Received 19March 2002 Accepted 15 May 2002 Publishedonline 6August 2002

Hybridizationof the marineseaweeds, Fucus serratus and Fucus evanescens (Heterokontophyta: Phaeophyceae)in a 100-year-oldzone of secondarycontact J. A. Coyer1* ,A.F.Peters 2, G. Hoarau1, W. T. Stam1 and J.L.Olsen 1 1Departmentof Marine Biology, Centre of Ecology and EvolutionaryStudies, University ofGroningen, POBox14, 9750AA Haren,The Netherlands 2Marine O¨ kologie,Institut fu¨rMeereskunde, Du¨sternbrookerWeg 20, D-24105 Kiel, Germany Historically, theintertidal seaweeds Fucus serratus (Fs) and Fucus evanescens (Fe)weresympatric only along thewestern coast of Norway. In themid-1890s, Fe (monoecious)was accidentally introducedinto theOslofjord. Putative hybridization with theendemic Fs (dioecious)was observed in Oslofjordby 1977 andin theKattegat andwestern Baltic Seasby 1998. AtBlushø j, Denmark (Kattegat Sea)putative Fs ´ Fe hybrids werepresent only whendensities of Fe and Fs exceeded14 and2 m 22 ,respectively. All ofthe 58 putative hybrids that werecollected in 1999 weredioecious and intermediate in morphology. Essentially all (57 outof 58) werereproductively mature,but the oogonia possessedfewer and more variably sizedeggs than either parent.Examination ofeach parental speciesand putative hybrids with nuclear,mitochondrial andchloroplast molecular markers conŽrmed theoccurrence of hybridization. Furthermore, all ofthe hybrids possessed Fe-typechloroplasts andmitochondria, indicating that only the Fe egg ´ Fs sperm pair- ing wassuccessful in theŽ eld.The reciprocal crossof Fs egg ´ Fe sperm wasabsent in theŽ eldand signiŽcantly lesssuccessful in laboratory crossings.Asymmetrical hybridization has also beenreported for several speciesof plants andanimals. Keywords: Fucus;hybridization; introducedspecies; seaweed

1. INTRODUCTION placedmany native halophytes asit rapidly expandedits range (Ranwell 1964; Scholten& Rozema 1990). Expandedlevels ofmarine shipping andŽ shing have Intra- andinter-generic hybridization have beendocu- resultedin theintroduction of non-native species to mentedfor many speciesof redand brown algae, butvir- coastal habitats (Carlton 1996). With respectto marine tually all studieshave beenbased on laboratory cultures macroalgae, mostintroductions result in thedisplacement (Rueness1978; Lewis1996 a).Hybridization has been ofexisting species,which in turn,often alter local com- veriŽed with cytological or molecular markers for only two munity dynamicsand trophic interactions.Several intro- hybrids: chromosomecounts for Želd-collected ductionsof macroalgae andtheir negative consequences Pelagophycus ´ Macrocystis (Laminarales) hybrids (Lewis& have beenwell studied: Undaria pinnatiŽda (Hay 1990; Neushul1995) andPCR detectionof species-speciŽ c Floc’h et al. 1995); Sargassummuticum (Ambrose& Nel- ITS1 nrDNA fragments for laboratory produced son1982; Viejo 1997); Codium fragile subsp. tomentosoides Alaria ´ Lessoniopsis (Laminarales) hybrids (Liptack & (Carlton &Scanlon1985); and Caulerpataxifolia Druehl 2000). Unsurprisingly, thereare nostudies of (Acade´mie desScience-Paris 1997). hybrid zonesin marine macroalgae. The potential ofthe exotic speciesto hybridize with Several speciesof Fucus (Fucales)dominate the inter- endemicspecies is an often-overlooked aspect of species tidal andshallow subtidal biomass along northernEuro- introductions.In somecases, hybridization leadsto speci- peanshorelines (Lu ¨ning 1990), wherethey exhibit a ation andthe new species signiŽ cantly impacts local com- considerabledegree of morphological variation both munities.For example, theestuarine cord grass Spartina within andbetween populations. Putative Fucus hybrids alterniora wasaccidentally introducedto the British Isles have beenreported from theŽ eldfor Fucus from easternNorth America in theearly 1800s, andin the vesiculosus ´ Fucus spiralis , Fucus vesiculosus ´ Fucus serratus newhabitat, producedrare andsterile hybrids with the and Fucus ceranoides ´ Fucus spiralis (Sauvageau 1909; local species, Spartinamaritima (Marchant 1967, 1968). Gard1910; Stomps1911; Burrows& Lodge1953; Asa resultof chromosomal doubling in thesterile hybrid, Scott& Hardy 1994). anewspecies emerged ( Spartinaanglica ) (Gray et al. 1991; Fucus serratus (henceforth Fs)ranges from northernPor- Raybould et al. 1991), which subsequentlyinduced major tugal tothe White Sea,Iceland, Nova Scotia,and into geomorphological changesin thetidal habitats anddis- theKattegat andsouthern Baltic Seas(Lu ¨ning 1990). The cold-waterspecies F. evanescens (henceforth Fe) occurs off NewEngland (USA), Greenland, Iceland and northwest- ernNorway (Powell 1957; Rice& Chapman 1985; Lu¨n- *Authorfor correspondence ([email protected]). ing &tom Dieck 1990). Thus, Fs and Fe are sympatric

Proc.R. Soc.Lond. B (2002) 269, 1829–1834 1829 Ó 2002 TheRoyal Society DOI10.1098/ rspb.2002.2093 1830J. A.Coyer andothers Hybridizationin Fucus along portionsof Iceland and the northwestern Norweg- ian coast. Fe wasaccidentally introducedinto the Oslofjord (easternNorway) in themid-1890s andthen expanded south,appearing in theSkagerrak Seain 1924, thecentral Kattegat Sea(west coast of Sweden) in 1933, andin the Kiel Bight (westernBaltic) by 1992 (seeSchueller &Pet- ers1994). Putative Fs and Fe hybrids reaching fertility in the Ž eldinitially werereported in Oslofjordin 1977 (Lein 1984; Rice& Chapman 1985) andmore recentlyat sev- eral locationsin theKattegat towesternBaltic Seasfrom 1998 to2000 (J.A.Coyerand A. F.Peters,unpublished data).Therefore, any hybridization betweenthe native Fs andintroduced Fe in theareas occurredwithin thelast 100 years. In thepresent study we usemolecular markers toverify theoccurrence of natural hybridization between Fs and Fe. Using Ž eld-collectedspecimens, we examined DNA polymorphism in Fs, Fe andputative Fs ´ Fe hybrids with Ž ve microsatellite loci (nuclear),as well asITS1 (nuclear), rubisco(chloroplast) and nad11 dehydrogenase (mitochondria) genes.We believe that ourstudy is the Ž rst toapply microsatellite genotyping andsingle-strand con- formation polymorphism (SSCP)towards veri Ž cation of hybridization in Ž eldpopulations ofmarine macroalgae.

2. MATERIALAND METHODS The dioecious Fs ischaracterizedby broad thalliwith serrated marginsand adistinctmidrib, whereas the hermaphroditic Fe (allconceptacles on asingleindividual contain both antheridia and oogonia)have muchnarrower thalli with no serrationsand an indistinctmidrib ( Ž gure1). The chromosomenumber for eachspecies is 32(or34) (Lewis 1996 b).Both speciesare peren- nial.In the KielBight, Fe isreproductiveduring late winter and spring, whereas Fs isreproductiveall year long (Schueller& Pet- ers1994). Eggs and spermof the closelyrelated F.vesiculosus typicallydisperse only 0.5 and 2m,respectively, from the par- ents (Lifvergren1996). Long-distance transport may occur, however,as detachedindividuals may driftto other areas. Putative Fs ´ Fe hybrids wereobserved in April 1998, 1999 and 2000at Blush øj,near Elsega Ê rde,Denmark (56 °109 N, 10°439 E).InApril 2000, the spatial distributionsof Fs, Fe and putative Fs ´ Fe hybrids weredetermined with 0.5m 2 quadrats Figure 1. Photographs of ( a) Fs, (b) Fe and (c) a Fs ´ Fe sequentiallyplaced along threetransects (19,24, 26 minlength; hybrid collected from Blushoj, Denmarkin April 1999. The 0.5m inwidth) traversing the entire Fucus beltfrom 0 to ca. size of thepencil in eachphotograph is 14.5 cm. 22mindepth. Allpost-recruit Fucus individuals(more than 10cmin length) within the quadrats werecollected, counted ted.The egg diameterswere measured with alight microscope and weighed.Very few recruitswere present. Tissue samples for at ´ 100 magniŽ cationafter theirrelease from the oogonia. DNA extraction werecollected from 72 Fs, 70 Fe and 58 puta- The spermbehaviour was observedat low magni Ž cation dur- tivehybrids and processedas describedin Coyer et al. (2002a). ing the crossingexperiments. The formationof spermclouds Laboratory crossesfor allcombinations of maleand female concentratedaround the eggs was interpretedas evidencefor Fs, Fe and Fs ´ Fe hybrids wereconducted as describedin Coyer chemotaxis,that has beenwell studied in Fucus spp. (Mu¨ller & et al. (2002b).Negative (eggs only,no sperm added) and posi- Seferiadis1977). In addition, the detectionof motilesperm in tive(conspeci Ž ccrosses)controls were included for eachcross. light microscopicpreparations at ´ 400 or ´ 1000 magniŽ cation Mitoticdivisions were present in all fertilized eggs, but not in was taken as beingevidence for the presenceof competent negativecontrols, and parthenogenesis doesnot occurin Fucus. sperm.Chemotaxis orsperm motility was not quanti Ž ed. Reproductivesuccess of the crosseswas de Ž nedas development DNA fromfresh ordried tissue was extracted as describedin to four-week-oldembryos. Coyer et al. (2002a).The following Ž vemicrosatellite loci were The numbersof eggs peroogonium and the sizesof eggs were genotyped: FsA198,FsB113, FsB128, FsD39 and FsF4(Coyer quantiŽ edin one individual of each Fe, female Fs and female et al. 2002a).ITS1 allelesand haplotypes ofchloroplast (rubisco hybrid. Batches of oogonia werecollected after releasefrom the spacer)and mitochondria( nad11)were determined by SSCP receptaclesand the numbersof cellsper oogonium were coun- analysis (Coyer et al. 2002b).

Proc.R. Soc.Lond. B (2002) Hybridizationin FucusJ. A.Coyer andothers 1831

3 50 2

40 ) %

7 1 . 4 (

2

) s . i

e 30 0 x . a s (

y t i

s _

n 1 e 20 d

_2 _2 _1 0 1 2 10 axis 1 (19.8 %)

Figure 3. Two-dimensional FCAof all Ž eld individuals. The 0 _ _ _ _ _ two dimensions depicted accounted for 24.5% of the 0 4 5 9 10 14 15 19 20 24 variability. Thelaboratory crosses were included in theFCA distance from shore (m) assupplementary elements for visualization only and were not used to compute thefactor axes. Black circles, Fs; white Figure 2. Density( 1s.e.) of Fs (light grey bars), Fe (dark circles, Fe;stippled squares, Ž eld hybrids; stippled diamonds, grey bars) and putative Fs ´ Fe hybrids (blackbars) at laboratory hybrids. Blushøj, Denmark. Themeans are derived from 0.5 m 2 quadrats along three transects in April 2000. displayed vigorous movementwith chemotaxis towards the eggs. Microsatellite andSSCP analysis con Ž rmed that the58 The ITS1 and Ž vemicrosatellite loci were analysed as active putative Fs ´ Fe hybrids collectedat Blush øjin 2000 were elementsin afactorialcorrespondence analysis (FCA)(G enetix truehybrids ( Ž gure 3, table 1). Additionally, all the Ž eld 4.02;Belkhir et al. 2001),a multivariateanalysis that projects hybrids possessed Fe-typechloroplasts andmitochondria, allindividuals in a spacede Ž nedby the components.In FCA indicating that only the Fe egg ´ Fs sperm pairing wassuc- (Benze´cri1973), each individual can be represented using each cessfulin the Ž eld(table 1). In addition tobeing absent alleleas an independentvariable (She et al. 1987).Only individ- in the Ž eld,the reciprocal cross, Fs egg ´ Fe sperm, was ualswith completegeneotypes wereused and allelespresent less signiŽ cantly lesssuccessful in thelaboratory crossings than Ž vetimes in the entiredataset wereexcluded from the (Ž gure 4). analysis (Kotulas1989). 4. DISCUSSION 3. RESULTS Ahybrid zonebetween Fs and Fe has existedfor ca. Of the58 putative hybrids collectedat Blush øj in 2000, 100 years in theSkagerrak, Kattegat andwestern Baltic all weredioecious and all butone were reproductively Seasand therefore, represents the early stagesof asecond- mature.The receptaclesof 30 randomly selectedindivid- ary contactbetween previously allopatric populations of ualsrevealed 16 femalesand 14 males.The average wet thetwo species. Furthermore, thebarrier togene weights(s.d.) of the Fucus individuals collectedat Blush øj exchange between Fs and Fe at Blushøjisasymmetrical, were: Fs,135.7 g(66.3); Fe,51.9 g(32.0); putative asonly hybrids resulting from pairings of Fe eggs ´ Fs hybrids, 57.2 g(28.1). Fs wasdistributed between the sperm wereobserved. Asymmetrical fertilization success bimodal peaksof Fe,whereasputative Fs ´ Fe hybrids among hybrids isa phenomenonthat has beenreported werepresent only when Fe and Fs formedmixed stands in several speciesof plants andanimals (Kaneshiro1976; (Ž gure 2). Noputative hybrids werepresent if theden- Levin1978; Arnold et al. 1996; Coyne& Orr 1998; Funk sities of Fe and Fs wereless than 14 and2 m 22 , respect- 1998; TifŽ n et al. 2001). ively. Asymmetrical fertilization is usually attributable tosex- Eachoogonium of Fs and Fe containedeight eggs that ual selection(Wilson 1979; Lyons et al.1989; Arnold et weresimilar in size,as is thenorm for Fucus (Fritch 1945). al. 1996). Morespeci Ž cally, theasymmetrical mate choice The egg diameter ( n,mean,s.d.) in Fe (40, 98.7 m m, 7.7) hypothesis(Kaneshiro 1976) predictsthat femalesof a was signiŽ cantly (ANOVA, p , 0.0001) larger than in Fs newlyevolved or derivedtaxon readily will acceptmales of (40, 70.3 m m,4.7). The female Fs ´ Fe hybrids weredif- theancestral taxon becauseall ofthe male characteristics ferent,as their oogonia containedvariably sized(50, required by thefemales are present.Derived males,how- 78.7 m m,25.7) andfewer eggs ( n = 76, mean = 2.3, ever,will beunable to fertilize femalesof an ancestral s.d. = 1.6) than thegeneric norm.Nevertheless, eggs taxon becausethese males have ‘lost’ necessarystimuli in releasedfrom theoogonia ofhybrids chemotactically themate recognition system.Thus, females from aderived attracted sperm,as did eggs releasedfrom theoogonia of taxon will mate with males from anancestral taxon,but Fe and Fs.The antheridia andsperm ofmale Fe, Fs and notvice versa.The hypothesishas beensupported by lab- Fs ´ Fe hybrids, weremicroscopically similar andall sperm oratory and Ž eldstudies of Drosophila and Ž eldstudies of

Proc.R. Soc.Lond. B (2002) 1832J. A.Coyer andothers Hybridizationin Fucus

Table1. Genotypes (nuclear) and alleles (chloroplast, mitochondria) for Fs, Fe and Fs ´ Fe hybrids from the Ž eld (HF) and crossed in thelaboratory (L). (Abbreviations: f, female; m, male.)

nuclear (ITS1) chloroplast (rubisco) mitochondrial ( nad11)

a a a b b b g g A B 1 2

Fs 0 0 71 1 0 72 0 72 HF 0 58 0 0 58 0 58 0 Fe 70 0 0 0 70 0 70 0 HF f ´ HF m (L) 0 1 1 0 2 0 2 0 HF f ´ Fs m (L) 0 1 2 0 3 0 3 0 HF f ´ Fe m (L) 0 4 0 0 4 0 4 0 Fs f ´ Fe m (L) 0 3 0 0 0 3 0 3 Fe f ´ Fs m (L) 0 4 0 0 4 0 4 0

a 0.6 a a

a s s ) e .

c a e . c s u

, s 0.4

n e o v i i t t r c o u p d o o r r p ( p e r 0.2

b b b b

0.0 Ef×Em Ef×Hm Ef×Sm Hf×Em Hf×Hm Hf×Hm Sf×Em Sf×Hm Sf ×Sm 16 8 16 8 12 12 12 12 20 combinations of laboratory crosses

Figure 4. Reproductive success(the presence of four-week-old embryos) for all combinations of laboratory crosses between femaleand male Fs (Sf, Sm), Fe (Ef, Em) and Fs ´ Fe hybrids (Hf, Hm). Thesample sizes are indicated below thebars; the crosses identi Ž ed withthe same letter (aor b)are not signi Ž cantly different (proportion dataarcsine-square-root transformed; one-factor ANOVA,Tukey –Kramer post-hoc test, p , 0.05).

sticklebacks( Gasterosteus spp.)(McPhail 1969; Ohta adhesion,as Fs eggs readily attracted Fe sperm in thelab- 1978; Powell 1978; Arita &Kaneshiro1979). Although oratory, consistentwith observationsof others who noted sexual selectionis possiblein plants (Wilson1979; Lyons extensivecross-speci Ž cattraction becauseall Fucus spp. et al. 1989; Arnold1994), theasymmetrical mate choice eggs possessthe same sperm-attracting pheromone hypothesishas notbeen tested (Tif Ž n et al. 2001). (Mu¨ller &Seferiadis1977; Maier &Mu¨ller 1986). Asymmetrical hybridization in Fucus (passive mate Alternatively, twoother hypothesesmay explain the choice)is analogous tothe asymmetrical mate choice asymmetry ofhybrid successin Fucus from the Ž eld. First, hypothesisof Kaneshiro (1976) (active anddirect mate anasynchronous release of Fs eggs and Fe sperm (theher- choice).A recentITS-based phylogeny demonstratedthat maphroditic species),relative toa synchronousrelease of Fs is theancestral Fucus (Serra˜o et al. 1999). Field hybrids Fe eggs and Fs sperm,may prevent Fs egg–Fe sperm at Blushøjconsistedonly of Fe egg (derivedspecies) ´ Fs hybrids. Second,the total numberof Fe sperm in the Ž eld sperm (ancestral species);the reciprocal crosswas not may bemuch less than Fs sperm,simply because Fe is a foundand was signi Ž cantly lesssuccessful in thelabora- hermaphrodite andhermaphroditic speciesof Fucus pro- tory (Ž gure 4). The lowsuccess rate ofa Fs egg–Fe sperm duce signiŽ cantly fewersperms per egg (40 :1) than do unionmay bedue to events occurring after sperm dioeciousspecies (400 :1) (Vernet& Harper 1980).

Proc.R. Soc.Lond. B (2002) Hybridizationin FucusJ. A.Coyer andothers 1833

Fewer Fe sperm relative to Fs sperm logically shouldresult becauseof limited gene  ow. Thus, Fe individuals recently in smaller numbersof Fe sperm–Fs egg fertilizations. introducedto the Kattegat Seawould encounter no bar- However,the asymmetry ofhybridization successdemon- rier tohybridization with local populations of Fs. Barriers stratedby ourlaboratory crossingscannot be explained by tohybridization may beweak or non-existentin Fe, how- either ofthese Ž eld-basedhypotheses. ever,as the founding individuals probably came from an The ecological fateof Fs ´ Fe hybrids at Blush øj and area oflong-standing sympatry with Fs. elsewherein theSkagerrak, Kattegat andBaltic Seas The discoveryof Fs ´ Fe hybrids in ayoung zoneof sec- remains tobe determined, but three possibilities exist. ondary contactprovides auniqueopportunity tounder-

First, thehybrid population may consistonly ofF 1 indi- standthe importance ofhybridization in marine algae. viduals becauseF 2 sandbackcrosses are less Ž t.This situ- Whether thehybrids eventually disappear becausethey are ation isan evolutionary dead-endfor thehybrids and less Ž tor becauseisolation mechanismshave been shouldlead tothe reinforcement of pre-zygotic isolation reinforced,or evolve intoa newspecies with thepotential mechanisms.Second, hybrid introgression (transferof tocompete with either or both parent species,remain to genesbetween parental species)may occur,leading tothe bedetermined. The Fs ´ Fe hybrids in theSkagerrak, Kat- transferof adaptations or speciesmergers, both ofwhich tegat andBaltic Seaswarrant further studyby tracking are well documentedin animals andplants (Rieseberg all genotypesover multiple generations,and conducting 1998). Third, thehybrids may evolve intoa newspecies. detailedecological anddemographic investigations of It ishighly likely that all Ž eldhybrids at Blush øj were hybrids andparents.

F1 individuals, rather than amixture oflater generation hybrids andbackcrosses, and that thehybrids wereless Ž t. Theauthors thankJ. Veldsink for laboratory assistance, E. Laboratory backcrossesproduced individuals with ITS1 Stommel for Ž eld assistanceand J.Rueness for discussions. alleles from oneor both parents,whereas the Ž eld hybrids Theresearch wassupported bythe BIOBASE Project funded under EUMASTIII, control no. PL97-1267. only possessedan ITS1 allele from both parents(table 1). The striking absenceof overlap betweenthe Fs, Fe and Fs ´ Fe cloudsin theFCA also indicatesa lack ofintro- REFERENCES gressiondue to backcrossing ( Ž gure 3). This wassurpris- Acade´mie des Science-Paris 1997 Dynamique d’espe `ces marines ing becausethe fertility ofhybrid eggs with hybrid or invasives: application a` l’expansion de Caulerpa taxifolia en parental sperm (backcross)in thelaboratory wasvery low, Me´diterrane´e .Paris: Tecand Doc Lavoisier. butthe fertility ofhybrid sperm washigh ( Ž gure 4). The Ambrose, R.F.&Nelson, B.V.1982 Inhibition of giant kelp absence of F2 individuals in the Ž eldmay bea resultof recruitment byan introduced brown alga. Bot. Mar. 25, geneticincompatibilities (endogenousselection) expressed 265–267. in F2 individuals, aswell asreduced viability in thehabitat Arita, L.H.&Kaneshiro, K.Y.1979 Ethological isolation (exogenousselection). between two stocksof Drosophila adiastola Hardy. Proc. Despitethe apparent low Ž tness of Fs ´ Fe hybrids, the Hawaii Ent. Soc. 23, 31–34. evolutionof anewspecies may still occur.It isimportant Arnold, M.L.&Hodges, S.A.1995 Are natural hybrids Ž t or torealize that limited fertility in presenthybrid gener- unŽ trelative to their parents? TrendsEcol. Evol. 10, 67–71. Arnold, S.J.1994 Bateman principles and themeasurement ationsdoes not necessarily prohibit aprogression tonew of sexual selection in plantsand animals. Am. Nat. 144, evolutionary lineages (Arnold& Hodges1995). For S126–S149. example, genotypic classesof hybrids have beendocu- Arnold, S.J., Verrell, P.A.&Tilley, S.G.1996 Theevolution mented,each of which canpossess equivalent or higher of asymmetryin sexual isolation: amodel and atestcase. levels of Ž tnessrelative totheir parents(Arnold & Hodges Evolution 50, 1024–1033. 1995) andlater hybrid generationscan be as Ž t or more Belkhir, K., Borsa, P., Goudet, J., Chikhi, L.&Bonhomme, Ž tthan either parent (Rieseberg 1998). F. 2001 GENETIX, logiciel sousWindows Ô pourla genetique Fs ´ Fe hybrids werenot found near Tr ømso, north- des populations .Laboratoire Genome et populations, Univer- westernNorway (Lein1984), where Fs and Fe have been sityof Montpellier. sympatric (= primary contactzone) for hundredsor per- Benze´cri, J.P.1973 L’Analyse des Donne´es:T. 2, l’Analyse des haps thousandsof years. How is it then,that in a100- correspondances .Paris: Dunod. Burrows, E.M.&Lodge, S.M.1953 Culture of Fucus hybrids. year-old zoneof secondary contact, Fs ´ Fe hybrids have Nature 172, 1009–1010. persistedfor several years,attained aweight intermediate Carlton, J.T.1996 Marine bioinvasions: thealteration of mar- toboth parents,and reached a high level ofreproductive ine ecosystemsby nonindigenous species. Oceanography 9, maturity? The answermay bethat Fs populations in Eur- 36–43. opeexhibit signi Ž cantgenetic structure. In astudyof 35 Carlton, J.T.&Scanlon, J.1985 Progression and dispersal of northernEuropean populations of Fs,estimatesof pair- an introduced algal: Codiumfragile spp. tomentosoides wise FS T -valuesderived from sevenmicrosatellite loci (Chlorophyta) on theAtlantic coast of north America. Bot. revealed signi Ž cantdifferentiation (implying lowgene Mar. 82, 155–165.  ow)among all populations separatedby > 2 km (Coyer Coyne, J.A.&Orr, H.A.1998 Theevolutionary genetics of et al.2003). For example, values betweenBergen (western speciation. Phil. Trans.R. Soc. Lond. B 353, 287–305. (DOI 10.1098/rstb.1998.0210.) Norway) andeach of 19 populations within theSkagerrak Coyer, J.A., Veldsink, J.H., Stam,W. T.&Olsen, J.L.2002 a andKattegat Seas,including Blush øj,ranged from 0.224 Characterization of microsatellite loci in themarine to 0.351. If Fs evolved barriers tohybridization with Fe rockweeds, Fucus serratus and F. evanescens along thenorthwestern coast of Norway (e.g.reinforce- (Heterokontophyta; Fucaceae). Mol.Ecol. Notes 2, 35–37. mentof isolation in aprimary zoneof contact), the bar- Coyer, J.A., Peters, A.F., Hoarau, G., Stam,W. T.&Olsen, riers are notlikely tohave spreadto other Fs populations J. L. 2002b Inheritance patterns of ITS1, chloroplasts, and

Proc.R. Soc.Lond. B (2002) 1834J. A.Coyer andothers Hybridizationin Fucus

mitochondria in arti Ž cial hybrids of themarine rockweeds, Marchant, C.J.1967 Evolution of Spartina (Gramineae). I. Fucus serratus and F. evanescens (Heterokontophyta; Thehistory and morphology of thegenus in Britain. J. Linn. Fucaceae). Eur.J. Phycol. 37, 173–178. Soc. (Bot.) 60, 1–24. Coyer, J.A., Peters, A.F., Stam,W. T.&Olsen, J.L.2003 Marchant, C.J.1968 Evolution of Spartina (Gramineae). II. Post-Ice Age recolonization and differentiation of Fucus Chromosomes, basicrelationships and theproblem of serratus L.(Fucaceae: Phacophyta)populations in northern S. ´ townsendii agg. J.Linn.Soc. (Bot.) 60, 381–409. Europe. Mol. Ecol.(Submitted.) Mu¨ller, D.G.&Seferiadis, K.1977 Speci Ž city of sexual Floc’h,J.Y., Pajot, R.&Mouret, V.1995 Propagation of the chemotaxis in Fucus serratus and Fucus vesiculosus Japanese brown alga Undaria pinnatiŽda (Harvey) Suringar (Phaeophyeae). Z.Panzenphysiol. 84, 84–95. along theFrench Atlantic coasts. Crypto.Algol. 16, 157–158. Ohta,A. T.1978 Ethological isolation and phylogeny in the Fritch, F.E.1945 The structure and reproduction ofthe algae , grimshawispecies complex of Hawaiian Drosophila. Evolution vol. 2. Cambridge University Press. 32, 485–492. Funk, D.J.1998 Isolating arole for natural selection in speci- Powell, H.T.1957 Studies in thegenus Fucus L. I. Fucus ation: host adaptation and sexual isolation in Neochlamisus distichus L.emend. Powell. J.Mar.Biol. Assoc.UK 36, bebbianae leafbeetles. Evolution 52, 1744–1759. 407–432. Gard, M.1910 Sur un hybride es Fucus platycarpus et Fucus Powell, J.R.1978 TheFounder –Flush speciation theory: an ceranoides. C.R.Acad.Sci. Paris 151, 888–890. experimental approach. Evolution 32, 465–474. Gray, A.J., Marshall, D.F.&Raybould, A.F.1991 Acentury Ranwell, D.S.1964 Spartina saltmarshesin southern England. of evolution in Spartina anglica . Adv.Ecol. Res. 21, 1–62. II. Rate and seasonal pattern of sediment accretion. J. Ecol. Hay, C.H.1990 Thedispersal of sporophytes of Undaria pin- 52, 79–94. natiŽ da bycoastal shipping in NewZealand and implications Raybould, A.F., Gray, A.J., Lawrence, M.J.&Marshall, for further dispersal of Undaria in France. Br.Phycol. J. 25, D.F.1991 Theevolution of Spartina anglica C.E.Hubbard 301–313. (Gramineae): origin and genetic variability. Biol. J.Linn. Kaneshiro, K.Y.1976 Ethological isolation and phylogeny in Soc. 42, 111–126. theplanitibia subgroup of Hawaiian Drosophila. Evolution 30, Rice, E.L.&Chapman,A. R.O.1985 Anumerical taxonomic 740–745. study of Fucus distichus L.emend. Powell (Phaeophyta). J. Kotulas, G.1989 Differentiation geographique et structure Mar.Biol. Assoc.UK 65, 433–459. genetique des populations de Solea vulgaris .PhDthesis, Uni- Rieseberg, L.1998 Molecular ecology of hybridization. In versity of Montpellier II, France. Advances in molecular ecology (ed. R.Carvalho), pp. 243 –265. Lein, T.E.1984 Hybrider mellom sagtang( Fucus serratus L .) Amsterdam: IOSPress. og gjelvtang( Fucus distichus subsp. edentatus (Pyl.) Powell) Rueness, J.1978 Hybridization in red algae. In Modern iindre Oslofjord. Blyttia 42, 71–77. approaches to the taxonomy ofred and brownalgae (ed. D. E. Levin, D.A.1978 Theorigin of isolation mechanismsin G.Irvine &J.H.Price), pp. 247 –262. London: Academic.  owering plants. Evol. Biol. 11, 185–315. Sauvageau, C.1909 Sur l ’hybride des Fucus vesiculosus et Fucus Lewis, R.J.1996 a Hybridization of brown algae: compatibility serratus. C.R.Soc. Biol. Paris 67, 832–833. and speciation. In Cytology,genetics and molecular biology of Scholten, M.&Rozema, J.1990 Thecompetitive ability of algae (ed. B.R.Chaudhary &S.F.Agrawal), pp. 275 –289. Spartina anglica on Dutchsalt marshes. In Spartina—a Amsterdam: SPBAcademic. research review (ed. A.J.Fray &P.Benham), pp. 39 –47. Lewis, R.J.1996 b Chromosomes of thebrown algae. [Phyco- London: NERCInstitute for Terrestrial Ecology. logical Reviews16]. Phycologia 35, 19–40. Schueller, G.H.&Peters, A.F.1994 Arrival of Fucus evan- Lewis, R.J.&Neushul, M.1995 Intergeneric hybridization escens (Phaeophyceae) in Kiel Bight (Western Baltic). Bot. among Ž vegeneral of thefamily Lessoniaceae Mar. 37, 471–477. (Phaeophyceae) and evidence for polyploidy in afertile Scott, G.W.&Hardy, F.G.1994 Observations of theoccur- Pelagophycus ´ Macrocystis hybrid. J. Phycol. 31, 1012–1017. rence of hybrids between two sympatricspecies of fucoid Lifvergren, T.1996 Dispersalcapacities of Fucus vesiculosus algae. Crypto.Algol. 15, 297–305. gametesand recolonisation basedon natural regeneration. Serra˜o, E.A., Alice, L.A.&Brawley, S.H.1999 Evolution of MScthesis, University of Kalmar, Sweden. theFucaceae (Phaeophyceae) inferred from nrDNA –ITS. J. Liptack,M. K.&Druehl, L.D.2000 Molecular evidence for Phycol. 35, 382–394. an interfamilial laminarialean cross. Eur.J. Phycol. 35, She, X., Autem, M.,Kotulas, G., Pasteur, N.&Bonhomme, 135–142. F.1987 Multivariate analysisof genetic exchanges between Lu¨ning, K.1990 Seaweeds: their environment, biogeography and Solea aegyptiaca and Solea senegalensis (Teleosts, Solediae). ecophysiology .NewYork: Wiley. Biol. J.Linn.Soc. 32, 357–371. Lu¨ning, K.&tom Dieck, I.1990 Thedistribution and evol- Stomps, T.J.1911 Etudes topographiues dur la variabilite de ution of theLaminariales: North Paci Ž c–Atlantic relation- Fucus vesiculosus L.platycarpusThur., et ceranoides L. Recueil ships. In Evolutionary biogeography ofthe marine algae ofthe l’Inst. Bot. ‘LeoErrera’ , Bruxelles 8, 326–377. NorthAtlantic (ed. D.J.Garbary &G.R.South), pp. 187 – TifŽ n, P., Olson, M.S.&Moyle, L.C.2001 Asymmetrical 204. Berlin: Springer. crossing barriers in angiosperms. Proc.R. Soc. Lond. B 268, Lyons, E.E., Waser, N.M., Price, M.V., Antonovics, J.& 861–867. (DOI10.1098/rspb.2000.1578.) Motten, A.F.1989 Sources of variation in plant repro- Vernet, P.&Harper, J.L.1980 Thecost of sexin seaweeds. ductive successand implications for concepts of sexual selec- Biol. J.Linn.Soc. 13, 129–138. tion. Am. Nat. 134, 409–433. Viejo, R.M.1997 Theeffects of colonization by Sargassum McPhail, J.D.1969 Predation and theevolution of astickle- muticum on tidepool macroalgal assemblages. J.Mar.Biol. back (Gasterosteus ). J.Fish Res.Bd Can. 26, 3183–3208. Assoc. UK 77, 325–340. Maier, I.&Mu¨ller, D.G.1986 Sexual pheromones in algae. Wilson, M.F.1979 Sexual selection in plants. Am. Nat. 113, Biol. Bull. 170, 145–175. 777–790.

Proc.R. Soc.Lond. B (2002)