.
Potentialimpactofanexoticmammalonrocky intertidalcommunitiesofnorthwesternSpain
MiguelDelibes1,∗, MiguelClavero1,2,Jose´Prenda2,Mar´ıadelCarmenBla´zquez3PabloFerreras4
1DepartmentofAppliedBiology,Estacio´nBiolo´gicadeDon˜ana,CSIC,Avda.Mar´ıaLuisa,s/n,41013Sevilla,
Spain;2DepartmentofBiolog´ıaAmbientalySaludPu´blica,UniversidaddeHuelva,C.U.ElCarmen,Avda
Andaluc´ıa,21071,Huelva,Spain;3CentrodeInvestigacionesBiolo´gicasdelNoroeste(CIBNOR),P.O. Box 128,LaPaz,BajaCaliforniaSur,23000Mexico;4InstitutodeInvestigacio´nenRecursosCinege´ticos, UCLM-CSIC-JCCM,P.O.Box535,13080CiudadReal,Spain;∗Address forcorrespondence(e-mail: ;fax:+34-95-4621125)
Keywords:alienmammal,Americanmink,Blennidae,coast,crab,invasion,Mustelavison,
Pachygrapsusmarmoratus,rockyintertidal
Abstract
Beingtheinterfaceofseaandland,thecoastcanbeinvadedbyintroducedspeciescomingfromeitherofthesetwo worlds.Recentreviewsofcoastalinvasionsemphasizethehuman-mediated transportofnon-indigenousmarine plantsandinvertebrates,forgettingthepotentialroleofinvadersofterrestrialorigin.Bystudyingthedietof theintroducedAmericanmink(Mustelavison)onarockyshoreofsouthwesternEurope,wedrawattentionto thepotentialimpactonintertidalcommunitiesofexoticspeciescomingfrominland.Weanalysed199minkfaeces collectedinAugust1997andAugust1999inBaiona,acoastalandurbanareaofnorthernSpainrecentlyinvadedby minks.Thedietofthespecieswasbasedalmostexclusivelyoncrabs(45.4%ofindividualprey)andfish(53.3%). Mostcrabsweremarbledcrabs(Pachygrapsusmarmoratus)andmostfishwereadultblennies(Coryphoblennius galeritaandLipophryspholis).Givenitsenergyrequirements(about1250kJ/day),asingleminkwillconsume duringthemonthofAugustapproximately945blennies and496crabs.Althoughwelackaccurate dataonmink abundance,acautiousestimation(4mink/kmbeforedispersal),supportedbyfieldobservations,suggeststhat predation inAugustmayreach3780blenniesand1984crabsperkmofshoreline.Thispredation pressurecould affectthenumbersofblenniesand(lessprobably)crabs,indirectlybenefitingthepopulationsoftheirprey,thatis, sessileinvertebratesandsnails.Morefieldresearchisneeded,butourresultssuggestthatanexoticnon-marinetop predatorsuchastheAmericanminkcouldaffectintertidalcommunitiesinEurasia.
Introduction
Biologicalinvasionsareasignificantcomponentof humancausedglobalchange(Vitouseketal.1997) andamaincauseofthecurrentlossofbiodiversity (Diamond 1989).Exotic,introduced speciesarecon- sideredthesecondmostimportantcauseofextinctions, immediately behindhabitatdestruction(Macdonald andThom2001).For example,morethan18%of941
endangeredvertebratetaxaarethreatenedinsomeway byexoticspecies(Macdonald etal.1989),bothin terrestrialandinaquaticecosystems.Moreover,extinc- tionisonlythemostdramaticpotentialresultofpreda- tionorcompetitionbyintroducedspecies,whichmore oftenmodifythedistribution,abundance,behaviour and evolutionary trajectories of native flora and fauna(Simberloff1981)and consequentlythespecies compositionanddynamicsofnativecommunities.
Historically,thestudyofcoastalinvasions has receivedlessattentionthansimilarstudiesofterrestrial andfreshwater ecosystems.However,coastalmarine habitatsareamongthemostheavilyinvadedsystemson Earth,withmanyecologicalandevolutionaryconseq- uences(Grosholz2002).Usually,reviewsofcoastal invasions (e.g.Ruizetal.2000;Grosholz 2002)take into accountalmostexclusivelyplantsand invertebra- tesasinvadersandconsiderthehuman-mediatedtrans- portofnon-indigenousspeciesbyshipping(resulting fromballastwaterorhullfouling) andtheintentional orunintentional resultofaquacultureandfisheriesas thedominantmechanismsproducinginvasions.How- ever, thecoastrepresentstheinterfaceofland and sea andcoastalinvaderscouldbealsosemi-aquaticspecies comingfrominland.Theprimepurposeofthispaperis tocontributetotheunderstandingofcoastalinvasions bydrawingattentiontothepotentialecologicalimpact onintertidalcommunitiesofsomeexoticspeciesfrom mainlyinlandorigin.Wewilluseasanexample the caseofthealienAmericanmink(Mustelavison)inthe rockyshoreofnorthwesternSpain.
TheAmericanminkisamustelidofabout1kgbody weightwhosenativerangeincludes Canadaandthe UnitedStatesexcepttheextremenorthandthearid southwest.Sincethelate19thcenturythespecieshas beenbredatfurfarmsinitsnativecountriesandin the1920sthefirst Americanminkswerebroughtto Europeforcommercialfarming.Duetoeitherdeliber- atereleases orescapes fromfurfarms,theAmerican minkisnowwidespread inmostEuropeanandsome AsianandSouthAmericancountries(Dunstone1993). Initsnativerangeandwhereverintroduced,thespecies occupies manymesicandwethabitats,fromriversto marshes,lakesand coasts. Atpresent,itisreportedin
mostEuropeanAtlanticcoasts northof48◦Nlatitude
andinsomeMediterraneancoasts(Mitchell-Jonesetal.
1999).Itspresenceisincreasingincoastalareasof northernSpain(PalomoandGisbert2002).
TheintroducedAmericanminkisanopportunis- ticpredator(Dunstone 1993).Wherever itbecame established inEurasiaandSouthAmericaithasbeen recognisedasaseriousthreattotheconservation of nativespecies,bothtoitscompetitors (e.g.European mink,Mustelalutreola,Patagonianhuill´ın,anendemic otter, Lutraprovocax)and to its prey (e.g. water voles,Arvicolaterrestris,riverineandseabirds,white- clawedcrayfish,Austropotamobiuspallipes)(Ferreras andMacdonald1999;MacdonaldandStrachan1999; MacdonaldandThom2001;Nordstro¨metal.2003).
Wesuspectthatpredationbyshore-livingAmerican minkscouldinfluencethedistributionandabundance ofitsintertidalpreyandhenceforththecomposition ofintertidalcommunitiesinEurope.Bydescribingthe summer dietoftheminkinarecently invaded rocky coastalareaofnorthern Spain,wewilltrytoesti- matetheabsolutenumberoffishandcrabsthatminks removedaily.Thiscouldsuggestapotentialimpactof minkpredationontheinvadedintertidalcommunities.
Studyareaandmethods
Thedietoftheminkwasestimatedthroughtheanalysis ofitsfaeces.WecollectedminkfaecesinAugust1997 and1999alonga2kmstretchofexposedAtlanticrocky shoreinthetownofBaiona(NWSpain;42◦10tN,
8◦50tE),arelativelysmallcitywhichreceivesthou- sandsofvisitors duringthesummer.Theintertidal fringeisanalmostcontinuous(therearetwosandy beachesabout100mlong)rockyplatformbetween
20mandmorethan100mwide,includinglargenatural boulders andfrequent poolsduringlowwateronthe tidalrange.Theupperrockyborderprovidesnumerous densitesforthemink.Theshoreisseparatedfromthe hinterlandbyroadsandurbanwalls.Theclimateis temperatewithMediterraneaninfluence;annualaver- agetemperatureandrainfall(1931–1980)are15◦C and1337mm,andinAugustaveragetemperatureis
21.6◦Candrainfall20.3mm.
Minkswerecommoninthearea.Atleasttwodiffer- entfamilygroupswiththreeandfourindividuals(pre- sumably onefemalewithalmostfullygrownyoung) andsomesolitaryindividualwererepeatedlyobserved inAugust1999.Otters(Lutralutra)arelackinginthis portionofthecoast.
Wecollected199minkfaeces(102in1997and
97in1999),probablycorrespondingto6–10differ- entindividuals eachsummer.Dietanalysesfollowed standardmethodology(Beja1997).Preyremainswere identified usingpublishedkeys(Webb1980;Rosello´
1986)andourowncollectionforcomparison.Each identifiedpreyclassinascatwasconsideredasan
‘occurrence’,beingtherelativefrequencyofoccur- rence(RFO)thepercentage ofthetotalnumberof occurrencescorrespondingtoacertainpreyclass.Diet diversitywasquantifiedusingtheShannon–Weaver
index(Ht)withRFOdataofthegeneralpreyitems
(fish,crabs,birds,insectsandamphibians)toallow appropriatecomparisonswithotherstudies.
Theminimumnumberofdifferentindividuals of eachpreyclasswasestimated fromthenumberand position(left–right) ofdiagnostichardparts(mainly mouthbonesforfishandthirdmaxillipedsforcrabs) (Beja1996),whichweremeasured withacalliperto thenearest0.1mm.Inthosecaseswhendiagnostic piecesdidnotappear,remains ofacertainpreyitem wereconsideredtobelongtoasingleindividual(i.e. aminimumestimate).
Weappliedregression equationstoestimatethe original weightofthepreyconsumedbythemink.In thecaseoffish,regressions betweenthesizeofkey bonesandoriginallength(PRBeja,unpubl.data),and betweenlengthandweight(ownunpubl.data)were computed.Theweightsofingestedcrabs(atleastfour differentspecies)wereestimated bydirectregression betweenthethirdmaxilliped’s merossizeandcrab weightwithoutclaws(theywererarelyfoundinscats) computed fromasampleofshorecrabs(Carcinus maenas) (ownunpub.data).Thelengthsandweights ofcrabandfishindividualswithoutmeasuredkeyhard partswereassumedtofollowthefrequencydistribution intheestimated samplesofeachpreyclass.Constant weightswereassignedtotheremaining preygroups: insects,1g;amphibians(Ranasp.),15g;birds,30g.
Theingested biomass ofeachpreyclasshasbeen estimatedthroughtwomethods. First, wesummed theestimatedbiomassofconsumedindividualsofeach sizeclassforfish andcrabs,asexplainedbefore. Becausethismethodcanoverestimatethecontribu- tionofsomepreygroupsproducingmoreremains(e.g. crabsorbirds),wehavealsoweighedthedryremainsof eachpreyclassinthefaeces,assumingthattherelative defecatedbiomassisproportionaltotheingestedfresh biomass.Thisestimation wascarriedoutonlyforthe largerpreycategories(i.e.fish,crabs,etc.).
Theestimated ingestedbiomassbelonging toeach preytypewastransformedto theirenergeticcontribu- tionaccordingtothecalorificvalues(kJ/gwetweight) givenbyDunstone(1993).Thedietanalysisresults wereexpressedasRFO,percentageofindividuals,per- centageofbiomassingestedandpercentage ofener- geticcontributionofthedifferentpreyclasses.Thelast tworesultswereexpressedasarange,becausethey includetheestimationsobtainedbytwomethods.
Tocalculatethepreyrequirementsofanindivid- ualminkweassumedadailyenergyrequirementof
1250kJ,themeanofthevaluesgivenbyDunstone (1993)foranadultfemale(1000kJ)andanadultmale (1500kJ).Thepercentageofenergyobtainedfrom
eachpreyclasswasusedtocalculate theingested biomassofthedifferentclasses.Thisvaluewasdivided proportionally amongthedifferentweightcategories toestimatethenumberofindividuals ofapreyclass predateddailybyasinglemink.
Results
Frequencies ofoccurrenceofmainpreytypesinthe minkfaecesdidnotshowsignificantvariationsbetween thetwoyears(χ2 =4.99; df =2; P 0.08), therefore wepooledtheresults.Minkdietwasbased almostexclusively oncrabs(RFO51.6%)andfish (RFO46.3%)(Table1).Birdsappearedinfourfae-
ces,insectsintwo,andonlyonecontainedamphibian remains.Wehavenotconsideredanyotherkindofprey, althoughduringtheanalysesremainsofsmallmussels, snailsandlimpetswerefoundinmorethan40%ofthe samples.Theseoccurrencescorresponded toanimals predatedbyblennyfish, whichwereinturntakenby mink.Theassociationofremainsofblenniesandmol-
luscsinthefaeceswashighlysignificant(χ2=58.33; df=1;P 0.001).Inaddition,wefoundmollusc remainsofthesame taxaandsizes inthestomachsof
someblenniescapturedinthestudyarea.
Themarbledcrab(Pachygrapsusmarmoratus)was themostfrequentlyconsumedspeciesintermsofRFO (34.9%),becausetheclass‘blennies’(38.1%)included morethanonespecies(atleasttwowereidentified: theMontagu’sblenny,Coryphoblenniusgalerita,and theshanny,Lipophrys pholis,whichaccording toour fieldobservationsandtheliteratureare,withagreat difference,themostfrequent blennies intheupper intertidal zoneoftheAtlanticcoastontheIberian Peninsula: Arruda1979; Iba´n˜ezetal. 1989; Faria andAlmada1999).Otheridentifiedcrabspecieswere velvetswimmingcrab(Necorapuber),shorecrab,and wartycrab(Eriphiaverrucosa).Regarding otherfish species,rocklingswereprobablyGaidropsarussp.and gobieswereGobiuscobitis.Dietdiversitywaslower
(Ht =0.80)thanreportedinotherstudiesofAmerican
minkdietinEurope(Jedrzejewskaetal.2001),dueto
thelowproportionofoccurrenceofpreyclassesother thanfishorcrab.
Inthestudyareaminkfedmainly onsmallprey (Figure1).Medianlengthofconsumedfishwas7.8cm (mean±SD = 7.9±1.87cm; n = 131); the largestrecordedfishwasa14cmlonggoby.Usually, Montagu’sblenniesandshanniesareconsideredjuve- nileswhentheyarelessthan3cmtotallength(Faria
Table1.Relativefrequencyofoccurrenceandpercentageofindividuals,ingestedbiomass andobtained energyfromthedifferent preyclassesfoundinminkfaecescollected inthe studyareainsummerof1997and1999.
RFO%Individuals%Biomass%Energy
Blennies / 38.1 / 47.7 / 34.2 / 52.7Rocklings / 2.9 / 2.3 / 2.6 / 4.0
Gobies / 2.1 / 1.3 / 1.8 / 2.7
Unidentifiedfish / 3.2 / 2.1 / 1.6 / 2.4
Totalfish46.353.342.8–68.161.8–82.4
Marbledcrabs / 34.9 / 33.0 / 36.9 / 24.0
Swimmingcrabs / 5.6 / 3.9 / 4.6 / 3.0
Shorecrabs / 4.4 / 4.1 / 4.2 / 2.8
Wartycrabs / 0.6 / 0.4 / 0.4 / 0.3
Unidentifiedcrab / 6.2 / 3.9 / 4.4 / 2.9
Totalcrab51.645.430.0–53.915.6–33.3
Birds / 1.2 / 0.8 / 1.5–2.9 / 1.9–4.3
Insects / 0.6 / 0.4 / 0.0 / 0.0
Amphibians / 0.3 / 0.2 / 0.1–0.4 / 0.1–0.5
Totalotherprey / 2.1 / 1.4 / 1.6–2.3 / 2.0–4.8
Totalnumbers / 341occurrences / 533individuals
andAlmada1999);hence,theblenniescapturedbythe minkshouldbeadults(Figure1A).Themedianweight ofpreywas5.2gforfish(mean±SD=6.4±4.5g;
n=131)and5.1gforcrabs(mean±SD=8.0±
8.46g;n=139).Meanweightvalueswerelargerthan
medianvaluessincethedistributionofweights,both
forfish andcrabs,washeavilyskewedtowardlarger individuals(Figures1BandC).
Fish representedbetween 43% and 68% of the biomass ingested by mink and crabs represented between30%and54%.Accordingtoourestimates, fishrepresentedbetween62%and82%andcrabs between16%and33%oftheenergyintake(Table1). Duetothesmallsizeofconsumedprey,minksmust capturemanyindividuals(between30and50)ofthe differentpreyclassestosatisfytheirdailyenergetic requirements(1250kJforanaverageweightindivid- ual).Onaverage,asingleminkmustconsume110–
147goffish(26–35individuals)and65–139gofcrabs (10–22individuals)daily,dependingonthemethod usedtoestimatetheproportions ofingestedbiomass. Assumingtheseresultscanbeextrapolatedtothewhole monthofAugust (31days),asingleminkwould consumeabout945fishand496crabs.
Discussion
ItiswellknownthattheAmerican minkisanoppor- tunisticandadaptablepredator, whichcanfeedon
mammals,birds,amphibians,fishandinvertebrates (Dunstone1993).Ourresultscorroborate thisoppor- tunisticforagingbehaviour,showingthatthespecies isabletorelyexclusively onalowdiversitydietof marinepreycaptured inanarrowshorestripofan urbanarea.Rathersimilarresultswereprovidedby Skirnisson (1979)fromacoastalpopulation inIce- land,whereminkfedmainlyonfishesandbirds.Also, DunstoneandBirks(1987)foundrockypoolfishin
41%ofminkfaecescollected onthecoastofsouth- westScotland,whererabbits(Oryctolaguscuniculus) comprised thebulkofminkfood.Overall,fishseems tobethemostimportant preyofcoastalminkswher- evertheirdiethasbeenstudiedinEurasia(Macdonald andStrachan1999;Jedrzejewskaetal.2001).Hence, thephenomenon onwhichwepayattentioncouldbe rathercommoninrockyshoreareasofEurasiainvaded bymink.Bycomparing stableisotoperatiosofmink tissueswiththoseofthepotentialprey,Ben-Davidetal. (1997)alsoconcludedthat,insoutheast Alaska,diets ofnativecoastalminksinspringandsummerconsisted largelyofintertidalfishes.TheRFOofcrabsinBaiona isthehighestreportedinminkdietstudiesfromcoastal habitats,butshorecrabsareanimportantpreyalsoin Scotland(DunstoneandBirks1987).
Inourstudy,minkseemstoforagealmostexclu- sivelyonintertidalpools,where,according toour unsystematicobservations,bothblenniesandmarbled crabswerebyfarthemostabundantavailableprey. Weobservedthatsomeshrimp(e.g.Palaemonspp.)
A
45
30
15
0
B 50
40
30
20
10
0
C 80
60
40
20
0
30507090110130
Lengthoffish(mm),N=131
2.57.512.517.522.527.5
Weightoffish(g),N=131
5152535455565
Weightofcrabs(g),N=139
1997).Theminkdietwereportcouldberathersimilar tothedietofcoastalyoungotters,whichhavealower divingsuccessratethanadultsandpredatemoreoften oncrabsandsmallfish(Watt1993).
Itisusuallyrecognizedthatmostpredationupon rockyintertidalfishassemblagesisexertedby‘exter- nal’predators,thatis,largersubtidalfish anddiving birdsinhightide,andterrestrialpredators,suchas birds,reptiles andmammals,whenthetideisout (GibsonandYoshiyama1999).However,theinflu- enceofpredationonlocalabundanceanddistribution ofintertidalfish isessentiallyunknown(Gibsonand Yoshiyama1999;Pfister1999).Wecannotestimate theimpactofminkpredationonintertidalfishorcrabs inourstudyareabecausewelackaccurateinformation abouttheabundance ofminkandtheabundance and renewal ratesoftheirprey.However,wecanpropose anapproach.
Minkfemalehomerangesofabout1kmlongand winteradultabundanceofabout2mink/kmofcoastline havebeenreportedinScotland(DunstoneandBirks
1983).Aswerepeatedly saw2familygroupswith3 and4individualsinourstudyareainAugust1999, itwaspossibletoestimateconservatively that8dif- ferentindividuals(adultsandyoung)werelivingin our2kmofcoastline(i.e.apredispersalabundance of4mink/km). Basedonourresults, theseminks couldconsumeabout3780blennies(3240–4340)and
1984crabs(1240–2728)perkmofcoastlineduring
August.
Weareawarethesenumberscannotbeextrapolated
Figure1.Distributionofestimatedlengths(A)andweights(B)of
fishandweightsofcrabs(C)consumedbytheAmericanminkinthe studyarea.Numbersin horizontalaxesindicatethecentralvalueof theintervals.
werecommonbutnotpredated.Dunstone andBirks (1987)alsofoundthatshrimpwereratherabundant inpools, butrarelyeateninScotland. Maybethis isbecause shrimparetoosmallorbecause theyare moreabundant inthemiddleandlowtidalareas, whereas minkforagemostlyinthehightidalzone becauseofitsreducedabilitytohuntunderwater(Poole andDunstone 1976).Infact,someblenniescouldbe capturedoutofwater,astheyshownocturnalemer- gencebehaviour (Louisy1987)andcoastalminkfor- agemainlyduringdarkness andwhenthetideislow (Dunstone 1993).OttersfeedinginsimilarIberian rockyshoreareasselectbigger andmoreprofitable subtidalfishes,suchaswrasses(Symphodussp.)(Beja
tothewhole year,foratleasttworeasons.First, blennies alongtheIberianAtlantic coastseemtobe moreavailable topredators insummer(Beja1997) andminkdietcouldbedifferentinotherseasons.Sec- ond,energyrequirements(i.e.preyconsumption)of thebreedingminkpopulation arehighestinAugust, whentheyoung arealmost fullygrown butstillliv- ingintheirmother’shomerange(Dunstone 1993). Besides,feralminkinvaded thestudyareaonlysome yearsagoandahighpredation pressureseemstobe characteristicofrecentlyintroducedpredators,which meetna¨ıveprey(e.g.BreitenmoserandHaller1993). Regardless,evenifabsolutepredation islowerin otherperiods,ourresultsseemtosuggest apotential impactoftheAmericanminkonintertidalfishandcrab populations.
Thispopulation effectofpredationdependsheav- ily on the numbers of blennies and crabs exist- inginthearea. Intertidalfishesareusuallyscarce,
densitiesrarelyexceeding onaverageafewindi- viduals persquaremeter,evenwhentheyarecon- centrated inlow-tiderefugessuchasrockypools (Gibson 1982).Besides,tidepoolsexperimentally depopulatedoffishesshowrelativelyslowrecoloniza- tionrates,theeffectsofdepopulation beingappre- ciated for weeks (Gibson 1982). However, Faria andAlmada(1999)stronglysuggestedthatdensity- dependentmechanisms controlthenumbersofadult blenniesfoundinPortuguese tidepools.Inthiscase, mortalitybyminkpredationcouldbeat leastpartially compensatory.Clearly,morefieldstudiesareneeded toaccurately determine theimpactofpredationby non-indigenouscoastalmink,butremovalofseveral thousandindividualblenniespermonthandkilometre of shorelinecouldhaveaseriouseffecton theirpopu- lations.Maybeminkpredationoncrabshasalesser impact,althoughnotnegligible,inourarea,because atleastmarbled crabsseemtobeextraordinarily abundant.
Byreducingnumbersofshore crabs andfish,mink predationcouldinfluenceintertidalcommunitiesasa whole.Predationhaslongbeenconsidered amain factorin structuringintertidalassemblages(e.g.Paine
1974).Forinstance,crabsareknown tobeableto alterthesurrounding habitatsthroughpredationon snailsandmussels(e.g.Leonardetal.1998;Gerard etal.1999).Similarly,anincreasing bodyofknowl- edgeindicatesthatpredationbyintertidalfishcanhave dramaticeffectsontheirprey(Anderson andConnell
1999;NortonandCook1999).Theinvasion byatop predatorsuch astheAmericanmink canproducecas- cadingeffectsindifferent trophiclevelsofthecom- plexstructure ofthemarineintertidal foodwebs,by releasingthepredationpressureofcrabsandfishupon snailsandsessileinvertebrates(PolisandStrong1996). Additionally,thepresenceofexotic minkcouldinflu- enceintertidalcommunitiesbyalteringthebehaviour ofitsprey,resultingincascadingindirecteffectson competitiveinteractionsandfoodchains(Trusselletal.
2003).
Insummary,theimportanceforcoastalinvasionsof exoticspeciesarriving through theseaiswellknown (Grosholz2002),butthosearrivingthroughinlandcan bealsosignificant. Thisisbecauseatlowtideshore birdsandmammals(includingnon-indigenous rats, feralcats, dogsandpigs, minks,mongooses,etc.) can consumeintertidalorganismsathighrates,probably exceedingthosefromfullyaquaticpredators(Edwards etal.1982).
Acknowledgements
I. Guilarte gave us her hospitality in Baiona and J.Mateos,F.Mateos andmainly M.Delibes-Mateos collaboratedinthefieldwork.A.Delibesandhissons helpedinthecaptureoffishandcrabs.I.Munilla, L.J.AlbertoandP.Nietoprovideduswithbibliography. I.Lo´pez,G.Mun˜oz,S.Rodr´ıguezandmainlyP.Beja gave helpful support for the identificationof fish andcrabremains.Themanuscriptgreatlybenefited fromcommentsbyN.Ferna´ndez, E.Macpherson, F.Palomares,E.Revilla,A.Rodr´ıguezandananony- mousreferee.S.Conradihelpedinmanyways.
References
AndersonMJandConnellSD(1999)Predationbyfishonintertidal oysters.MarineEcologyProgressSeries187:203–211
ArrudaLM(1979)Specificcompositionandrelativeabundance of intertidalfish attwoplacesonthePortuguesecoast(Sesimbra andMagoito,1977–78).ArquivosdoMuseuBocage (2aseries)
6:325–342
BejaPR(1996) AnanalysisofotterLutra lutrapredationonintro- ducedAmericancrayfishProcambarusclarkiiinIberianstreams. JournalofAppliedEcology33:1156–1170
BejaPR(1997)Predationbymarine-feedingotters(Lutralutra) insouth-westPortugalinrelationtofluctuatingfoodresources. JournalofZoologyLondon242:503–518
Ben-DavidM,HanleyTA,KleinDRandSchellDM(1997)Seasonal changesindietsofcoastalandriverinemink:theroleofspawning Pacificsalmon.CanadianJournalofZoology75:803–811
BreitenmoserUandHallerH(1993)Patterns ofpredationbyrein- troducedEuropean lynxintheSwissAlps.JournalofWildlife Management57:135–144
DiamondJ(1989)Overviewofrecentextinctions.In:WesternD andPearlsN(eds)Conservation fortheTwenty-First Century, pp37–41.OxfordUniversityPress,Oxford,UK
DunstoneN(1993)TheMink.TandADPoyser,London,232pp
DunstoneNandBirksJDS(1983)Activitybudgetandhabitatusage bycoastal-livingmink(MustelavisonSchreber).ActaZoologica Fennica174:189–191
DunstoneNandBirksJDS(1987)Thefeedingecologyofmink
(Mustelavison) inacoastal habitat.JournalofZoologyLondon
212:69–83
EdwardsDG,Connover DOandSutterF(1982)Mobilepredators andthestructureofmarineintertidalcommunities.Ecology63:
1175–1180
FariaCandAlmadaV(1999)Variationandresilienceofrockyinter- tidalfishinwesternPortugal.MarineEcologyProgressSeries
184:197–203
FerrerasPandMacdonaldDW(1999)TheimpactofAmericanmink Mustela visononwaterbirdsintheupperThames. Journalof AppliedEcology36:1–8
GerardVA,CerratoRMandLarsonAA(1999)Potentialimpactsof aWesternPacificgrapsidcrabonintertidalcommunitiesofthe NorthwesternAtlanticOcean.BiologicalInvasions1:353–361
GibsonRN(1982)Recentstudiesonthebiologyofintertidalfishes.
OceanographicandMarineBiology20:363–414
GibsonRNand YoshiyamaRM(1999)Intertidalfishcommunities.
In: HornMH,MartinKLMandChotkowsiMA(eds)Inter- tidalFishes:LifeinTwoWorlds,pp264–296. Academic Press, SanDiego,California
GrosholzE(2002)Ecologicalandevolutionaryconsequencesof coastalinvasions. TrendsinEcologyandEvolution17:22–27
Iba´n˜ezM,MiguelI,SanMilla´nMDandRipaMI(1989) Intertidal ichthyofauna oftheSpanishAtlanticcoast.ScientiaMarina53:
451–455
Jedrzejewska B,SidorovichVE,PikulikMMandJedrzejewski W (2001)FeedinghabitsoftheotterandtheAmericanminkin BialowiezaPrimevalForest(Poland)comparedtootherEurasian populations.Ecography24:165–180
Leonard GH,LevineJM,Schmidt PRandBertness MD(1998) Flow-driven variationinintertidalcommunitystructureina Maineestuary.Ecology79:1395–1411
LouisyP(1987)Observations surl’emersionnocturnededeux blenniesMediterraneens:Coryphoblennius galeritaetBlennius trigloides(Pisces,Perciformes).Cybium11:55–73
MacdonaldDWandStrachanR(1999)TheMinkandtheWaterVole: AnalysesforConservation.WildlifeConservationResearchUnit andtheEnvironmentAgency,Oxford,161pp
MacdonaldDWandThomMD(2001)Aliencarnivores:unwelcome experimentsinecologicaltheory.In:GittlemanJL,FunkSM, MacdonaldDandWayneRK(eds)CarnivoreConservation, pp93–122.CambridgeUniversityPress,Cambridge,UK
Macdonald IAW,LoopeLL,UsherMBandHamannO(1989) Wildlifeconservation andtheinvasionofnaturereservesby introducedspecies:aglobalperspective.In:DrakeJ,MooneyH, diCastriF,GrovesR,KrugerF,RejmanekMandWilliamsonM (eds)Biological Invasions:AGlobalPerspective,pp215–255. JohnWileyandSons,Chichester,UK
Mitchell-Jones AJ,AmoriG,BogdanowiczW,KrystufekB, ReinjdersPJH,Spitzenberger F,StubbeM,ThissenJBM, Vohral´ıkVandZimaJ(1999)TheAtlasofEuropeanMammals. TandADPoyser,London
Nordstro¨mM,Ho¨gmanderJ,LaineJ,NummelinJ,LaanetuN andKorpima¨ki E(2003)Effectsofferalminkremovalon seabirds,wadersandpasserinesonsmallislandsintheBaltic sea.BiologicalConservation109:359–368
NortonSFandCookAE(1999)Predationbyfishes intheinter- tidal.In:HornMH,MartinKLMandChotkowsiMA(eds)Inter- tidalFishes:LifeinTwoWorlds,pp223–263. Academic Press, SanDiego,California
PaineRT(1974)Intertidalcommunity structure.Experimental studiesontherelationship betweenadominantcompetitorand itsprincipalpredator.Oecologia39:1–24
PalomoLJandGisbertJ(2002)Atlasdelosmam´ıferos terrestres deEspan˜a.Direccio´nGeneraldeConservacio´ndelaNaturaleza, SECEM,SECEMU,Madrid,Spain
PfisterCA(1999)Recruitmentofintertidalfishes.In:HornMH, Martin KLM andChotkowsiMA(eds)IntertidalFishes:Lifein TwoWorlds,pp181–196.AcademicPress,SanDiego,California PolisGAandStrongDR(1996)Foodwebcomplexityandcommu-
nitydynamics.TheAmericanNaturalist147:813–846
PooleTBandDunstoneN(1976)Underwaterpredatorybehaviourof theAmericanmink(Mustelavison).JournalofZoologyLondon
178:395–412
Rosello´E(1986)Atlasosteolo´gicodelosteleo´steosibe´ricos.I.
Mand´ıbulainferior (dentarioyarticular).EdicionesdelaUAM, Madrid,Spain
RuizGM,Fofonoff PW,CarltonJT,Wonham MJandHinesAH (2000)InvasionofcoastalmarinecommunitiesinNorthAmerica: apparentpatterns,processes,andbiases.AnnualReviewEcology andSystematic31:481–531
SimberloffD (1981)Communityeffectsof introducedspecies.
In:Nitecki M(ed)BioticCrisesinEcologicalandEvolutionary
Time,pp53–81.AcademicPress,NewYork
Skirnisson K (1979) Food habits of mink (Mustela vison) in
Grindavik,southwestIceland.Natturufraedingurinn49:194–203
TrussellGC,EwanchukPJandBertnessMD(2003)Trait-mediated effectsinrockyintertidalfoodchains:predatorriskcuesalter preyfeedingrates.Ecology84:629–640
VitousekPM,DantonioCM,LoopeLL,Rejmanek Mand WestbrooksR(1997)Introducedspecies:asignificantcomponent ofhuman-causedglobalchange.NewZealandJournalofEcology
21:1–16
WattJP(1993)Ontogenyofhuntingbehaviourofotters(Lutralutra) in a marineenvironment.Symposiaof theZoologicalSocietyof London65:87–104.
WebbJB(1980)OtterSpraintAnalysis.Occasional Publication of theMammalSociety13,London