended fromalleles in the founding population (Hewitt 1996 2000 Aus-terlitz et al 1997 Johnson et al 2000 Zink et al 2000)Second recently founded populations should show the ge-netic signature of a rapidly expanding population (eg Neeet al 1995 Kuhner et al 1998) in part because they areunlikely to have reached an equilibrium between allelic di-versity (mutation) and demographic changes (genetic drift)

In this paper we use genetic markers to describe the post-glacial reassembly of the intertidal marine community in NewEngland and the Canadian Maritimes We will focus on thehypothesis that substratum requirements have played a majorrole in North Atlantic historical biogeography Because thereare recognizable patterns that we expect in recently foundedpopulations we test these predictions using both standardpopulation genetic measures as well as recently developedcoalescent methods Coalescent methods (for review seeHudson 1990) are expected to more accurately estimate thesepatterns than methods using pairwise sequence comparisons







M NS W Ic NoIr Fr


627 bpAF240022ndash240081






516 bpAF241889ndash241935





463 bpAF241936ndash142035





537 bpAF242067ndash242117





615 bpAF242660ndash242728






















2(gm)tN 5 Qe t (1)


RESULTS















0597 6 00630771 6 0052

0275 6 0015 0662 6 01160893 6 00320186 6 01100145 6 00890257 6 0142

0893 6 00380936 6 0033

0923 6 00250933 6 00250434 6 01010942 6 0038




484 3 1028 (877 3 1029)951 3 1028 (196 3 1028)249 3 1028 (13 3 1028)443 3 1028 (107 3 1028)276 3 1028 (632 3 1029)


36 3 1028 (see text)21 3 1028

19 3 1028

38 3 1028










Asterias rubens



TA

BL

E4

Est

imat

esof

Q(t

heco

mpo

und

para

met

erre

pres

enti

ngth

eef

fect

ive

popu

lati

onsi

zean

dsu

bsti

tuti

onra

te)

and

g(t

heex

pone

ntia

lgr

owth

para

met

er)

for

spec

ies

inth

isst

udy

usin

gF

luct

uate

(Kuh

ner

etal

19

98)

Eac

hsp

ecie

sis

divi

ded

into

Nor

thA

mer

ican

and

Eur

opea

npo

pula

tion

sto

calc

ulat

eth

era

nge

ofpr

ojec

ted

ages

for

each

regi

on

The

Flu

ctua

tees

tim

ates

(the

com

posi

teof

five

repl

icat

ean

alys

es)

indi

cate

the

max

imum

-lik

elih

ood

esti

mat

eof

the

join

tpr

obab

ilit

yof

Qan

dg

The

esti

mat

esof

Ndu

ring

the

mos

tre

cent

glac

ial

max

imum

(20

000

year

sag

o)re

lati

veto

pres

ent-

day

esti

mat

esof

Nar

eba

sed

ong

and

m

The

stan

dard

devi

atio

nis

pres

ente

daf

ter

each

esti

mat

or

Est

imat

edpa

ram

eter

sfo

rId

otea

()

base

don

resu

lts

usin

gal

lco

don

posi

tion

sas

the

Nor

thA

mer

ican

popu

lati

onis

inva

rian

tat

thir

dpo

siti

onH

owev

er

the

max

imum

-lik

elih

ood

estim

ate

ofg

ism

axim

alus

ing

eith

eral

lco

don

posi

tion

sor

ada

tase

tsi

mul

ated

usin

gth

esa

me

mut

atio

nal

para

met

ers

asth

eId

otea

thir

dpo

siti

onda

ta

and

this

esti

mat

eis

conc

orda

ntw

ith

high

esti

mat

esof

gin

othe

rsp

ecie

sw

ith

very

low

Nor

thA

mer

ican

alle

lic

dive

rsit

y(L

itto

rina

N

ucel

la)

Spe

cies

Pop

ulat

ion

Q(n

ogr

owth

)Q

(joi

ntly

esti

mat

ed)

gT

ime

to1

rela

tive

NR

elat

ive

N20

000

year

sag

o

Ast

eria

sru

bens

Idot

eaba

lthi

ca

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

000

429

026

63

000

398

005

004

000

566

000

2

041

906

030

1

034

046

032

80

1883

60

012

875

36

437

4

962

16

314

2

100

00

278

36

186

104

800

(59

500

ndash25

390

0)95

400

(61

200

ndash17

160

0)13

400

482

700

017

ndash0

99

022

ndash0

72

25

310

25ndash

001

10

81ndash

085

Myt

ilus

edul

is

Lit

tori

naob

tusa

ta

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

002

965

003

264

000

248

003

137

012

596

004

5

031

806

009

53

001

346

001

15

011

706

002

28

109

46

494

356

66

553

100

006

(19

46)

493

66

393

184

310

6

(11

2ndash

389

310

6 )5

663

105

(43

2ndash

774

310

5 )18

500

(11

047

ndash38

389

)3

743

105

(24

7ndash

680

310

5 )

091

ndash0

99

081

ndash0

89

68

310

23ndash

68

310

23

075

ndash0

82

Sem

ibal

anus

bala

noid

esN

orth

Am

eric

a

Sou

thcl

ade

Nor

thcl

ade

na

000

309

002

341

003

256

000

23

715

316

392

3

044

566

059

94

308

65

9

2326

36

460

2

100

006

(11

38)

514

310

6

(35

4ndash

814

310

6 )68

000

(46

685

ndash10

840

1)15

800

(11

679

ndash22

826

)

098

ndash0

99

015

ndash0

44

29

310

23ndash

29

310

23

Nuc

ella

lapi

llus

Eur

ope

Nor

thA

mer

ica

Eur

ope

006

347

000

434

000

534

018

266

001

73

009

666

007

49

053

796

053

42

176

06

263

100

006

(11

61)

4391

46

396

7

899

310

5

(64

23

105ndash

135

310

6 )10

400

(747

9ndash

154

96)

236

00(1

740

0ndash

343

00)

087

ndash0

93

14

310

23ndash

14

310

23

001

ndash0

04









Mytilus edulis




Idotea balthica



DISCUSSION












































ACKNOWLEDGMENTS


LITERATURE CITED











































































































APPENDIX 2





(1)(5)

(41)

911123

8966

FCT 5 0091FSC 5 0013FST 5 0103


(1)(8)

(78)

61621907194

FCT 5 0062FSC 5 0233FST 5 0281


(1)(3)

(28)

2143766914747

FCT 5 20144FSC 5 0585FST 5 0525


(1)(5)

(51)

2117765114666

FCT 5 20117FSC 5 0582FST 5 0533


(1)(5)

(38)

4189387

5424

FCT 5 0418FSC 5 0066FST 5 0457


(2)(6)

(59)

(1)(2)

(60)

6023684

3293

7322606

2072

FCT 5 0602FSC 5 0172FST 5 0671

FCT 5 0732FSC 5 0226FST 5 0792

P 005







M NS W Ic NoIr Fr


627 bpAF240022ndash240081






516 bpAF241889ndash241935





463 bpAF241936ndash142035





537 bpAF242067ndash242117





615 bpAF242660ndash242728






















2(gm)tN 5 Qe t (1)


RESULTS















0597 6 00630771 6 0052

0275 6 0015 0662 6 01160893 6 00320186 6 01100145 6 00890257 6 0142

0893 6 00380936 6 0033

0923 6 00250933 6 00250434 6 01010942 6 0038




484 3 1028 (877 3 1029)951 3 1028 (196 3 1028)249 3 1028 (13 3 1028)443 3 1028 (107 3 1028)276 3 1028 (632 3 1029)


36 3 1028 (see text)21 3 1028

19 3 1028

38 3 1028










Asterias rubens



TA

BL

E4

Est

imat

esof

Q(t

heco

mpo

und

para

met

erre

pres

enti

ngth

eef

fect

ive

popu

lati

onsi

zean

dsu

bsti

tuti

onra

te)

and

g(t

heex

pone

ntia

lgr

owth

para

met

er)

for

spec

ies

inth

isst

udy

usin

gF

luct

uate

(Kuh

ner

etal

19

98)

Eac

hsp

ecie

sis

divi

ded

into

Nor

thA

mer

ican

and

Eur

opea

npo

pula

tion

sto

calc

ulat

eth

era

nge

ofpr

ojec

ted

ages

for

each

regi

on

The

Flu

ctua

tees

tim

ates

(the

com

posi

teof

five

repl

icat

ean

alys

es)

indi

cate

the

max

imum

-lik

elih

ood

esti

mat

eof

the

join

tpr

obab

ilit

yof

Qan

dg

The

esti

mat

esof

Ndu

ring

the

mos

tre

cent

glac

ial

max

imum

(20

000

year

sag

o)re

lati

veto

pres

ent-

day

esti

mat

esof

Nar

eba

sed

ong

and

m

The

stan

dard

devi

atio

nis

pres

ente

daf

ter

each

esti

mat

or

Est

imat

edpa

ram

eter

sfo

rId

otea

()

base

don

resu

lts

usin

gal

lco

don

posi

tion

sas

the

Nor

thA

mer

ican

popu

lati

onis

inva

rian

tat

thir

dpo

siti

onH

owev

er

the

max

imum

-lik

elih

ood

estim

ate

ofg

ism

axim

alus

ing

eith

eral

lco

don

posi

tion

sor

ada

tase

tsi

mul

ated

usin

gth

esa

me

mut

atio

nal

para

met

ers

asth

eId

otea

thir

dpo

siti

onda

ta

and

this

esti

mat

eis

conc

orda

ntw

ith

high

esti

mat

esof

gin

othe

rsp

ecie

sw

ith

very

low

Nor

thA

mer

ican

alle

lic

dive

rsit

y(L

itto

rina

N

ucel

la)

Spe

cies

Pop

ulat

ion

Q(n

ogr

owth

)Q

(joi

ntly

esti

mat

ed)

gT

ime

to1

rela

tive

NR

elat

ive

N20

000

year

sag

o

Ast

eria

sru

bens

Idot

eaba

lthi

ca

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

000

429

026

63

000

398

005

004

000

566

000

2

041

906

030

1

034

046

032

80

1883

60

012

875

36

437

4

962

16

314

2

100

00

278

36

186

104

800

(59

500

ndash25

390

0)95

400

(61

200

ndash17

160

0)13

400

482

700

017

ndash0

99

022

ndash0

72

25

310

25ndash

001

10

81ndash

085

Myt

ilus

edul

is

Lit

tori

naob

tusa

ta

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

002

965

003

264

000

248

003

137

012

596

004

5

031

806

009

53

001

346

001

15

011

706

002

28

109

46

494

356

66

553

100

006

(19

46)

493

66

393

184

310

6

(11

2ndash

389

310

6 )5

663

105

(43

2ndash

774

310

5 )18

500

(11

047

ndash38

389

)3

743

105

(24

7ndash

680

310

5 )

091

ndash0

99

081

ndash0

89

68

310

23ndash

68

310

23

075

ndash0

82

Sem

ibal

anus

bala

noid

esN

orth

Am

eric

a

Sou

thcl

ade

Nor

thcl

ade

na

000

309

002

341

003

256

000

23

715

316

392

3

044

566

059

94

308

65

9

2326

36

460

2

100

006

(11

38)

514

310

6

(35

4ndash

814

310

6 )68

000

(46

685

ndash10

840

1)15

800

(11

679

ndash22

826

)

098

ndash0

99

015

ndash0

44

29

310

23ndash

29

310

23

Nuc

ella

lapi

llus

Eur

ope

Nor

thA

mer

ica

Eur

ope

006

347

000

434

000

534

018

266

001

73

009

666

007

49

053

796

053

42

176

06

263

100

006

(11

61)

4391

46

396

7

899

310

5

(64

23

105ndash

135

310

6 )10

400

(747

9ndash

154

96)

236

00(1

740

0ndash

343

00)

087

ndash0

93

14

310

23ndash

14

310

23

001

ndash0

04









Mytilus edulis




Idotea balthica



DISCUSSION












































ACKNOWLEDGMENTS


LITERATURE CITED











































































































APPENDIX 2





(1)(5)

(41)

911123

8966

FCT 5 0091FSC 5 0013FST 5 0103


(1)(8)

(78)

61621907194

FCT 5 0062FSC 5 0233FST 5 0281


(1)(3)

(28)

2143766914747

FCT 5 20144FSC 5 0585FST 5 0525


(1)(5)

(51)

2117765114666

FCT 5 20117FSC 5 0582FST 5 0533


(1)(5)

(38)

4189387

5424

FCT 5 0418FSC 5 0066FST 5 0457


(2)(6)

(59)

(1)(2)

(60)

6023684

3293

7322606

2072

FCT 5 0602FSC 5 0172FST 5 0671

FCT 5 0732FSC 5 0226FST 5 0792

P 005








2(gm)tN 5 Qe t (1)


RESULTS















0597 6 00630771 6 0052

0275 6 0015 0662 6 01160893 6 00320186 6 01100145 6 00890257 6 0142

0893 6 00380936 6 0033

0923 6 00250933 6 00250434 6 01010942 6 0038




484 3 1028 (877 3 1029)951 3 1028 (196 3 1028)249 3 1028 (13 3 1028)443 3 1028 (107 3 1028)276 3 1028 (632 3 1029)


36 3 1028 (see text)21 3 1028

19 3 1028

38 3 1028










Asterias rubens



TA

BL

E4

Est

imat

esof

Q(t

heco

mpo

und

para

met

erre

pres

enti

ngth

eef

fect

ive

popu

lati

onsi

zean

dsu

bsti

tuti

onra

te)

and

g(t

heex

pone

ntia

lgr

owth

para

met

er)

for

spec

ies

inth

isst

udy

usin

gF

luct

uate

(Kuh

ner

etal

19

98)

Eac

hsp

ecie

sis

divi

ded

into

Nor

thA

mer

ican

and

Eur

opea

npo

pula

tion

sto

calc

ulat

eth

era

nge

ofpr

ojec

ted

ages

for

each

regi

on

The

Flu

ctua

tees

tim

ates

(the

com

posi

teof

five

repl

icat

ean

alys

es)

indi

cate

the

max

imum

-lik

elih

ood

esti

mat

eof

the

join

tpr

obab

ilit

yof

Qan

dg

The

esti

mat

esof

Ndu

ring

the

mos

tre

cent

glac

ial

max

imum

(20

000

year

sag

o)re

lati

veto

pres

ent-

day

esti

mat

esof

Nar

eba

sed

ong

and

m

The

stan

dard

devi

atio

nis

pres

ente

daf

ter

each

esti

mat

or

Est

imat

edpa

ram

eter

sfo

rId

otea

()

base

don

resu

lts

usin

gal

lco

don

posi

tion

sas

the

Nor

thA

mer

ican

popu

lati

onis

inva

rian

tat

thir

dpo

siti

onH

owev

er

the

max

imum

-lik

elih

ood

estim

ate

ofg

ism

axim

alus

ing

eith

eral

lco

don

posi

tion

sor

ada

tase

tsi

mul

ated

usin

gth

esa

me

mut

atio

nal

para

met

ers

asth

eId

otea

thir

dpo

siti

onda

ta

and

this

esti

mat

eis

conc

orda

ntw

ith

high

esti

mat

esof

gin

othe

rsp

ecie

sw

ith

very

low

Nor

thA

mer

ican

alle

lic

dive

rsit

y(L

itto

rina

N

ucel

la)

Spe

cies

Pop

ulat

ion

Q(n

ogr

owth

)Q

(joi

ntly

esti

mat

ed)

gT

ime

to1

rela

tive

NR

elat

ive

N20

000

year

sag

o

Ast

eria

sru

bens

Idot

eaba

lthi

ca

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

000

429

026

63

000

398

005

004

000

566

000

2

041

906

030

1

034

046

032

80

1883

60

012

875

36

437

4

962

16

314

2

100

00

278

36

186

104

800

(59

500

ndash25

390

0)95

400

(61

200

ndash17

160

0)13

400

482

700

017

ndash0

99

022

ndash0

72

25

310

25ndash

001

10

81ndash

085

Myt

ilus

edul

is

Lit

tori

naob

tusa

ta

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

002

965

003

264

000

248

003

137

012

596

004

5

031

806

009

53

001

346

001

15

011

706

002

28

109

46

494

356

66

553

100

006

(19

46)

493

66

393

184

310

6

(11

2ndash

389

310

6 )5

663

105

(43

2ndash

774

310

5 )18

500

(11

047

ndash38

389

)3

743

105

(24

7ndash

680

310

5 )

091

ndash0

99

081

ndash0

89

68

310

23ndash

68

310

23

075

ndash0

82

Sem

ibal

anus

bala

noid

esN

orth

Am

eric

a

Sou

thcl

ade

Nor

thcl

ade

na

000

309

002

341

003

256

000

23

715

316

392

3

044

566

059

94

308

65

9

2326

36

460

2

100

006

(11

38)

514

310

6

(35

4ndash

814

310

6 )68

000

(46

685

ndash10

840

1)15

800

(11

679

ndash22

826

)

098

ndash0

99

015

ndash0

44

29

310

23ndash

29

310

23

Nuc

ella

lapi

llus

Eur

ope

Nor

thA

mer

ica

Eur

ope

006

347

000

434

000

534

018

266

001

73

009

666

007

49

053

796

053

42

176

06

263

100

006

(11

61)

4391

46

396

7

899

310

5

(64

23

105ndash

135

310

6 )10

400

(747

9ndash

154

96)

236

00(1

740

0ndash

343

00)

087

ndash0

93

14

310

23ndash

14

310

23

001

ndash0

04









Mytilus edulis




Idotea balthica



DISCUSSION












































ACKNOWLEDGMENTS


LITERATURE CITED











































































































APPENDIX 2





(1)(5)

(41)

911123

8966

FCT 5 0091FSC 5 0013FST 5 0103


(1)(8)

(78)

61621907194

FCT 5 0062FSC 5 0233FST 5 0281


(1)(3)

(28)

2143766914747

FCT 5 20144FSC 5 0585FST 5 0525


(1)(5)

(51)

2117765114666

FCT 5 20117FSC 5 0582FST 5 0533


(1)(5)

(38)

4189387

5424

FCT 5 0418FSC 5 0066FST 5 0457


(2)(6)

(59)

(1)(2)

(60)

6023684

3293

7322606

2072

FCT 5 0602FSC 5 0172FST 5 0671

FCT 5 0732FSC 5 0226FST 5 0792

P 005







0597 6 00630771 6 0052

0275 6 0015 0662 6 01160893 6 00320186 6 01100145 6 00890257 6 0142

0893 6 00380936 6 0033

0923 6 00250933 6 00250434 6 01010942 6 0038




484 3 1028 (877 3 1029)951 3 1028 (196 3 1028)249 3 1028 (13 3 1028)443 3 1028 (107 3 1028)276 3 1028 (632 3 1029)


36 3 1028 (see text)21 3 1028

19 3 1028

38 3 1028










Asterias rubens



TA

BL

E4

Est

imat

esof

Q(t

heco

mpo

und

para

met

erre

pres

enti

ngth

eef

fect

ive

popu

lati

onsi

zean

dsu

bsti

tuti

onra

te)

and

g(t

heex

pone

ntia

lgr

owth

para

met

er)

for

spec

ies

inth

isst

udy

usin

gF

luct

uate

(Kuh

ner

etal

19

98)

Eac

hsp

ecie

sis

divi

ded

into

Nor

thA

mer

ican

and

Eur

opea

npo

pula

tion

sto

calc

ulat

eth

era

nge

ofpr

ojec

ted

ages

for

each

regi

on

The

Flu

ctua

tees

tim

ates

(the

com

posi

teof

five

repl

icat

ean

alys

es)

indi

cate

the

max

imum

-lik

elih

ood

esti

mat

eof

the

join

tpr

obab

ilit

yof

Qan

dg

The

esti

mat

esof

Ndu

ring

the

mos

tre

cent

glac

ial

max

imum

(20

000

year

sag

o)re

lati

veto

pres

ent-

day

esti

mat

esof

Nar

eba

sed

ong

and

m

The

stan

dard

devi

atio

nis

pres

ente

daf

ter

each

esti

mat

or

Est

imat

edpa

ram

eter

sfo

rId

otea

()

base

don

resu

lts

usin

gal

lco

don

posi

tion

sas

the

Nor

thA

mer

ican

popu

lati

onis

inva

rian

tat

thir

dpo

siti

onH

owev

er

the

max

imum

-lik

elih

ood

estim

ate

ofg

ism

axim

alus

ing

eith

eral

lco

don

posi

tion

sor

ada

tase

tsi

mul

ated

usin

gth

esa

me

mut

atio

nal

para

met

ers

asth

eId

otea

thir

dpo

siti

onda

ta

and

this

esti

mat

eis

conc

orda

ntw

ith

high

esti

mat

esof

gin

othe

rsp

ecie

sw

ith

very

low

Nor

thA

mer

ican

alle

lic

dive

rsit

y(L

itto

rina

N

ucel

la)

Spe

cies

Pop

ulat

ion

Q(n

ogr

owth

)Q

(joi

ntly

esti

mat

ed)

gT

ime

to1

rela

tive

NR

elat

ive

N20

000

year

sag

o

Ast

eria

sru

bens

Idot

eaba

lthi

ca

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

000

429

026

63

000

398

005

004

000

566

000

2

041

906

030

1

034

046

032

80

1883

60

012

875

36

437

4

962

16

314

2

100

00

278

36

186

104

800

(59

500

ndash25

390

0)95

400

(61

200

ndash17

160

0)13

400

482

700

017

ndash0

99

022

ndash0

72

25

310

25ndash

001

10

81ndash

085

Myt

ilus

edul

is

Lit

tori

naob

tusa

ta

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

002

965

003

264

000

248

003

137

012

596

004

5

031

806

009

53

001

346

001

15

011

706

002

28

109

46

494

356

66

553

100

006

(19

46)

493

66

393

184

310

6

(11

2ndash

389

310

6 )5

663

105

(43

2ndash

774

310

5 )18

500

(11

047

ndash38

389

)3

743

105

(24

7ndash

680

310

5 )

091

ndash0

99

081

ndash0

89

68

310

23ndash

68

310

23

075

ndash0

82

Sem

ibal

anus

bala

noid

esN

orth

Am

eric

a

Sou

thcl

ade

Nor

thcl

ade

na

000

309

002

341

003

256

000

23

715

316

392

3

044

566

059

94

308

65

9

2326

36

460

2

100

006

(11

38)

514

310

6

(35

4ndash

814

310

6 )68

000

(46

685

ndash10

840

1)15

800

(11

679

ndash22

826

)

098

ndash0

99

015

ndash0

44

29

310

23ndash

29

310

23

Nuc

ella

lapi

llus

Eur

ope

Nor

thA

mer

ica

Eur

ope

006

347

000

434

000

534

018

266

001

73

009

666

007

49

053

796

053

42

176

06

263

100

006

(11

61)

4391

46

396

7

899

310

5

(64

23

105ndash

135

310

6 )10

400

(747

9ndash

154

96)

236

00(1

740

0ndash

343

00)

087

ndash0

93

14

310

23ndash

14

310

23

001

ndash0

04









Mytilus edulis




Idotea balthica



DISCUSSION












































ACKNOWLEDGMENTS


LITERATURE CITED











































































































APPENDIX 2





(1)(5)

(41)

911123

8966

FCT 5 0091FSC 5 0013FST 5 0103


(1)(8)

(78)

61621907194

FCT 5 0062FSC 5 0233FST 5 0281


(1)(3)

(28)

2143766914747

FCT 5 20144FSC 5 0585FST 5 0525


(1)(5)

(51)

2117765114666

FCT 5 20117FSC 5 0582FST 5 0533


(1)(5)

(38)

4189387

5424

FCT 5 0418FSC 5 0066FST 5 0457


(2)(6)

(59)

(1)(2)

(60)

6023684

3293

7322606

2072

FCT 5 0602FSC 5 0172FST 5 0671

FCT 5 0732FSC 5 0226FST 5 0792

P 005


TA

BL

E4

Est

imat

esof

Q(t

heco

mpo

und

para

met

erre

pres

enti

ngth

eef

fect

ive

popu

lati

onsi

zean

dsu

bsti

tuti

onra

te)

and

g(t

heex

pone

ntia

lgr

owth

para

met

er)

for

spec

ies

inth

isst

udy

usin

gF

luct

uate

(Kuh

ner

etal

19

98)

Eac

hsp

ecie

sis

divi

ded

into

Nor

thA

mer

ican

and

Eur

opea

npo

pula

tion

sto

calc

ulat

eth

era

nge

ofpr

ojec

ted

ages

for

each

regi

on

The

Flu

ctua

tees

tim

ates

(the

com

posi

teof

five

repl

icat

ean

alys

es)

indi

cate

the

max

imum

-lik

elih

ood

esti

mat

eof

the

join

tpr

obab

ilit

yof

Qan

dg

The

esti

mat

esof

Ndu

ring

the

mos

tre

cent

glac

ial

max

imum

(20

000

year

sag

o)re

lati

veto

pres

ent-

day

esti

mat

esof

Nar

eba

sed

ong

and

m

The

stan

dard

devi

atio

nis

pres

ente

daf

ter

each

esti

mat

or

Est

imat

edpa

ram

eter

sfo

rId

otea

()

base

don

resu

lts

usin

gal

lco

don

posi

tion

sas

the

Nor

thA

mer

ican

popu

lati

onis

inva

rian

tat

thir

dpo

siti

onH

owev

er

the

max

imum

-lik

elih

ood

estim

ate

ofg

ism

axim

alus

ing

eith

eral

lco

don

posi

tion

sor

ada

tase

tsi

mul

ated

usin

gth

esa

me

mut

atio

nal

para

met

ers

asth

eId

otea

thir

dpo

siti

onda

ta

and

this

esti

mat

eis

conc

orda

ntw

ith

high

esti

mat

esof

gin

othe

rsp

ecie

sw

ith

very

low

Nor

thA

mer

ican

alle

lic

dive

rsit

y(L

itto

rina

N

ucel

la)

Spe

cies

Pop

ulat

ion

Q(n

ogr

owth

)Q

(joi

ntly

esti

mat

ed)

gT

ime

to1

rela

tive

NR

elat

ive

N20

000

year

sag

o

Ast

eria

sru

bens

Idot

eaba

lthi

ca

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

000

429

026

63

000

398

005

004

000

566

000

2

041

906

030

1

034

046

032

80

1883

60

012

875

36

437

4

962

16

314

2

100

00

278

36

186

104

800

(59

500

ndash25

390

0)95

400

(61

200

ndash17

160

0)13

400

482

700

017

ndash0

99

022

ndash0

72

25

310

25ndash

001

10

81ndash

085

Myt

ilus

edul

is

Lit

tori

naob

tusa

ta

Nor

thA

mer

ica

Eur

ope

Nor

thA

mer

ica

Eur

ope

002

965

003

264

000

248

003

137

012

596

004

5

031

806

009

53

001

346

001

15

011

706

002

28

109

46

494

356

66

553

100

006

(19

46)

493

66

393

184

310

6

(11

2ndash

389

310

6 )5

663

105

(43

2ndash

774

310

5 )18

500

(11

047

ndash38

389

)3

743

105

(24

7ndash

680

310

5 )

091

ndash0

99

081

ndash0

89

68

310

23ndash

68

310

23

075

ndash0

82

Sem

ibal

anus

bala

noid

esN

orth

Am

eric

a

Sou

thcl

ade

Nor

thcl

ade

na

000

309

002

341

003

256

000

23

715

316

392

3

044

566

059

94

308

65

9

2326

36

460

2

100

006

(11

38)

514

310

6

(35

4ndash

814

310

6 )68

000

(46

685

ndash10

840

1)15

800

(11

679

ndash22

826

)

098

ndash0

99

015

ndash0

44

29

310

23ndash

29

310

23

Nuc

ella

lapi

llus

Eur

ope

Nor

thA

mer

ica

Eur

ope

006

347

000

434

000

534

018

266

001

73

009

666

007

49

053

796

053

42

176

06

263

100

006

(11

61)

4391

46

396

7

899

310

5

(64

23

105ndash

135

310

6 )10

400

(747

9ndash

154

96)

236

00(1

740

0ndash

343

00)

087

ndash0

93

14

310

23ndash

14

310

23

001

ndash0

04









Mytilus edulis




Idotea balthica



DISCUSSION












































ACKNOWLEDGMENTS


LITERATURE CITED











































































































APPENDIX 2





(1)(5)

(41)

911123

8966

FCT 5 0091FSC 5 0013FST 5 0103


(1)(8)

(78)

61621907194

FCT 5 0062FSC 5 0233FST 5 0281


(1)(3)

(28)

2143766914747

FCT 5 20144FSC 5 0585FST 5 0525


(1)(5)

(51)

2117765114666

FCT 5 20117FSC 5 0582FST 5 0533


(1)(5)

(38)

4189387

5424

FCT 5 0418FSC 5 0066FST 5 0457


(2)(6)

(59)

(1)(2)

(60)

6023684

3293

7322606

2072

FCT 5 0602FSC 5 0172FST 5 0671

FCT 5 0732FSC 5 0226FST 5 0792

P 005









Mytilus edulis




Idotea balthica



DISCUSSION












































ACKNOWLEDGMENTS


LITERATURE CITED











































































































APPENDIX 2





(1)(5)

(41)

911123

8966

FCT 5 0091FSC 5 0013FST 5 0103


(1)(8)

(78)

61621907194

FCT 5 0062FSC 5 0233FST 5 0281


(1)(3)

(28)

2143766914747

FCT 5 20144FSC 5 0585FST 5 0525


(1)(5)

(51)

2117765114666

FCT 5 20117FSC 5 0582FST 5 0533


(1)(5)

(38)

4189387

5424

FCT 5 0418FSC 5 0066FST 5 0457


(2)(6)

(59)

(1)(2)

(60)

6023684

3293

7322606

2072

FCT 5 0602FSC 5 0172FST 5 0671

FCT 5 0732FSC 5 0226FST 5 0792

P 005









































ACKNOWLEDGMENTS


LITERATURE CITED











































































































APPENDIX 2





(1)(5)

(41)

911123

8966

FCT 5 0091FSC 5 0013FST 5 0103


(1)(8)

(78)

61621907194

FCT 5 0062FSC 5 0233FST 5 0281


(1)(3)

(28)

2143766914747

FCT 5 20144FSC 5 0585FST 5 0525


(1)(5)

(51)

2117765114666

FCT 5 20117FSC 5 0582FST 5 0533


(1)(5)

(38)

4189387

5424

FCT 5 0418FSC 5 0066FST 5 0457


(2)(6)

(59)

(1)(2)

(60)

6023684

3293

7322606

2072

FCT 5 0602FSC 5 0172FST 5 0671

FCT 5 0732FSC 5 0226FST 5 0792

P 005



















































































APPENDIX 2





(1)(5)

(41)

911123

8966

FCT 5 0091FSC 5 0013FST 5 0103


(1)(8)

(78)

61621907194

FCT 5 0062FSC 5 0233FST 5 0281


(1)(3)

(28)

2143766914747

FCT 5 20144FSC 5 0585FST 5 0525


(1)(5)

(51)

2117765114666

FCT 5 20117FSC 5 0582FST 5 0533


(1)(5)

(38)

4189387

5424

FCT 5 0418FSC 5 0066FST 5 0457


(2)(6)

(59)

(1)(2)

(60)

6023684

3293

7322606

2072

FCT 5 0602FSC 5 0172FST 5 0671

FCT 5 0732FSC 5 0226FST 5 0792

P 005

phylogeography and historical ecology of the north atlantic intertidal

Documents