Trophicdiversityoftheotter(LutralutraL.)in temperateandMediterraneanfreshwaterhabitats
MiguelClavero1,2*,Jose´Prenda1andMiguelDelibes2 1DepartamentodeBiologı´aAmbiental ySaludPu´blica,UniversidaddeHuelva,CampusUniversitariodeElCarmen,Huelva,Spain and2DepartamentodeBiologı´aAplicada,Estacio´nBiolo´gicadeDon˜ana,CSIC,Pabello´ndel Peru´,Sevilla,Spain
Abstract
Aim To analyse thegeographicalpatterns inthe compositionand diversity ofotter’s
(LutralutraL.)dietandtheirrelationshipwithclimaticcharacteristics.
LocationEuropean freshwater habitats underMediterraneanandtemperateclimatic regimes.
MethodsThirty-sevenotterdietstudieswerereviewed,twenty-one fromtemperateand sixteenfromMediterraneanareas. Allstudieswerebasedonspraint analysisandtheir results expressed as relative frequency of occurrence of seven main prey categories. PrincipalComponent Analysiswasperformed toextract themaingradientsofdiet composition. Pearson’s correlation and t-tests were used to assess the relationship between diet characteristics(composition,diversity and taxonomicrichness) and geo- graphical andclimaticvariables.
ResultsAclearlatitudinalgradient indietcompositionwasobserved. Otter dietwas morediverseandfeatured morepreyclassesinsouthern localities,whilethespecieswas more piscivorous towards thenorth, where itpredatedupon ahigher number offish families. This patternwas similar when temperateand Mediterraneanlocalities of Europewerecompared.Mediterraneanottersbehavedasmoregeneralistpredatorsthan temperateones,relyinglessonfish,andmoreonaquatic invertebratesandreptiles.
MainconclusionsGeographicaldifferencesinotter feedingecologyinEurope seemto berelated with thetwo contrastedclimatic conditions affectingpreypopulations.The otter canactasahighlyspecializedpiscivorous predatorintemperatefreshwater eco- systems,whichdonotsufferadryseasonandhaveacomparativelystablewaterregime compared toMediterranean ones.However, theunpredictablepreyavailability in Mediterraneanareas,affectedbystrongspatialandtemporal watershortages, favoursa diversification oftheotter’sdiet.
Keywords
Feedinghabits, freshwater ecosystems, freshwater fishes,geographicalvariation,otter,
Lutralutra,Mediterraneanclimate.
INTRODUCTION
Thelatitudinalgradient indiversity(i.e.anincreasing rich- nessoffloraandfaunaspeciesfromthepolestothetropics) isconsideredtheoldest(Hawkins,2001)andthebestknown (Rosenzweig, 1995)biogeographical‘pattern’inecology.As
*Correspondence:MiguelClavero, DepartamentodeBiologı´aAmbiental y
SaludPu´blica,Universidad deHuelva, Campus Universitario deElCarmen, Avd/FuerzasArmadas s/n,21007 Huelva, Spain.E-mail:
abundanceand diversity ofpotentialpreyarethemostim- portant featuresdetermining apredator’s nichewidth (MacArthurPianka, 1966; Schoener, 1971), we could expect, onalarge scale,that predatorspeciesoccupying a broad geographicalrangemayexperiencevariations oftheir food-niche breadth following changesinpreycommunities. Attheintraspecificlevel, suchdifferentialuseoftrophic resources among allopatric populations may occur with minor or no apparent morphological or physiological changes(Futuyma Moreno,1988; Martı´n etal.,1995).
Studiesonthetrophic ecologyofseveralspeciesofbirds andmammals support theprediction ofadecreasingtrophic diversityathigherlatitudes. Aclassicalexampleisthepuma (PumaconcolorL.),whoserangeextendsfromAlaskainthe north to Tierra delFuego inthe south and whose highest food-niche breadths arereached intropical America(Iriarte etal., 1990). In Europe, an increasing dietary diversity towards thesouthhasbeensuggestedforthebarnowl(Tyto alba Scop.;Herrera,1974), Montagu’sharrier (Circus pigargusL.;Arroyo, 1997)andotherraptors (Korpimaki Marti, 1995),amongthebirds,andcommon genet(Genetta genettaL.;Virgo´setal.,1999) among themammals.
Asabiotic environmentalfactors, suchasclimateregime,
areoneofthemost importantdeterminantsofthecompo- sition and structure ofprey communities (SmithSmith,
2000),theobserveddietarypatternsin Europecouldbe associated with the two principal climatic areas definedin the continent: temperatein central and part of northern Europe andMediterraneaninthesouth (Fig.1).Temperate climateischaracterizedbycoldwinters andmildandrainy summers, withoutasummer drought.Mediterraneanareas (theterm‘Mediterranean’is alwaysusedinthetextina climatic,andnotstrictlygeographical, sense),considered transitionalbetween coldtemperateanddrytropical zones, arecharacterizedbytheuniquecombination ofhotdry summersandcoolhumidwinters. Littleornosurfacewater is available during the summer, generating a period of effectivephysiologicaldrought.Addedtothisseasonalityisa
characteristicunpredictability,withdramatic variations of weather conditions between years,between seasonsofagi- ven yearandeven inthecourseofasingleday(BlondelAronson, 1999).
Freshwater ecosystemsareespeciallysensitivetopre- dictableandunpredictableenvironmentalvariations suchas thosecharacteristic ofMediterranean climates(PrendaGallardo,1996; Gasith Resh, 1999; Magalha˜es etal.,
2002). Consequently,theEurasian otter (LutralutraL.),a
top predatorofaquatic ecosystems, could experiment changesinitsfoodandforagingbehaviour according tothe compositionandstructure ofpreycommunities inMediter- ranean and temperateecosystems. The strong decline suf- fered by the populationsof the otter in Europe since the
1950s haslargelystimulated research into thespecies’dis- tributionand ecology (Mason Macdonald,1986), diet beingoneofthecentralissues(Carss,1995).Suchasituation providesagoodchancetoanalysethepossiblevariations in theotter’strophic diversitythroughoutEurope.
Mason & Macdonald (1986) suggested that otters in southern Europe featured more amphibiansand reptiles in their dietsthan those ofthenorth. Also,Adria´nDelibes (1987)indicatedthatfrequencyofoccurrenceof insects, amphibiansandreptilesinfaecesofottersinEuropeseemed toincreaseas latitudedecreased.Again,Ruiz-Olmo(1995) showedthatreptileswereacommon foodofMediterranean otters, but were very rare in other latitudes. However, a recentanalysisofthedietofottersinEurasiafailedtodetect
Figure1Distributionofthethirty-seven reviewedstudiesonthedietofotters (Lutra lutraL.).Thedotted lineseparates temperate (filledcircles)fromMediterraneanlocations (emptycircles).Thenumbers areassignedto originalworksasfollows.1:Canas(1999);2: Beja(1996); 3:Adria´nMoreno (1986);4,
5:Adria´nDelibes(1987);6:Lo´pez-Nieves
Hernando(1984); 7:Morales Lizana (1997); 8:Acera(1998); 9:Morales etal. (1998);10:CallejoDelibes(1987);11,12: Ruiz-Olmo etal.(1989);13:Arca´Prigioni (1987); 14–16: Prigionietal.(1991); 17: Lode´(1989); 18:Libois(1995); 19,20: Chanin (1981); 21:Wiseet al.(1981); 22: Webb(1975); 23:Henshilwood(1981); 24: Gormally Fairley(1982); 25:Kyneetal. (1989); 26:O’Neilletal.(1998); 27,29: Weber(1990); 28:Carssetal.(1990); 30: TaastrømJacobsen(1999); 31:Erlinge (1967); 32:Erlinge(1969); 33:Geidezis (1998);34:LanszkiKo¨rmendi (1996);35: KnollseisenKranz(1998);36:Wisniowska (1996); 37:Brzezin´skietal.(1993).
any of these trends (Je˛drzejewska etal., 2001). These authors stated that otter diets in Eurasia do not change withlatitude, butchangewithhabitat: fisharemorefrequent as prey on sea shores, followed by lakes, and rivers and streams;amphibiansandcrustaceans showjustthecontrary trend.
Theaim ofthispaper isto search forgeographicalpat- ternsindietcomposition anddiversityofotterslivingin Mediterraneanandtemperate climaticconditions inEurope. Toremovetheimportantbiasthat coastal otters canintro- duce,welimitedoursourcestofreshwater habitats.
MATERIALAND METHODS
Data onotter dietweretaken fromthirty-seven dietstudies fromtheavailableliterature (seeAppendix). Eachstudyarea wasassignedtoMediterraneanortemperateclimatecondi- tionsaccording tothemap ofEmberger etal.(1963). This resultedinsixteenMediterraneanandtwenty-one temperate localities (Fig.1). All the studies were based on spraint analysisandthedataexpressedas relativefrequencyof oc- currence(RFO)(numberofoccurrencesofacertainitemasa percentage ofthe total number ofoccurrences ofall prey items).Onlystudieswithmorethan 200occurrences ofthe different prey categories were considered. Usually, results from several places or streams from the same area were pooled to avoid pseudoreplication(Hulbert,1984). When- evernecessary, original data were transformedto RFO to allow acorrect comparison.Recent critical analyses (Carss
Parkinson, 1996;JacobsenHansen, 1996)haveshown that RFOisnotthebestmethod toassessotter diet,leading tooverestimationofmediumsize preyitemsandunder- estimation of the smaller and bigger ones. However, Ja- cobsen Hansen (1996) compared several methods and found that similarity of RFO results with those of more accurate methods wasbetween 80% and 90% (Renkonens IndexofSimilarity). Thus, beingthemostfrequent method usedinliterature,andforour aimofmaking comparisons, weconsider RFO anappropriatemethodologytoestablish dietary geographicalpatterns.
Seven basic prey categories were considered: fish, amphibians, reptiles,birds,mammals, crayfishandother aquatic invertebrates(beetles,damselflynymphs, small shrimps, etc).Foreachlocaldiet,wecalculated thenumber of preycategories(NPC)and,whenpossible(in thirty-five studies), thetotal number offishfamilies(NFF)present in the sample. To quantify the general trophic diversity for eachlocation, theShannon–Wiener index(H¢)appliedto frequencies ofoccurrence wasused.
Principal Component Analysis was performed to an arcsine transformed matrix of RFO·study locations (n¼37)inordertosummarizegeneralpatterns inotterdiet throughout thestudyarea.RFOwerearcsinetransformed beforethe analysisto homogenizevariances(Zar,1984). Pearson’scorrelationsbetweenoccurrence data, trophic and geographicalvariables and principal componentswerecal- culated. To assessdifferences between Mediterraneanand temperatediets we used t-tests. Whenever multiple t-test
wereperformed,significancelevelswerecorrected usingthe
Sequential Bonferroni test(Rice,1989).
RESULTS
Fish were the otter’s main food category, representing almost75% oftheconsumed preysamongotters inhabiting freshwater habitats(Table1).Theremaining categories can beconsidered assecondary inageneral approximationto otterdiet.Predation uponamphibians, crayfishandother invertebrateswas frequent, while birds, reptiles and mam- malswererather rareinthedietoftheotter.
Thefirstprincipal component (PC1)produced astrong ordinationofthedifferent studylocations, explaining 40% of the total observed variance. PC 1 defined a gradient running from high and almost exclusive fishconsumption torelativehighpredation uponaquatic invertebrates, amphibians,crayfishandreptiles(Table1).Thedietgradient representedbyPC 1showedamarkedgeographicalcom- ponent, being highly correlated with latitude (r¼0.63; P0.001) (Fig.2a). The second principal component explained only17.6%oftheobserved variance andcannot beinterpretedingeographical/ecologicalterms.
Latitude was also correlated with most prey items and trophic variables (Table2). Fish consumption showed a strong positive correlationwith latitude, a patternshared withavianprey,whilereptiles,crayfish andotheraquatic invertebrates wereconsumed moreatlowerlatitudes. The southwardintensification ofpredationuponthesefourprey categories occurred in parallel to an increase introphic diversity (Fig.2b)and inthe NPC inotter diet, andadecreaseintheNFF(Table2).Predation upon mam- malsandamphibiansdidnotshowanygeographicalpattern. Trophic diversitywasnegativelycorrelated withfish con- sumption and positively with the RFOs of amphibians, reptiles,crayfishandother aquatic invertebrates(P0.001 in all cases). The RFOs of mammals and birds were not related totrophic diversity.
Patternsinotterdietobtainedbytwo-sample comparisons betweenMediterranean andtemperate climaticareaswere similar to those shown by correlations with latitude
Table1meanrelativefrequency ofoccurrence forseveralotter (LutralutraL.)preyitemsinEurope andPearson’scorrelation(r) between themandtheprincipal component1
PreyitemsMean ±SDrFactor 1
Fish72.8±18.5 0.91*** Amphibians8.8±7.6)0.68*** Reptiles1.5±2.9 )0.61*** Birds 1.9±2.1 0.31(NS) Mammals 0.9±1.3 0.43** Crayfish 6.8±12.9 )0.49** Aquatic invertebrates 7.0±7.7 )0.79*** Eigenvalue 2.80
Explained variance (%)40.0
Significancelevels:***P0.001;**P0.01;*P0.05;NS,non- significant.
Figure2Relationshipbetween latitude and(a)principal compo- nent1(PC1)scores(r¼0.63;P0.001)and(b)trophic diversity, measured withtheShannon–Wienerindex(H¢),(r¼)0.57;
P0.001) forthirty-seven otter (Lutralutra)dietstudies.Filled
circles:temperatelocations; emptycircles:Mediterraneanlocations. Encircledpoint (Brzezin´skietal.,1993; point 37ofFig.1)iscom- mented inDiscussion.
(Table2).Mediterraneanotters hadmorediversedietsand fedonalarger NPC and asmaller NNF than those occu- pying temperatehabitats.The frequency of occurrence of fishand birds inotter dietwassignificantlyhigher intem- peratethaninMediterranean locations, whilethereverse happenswithreptilesandaquaticinvertebrates. Differences betweenMediterranean andtemperatelocations inthe consumption ofcrayfish,amphibians andmammals were smaller.
DISCUSSION
Althoughfisharetheotter’smainpreyeverywhereinEurope (Mason Macdonald,1986; Carss, 1995; Kruuk, 1995), their consumptionshows onaverage amarked decrease in southern localities.Thisiscompensatedforbymoreintense predationuponapoolofalternative prey,includingcrayfish, reptiles, amphibians and aquatic invertebrates. Some of thesealternative preycanconstitute thebulkofotter dietin some southern places. As a consequence, Mediterranean ottersshowhighertrophic diversity,predatinguponalarger NPCthan thoseoftemperatehabitats.
According to the latitudinal gradient in diversity, an increased trophic diversityoftheotter insouthern latitudes couldalsoberelated toahigherabundanceandavailability of non-fishpreyintheseareas.In fact,atleastreptilesand insectsareespeciallyabundant inMediterraneanEurope (Blondel &Aronson,1999),wherethewarmweathercon- ditionsallowthemtoreachlargesizesandtobeactivemost oftheyear. Moreover,therecent spread oftheintroduced American crayfish(Procambarusclarkii, Girard) haschan- gedthedietofottersandotherpredators inmanyMedi- terraneanhabitats(Delibes Adria´n, 1987; Beja, 1996; Correia, 2001), crayfishbecoming animportantprey.
However,ahigheravailabilityof alternative preymakes possible,althoughdoesnotimply,anincreaseof trophic diversity.Inaddition,thisusuallynecessitatesareductionin theabundance oravailability ofthepreferredprey(Stephens
Krebs,1986).Erlinge (1968)statedthatcaptiveotters preferred to predate upon fish,and apparentlyfisharethe otter’sstaplepreywheneverabundant, evenunderMedi- terraneanconditions.Infact,ottershaveanalmostexclusive piscivorousdietinsomeMediterranean localitieswhere alternative prey are probably abundant (see Ruiz-Olmo etal.,1989;Prigioniet al.,1991).Inthesameway,ottersin temperate areasfeedonahighdiversityofpreywhen availability offish isreduced,likeinBialowieza,Poland (Brzezin´ski etal., 1993; see Fig.2). Kruuk (1995) also relatedaincreaseof non-fishpreywithperiodsof lowfish abundanceinScottishrivers.
t-testresults(mean±SD)
Latitude (r) Temperate (n¼21) Mediterranean(n¼16) t
Table2 Pearson’s(r)significantcorrela- tionsbetween latitude andRFOsofgeneral preyitemsinotterdiet,trophic diversity(H¢),
number ofpreyitemsinotterdiet(NPC)and
number offishfamiliesinotter diet(NFF). SignificancelevelsforrvaluesasforTable1. Comparisonsofthedifferentvariablesunder different climaticconditions usingt-testare alsoshown.Significancelevelsformarked(*) tvalues,afterSequentialBonferroni test(ten tests)
NFF0.56***5.3±1.383.7±0.794.18*
Wearguethattheloworunpredictableavailability offish infreshwater Mediterraneanecosystems,morethanthehigh abundanceofalternative prey,favourstheincreasedtrophic diversityofMediterraneanotters. Theharsh environmental conditions duringthepronouncedsummerdrought areakey factorexplaining thecomposition anddynamicsofMedi- terranean freshwater communities (Prenda & Gallardo,
1996). Most streams and smallriversbecome dryorbreak into isolated pools during the summer, reducing the avail- abilityoffish(Piresetal.,1999).Besides,thehighintra-and interannualvariability intheprecipitationand temperature regimescharacteristicofMediterraneanareasseverelyaffects freshwater ecosystems,necessarilyresultinginunpredictable fishavailability (Mooney, 1981;Prendaet al.,2001).Infact, otter dietinsomeMediterraneanareashavebeenshown to reachits maximumdiversityduringthesummer,whenfish populationsare strongly affected by drought (Ruiz-Olmo etal.,2001). Thus,thewideningoftheotter’sfeedingniche inMediterraneanecosystemswould befavoured inanenvi- ronment wherefish populationsaretemporallyscarceand patchilyavailable,bothinspaceandtime(seeErlinge,1986). Sulkava (1996) described a very diverse diet of otters in central Finland (H¢¼1.16) (seeTable2), inataiga envi- ronmentwheretheextremelycold wintersproducegreat variations andunpredictabilityoffishavailability (thiswork was not included in the analyses because central Finland cannot beconsidered atemperatearea).
ThereviewonotterdietbyJe˛drzejewska etal.(2001)did
notfindanyrelationshipbetweendiversityorcompositionof otter dietand latitude, but concluded that otters behave as moregeneralistpredatorsinstreamsandriversthaninlakes andseashores,wherefisharemorefrequent prey.Thiscon- clusion supports our hypothesis in an indirect way, as it relateschangesin theotter’strophicdiversitywithhabitat features,especiallythestabilityof wateravailabilityandits effectsonfish abundance.Wehypothesize thatintra-and interannualabundanceandpredictabilityoffish resources couldrangefromamaximumin sea shoresandlakestoa minimum inMediterranean temporal streams,beinginter- mediate intemperatewater courses; thisshould beaccom- panied bychangesinthepatternofuseoffish,itsfavourite prey,bytheotter(Fig.3).
Severalreasons (thecriteria to selectthedietary studies, the covered range, the oversight ofincluding reptiles, etc.) help to explain why Je˛drzejewska etal.(2001) failed to describe the evident temperate–Mediterranean change in otter trophic niche breadth. However, the main cause probably wasthelowproportionofMediterraneanlocalities included intheir analysis: onlynineteen of102dietstudies camefromMediterranean areas,andonlytwelveofthem correspondtofreshwater ecosystems. Astheobserved lati- tudinal patternseemsnot tobegradual, but related tothe differentconditions inMediterranean andtemperateeco- systems,thepatternmaybeconcealed iftheproportionof localitiesisverybiasedtowards temperateplaces(seeFig.2 tonotethat inourstudyalatitudinaltrend canbedetected within Mediterraneanlocalities, but isnot apparentwhen only temperate ones are considered). Thus, Je˛drzejewska
Figure3Schematicdiagramrepresenting thesuggestedvariation in otter trophic diversityandfishcommunities characteristicsinEur- opeinrelationtowateravailabilityandstability.Anamplificationin otter trophic diversityisobserved asfishbecomescarceorunpre- dictably available. Numbers showthehabitats inwhichthereview byJe˛drzejewska etal.(2001) (1)andthisstudy(2)werecentred.
etal.(2001) could detect the increase inotter trophic di- versityfromsea-shoresandlakestotemperaterivers,butnot thegeographicalpatternpresented here,from temperateto Mediterraneanfreshwater habitats(Fig.3).
Thedecrease infishconsumptionand thecorresponding enlargement of the food-niche of Mediterranean otters can berelatedto theirsmallbodysize.Ruiz-Olmoetal.(1998) showedthatottersfromtemperateareaswerebetween35% and11%heavierthanottersinhabitingtheIberianPeninsula. Someauthors (KingMoody, 1982;Clevenger,1993)have relatedchangesinmustelids’bodysizewithdifferencesinfood abundanceand availability. Temporally reduced fishabun- danceandunpredictableavailability infreshwater Mediter- ranean environmentscould then favour smallbody sizesin otters, given that reduced energetic demands allow for a higherdietary flexibility(GittlemanPurvis,1998). Iriarte et al.(1990)alsorelated thereduced bodysizeofAmerican pumas inlowlatitudes with broadernichebreadths.Other authors (King,1991)provedthatbodysizeinsmallmustelids couldbelinkedtomeanpreysize.Although wedonothave dataonpreysizestotestthishypothesisinthecaseoftheotter, thehigherfrequencyofsmallandrelativelyunprofitableprey (i.e.insects)intheMediterraneanareacouldalsoberelatedto theotter’sreducedbodysize.
Theresultsofthispaper show acleardietary diversifica- tion ofotters inhabiting Mediterraneanfreshwater ecosys- temsinrelation withottersfromtemperateones,resultingin astrong latitudinalgradient. Similarresultshavebeenpub- lishedforother smalland medium sizepredatorsinMedi- terraneanareas, andarealsothought tobeinresponsetoa reductioninthediversityandabundanceoftheirmainprey (e.g. small mammals for the barn owl and the common genet; Herrera, 1974; Virgo´s etal., 1999). This pattern cannotbe extended,however,tobiggerMediterranean predators, such as the Iberian Lynx (Lynx pardinus
Temminck),thegoldeneagle(AquilachrysaetosL.)(Delibes,
1975)andthebadger (MelesmelesL.)(Martı´n etal.,1995; Goszczyn´ski etal., 2000), which in Mediterranean Iberia strongly predate upon rabbits (Oryctogaluscuniculus),an extremely abundant prey. Thus, the trend of increased dietarynichebreadth withreducedlatitudecannotbe gen- eralizedtoalltaxa.
ACKNOWLEDGMENTS
Wethank DrB.Je˛drzejewska, DrJ.Jime´nezNieva, DrA. Rodrı´guezandDrJ.Calzada fortheirhelpfulcomments on early versions ofthe manuscript.Two anonymousreferees provided usefulideasand corrections.Thestudywas financially supportedbyGIASA-CSIC,through the project
‘Medidas compensatoriasdelaAutovı´a A-381 Jerezdela
Frontera-LosBarrios’.
REFERENCES
Acera,F.(1998)Distribucio´nydietadelanutria(LutralutraL.,
1758) en elrı´o Francia, Salamanca. Tesis de Licenciatura, Universidad deSalamanca.
Adria´n, M.I. Delibes, M. (1987) Food habits of the otter (Lutralutra)intwo habitatsoftheDon˜ana NationalPark, SWSpain.JournalofZoology,London,212,399–406.
Adria´n,M.I.Moreno,S.(1986)Notas sobrelaalimentacio´n delanutria (Lutralutra)enelembalsedeMatavacas,Huelva. Don˜ana,ActaVertebrata,13,189–191.
Arca´,G.Prigioni, C.(1987) Foodoftheotter ontheFiora
River(Central Italy).ActaTheriologica,32,134–140. Arroyo, B.E.(1997)DietofMontaguharrier Circuspigargusin
CentralSpain.Analysisoftemporal andgeographicvariation.
Ibis,139,664–672.
Beja,P.R.(1996) Ananalysisofotter Lutralutrapredationon introducedAmerican crayfishProcambarusclarkiiinIberian streams. JournalofAppliedEcology,33,1156–1170.
Blondel, J.Aronson, J.(1999) Biologyandwildlife ofthe
Mediterraneanregion.Oxford UniversityPress,Oxford. Brzezin´ski,M.,Je˛drzejewski,W.Je˛drzejewska, B.(1993)Diet
of otters (Lutra lutra) inhabiting small rivers in the Bial-
owieza NationalPark, eastern Poland. JournalofZoology, London,230,495–501.
Callejo, A.Delibes,M.(1987)Dietadelanutria Lutralutra
(Linnaeus,1758)enlacuencedelaltoEbro,Norte deEspan˜a.
MiscellaneaZoologica,11,353–362.
Canas, C. (1999) Predac¸aopor lontra, Lutra lutra L., em habitatsestivais deuma ribeira intermitenteMediterraˆnica. Relato´rio de esta´gio do curso de Licenciatura em Biologia MarinhaePescas,Universidade doAlgarve,Algarve.
Carss,D.N. (1995) Foraging behaviour andfeedingecologyof theotterLutralutra:aselectivereview.Hystrix,7,179–194. Carss,D.N., Kruuk,H.Conroy, J.W.H. (1990)Predation on adult Atlantic salmon, SalmosalarL.,byotters, Lutralutra (L.),within the River Deesystem, Aberdeenshire,Scotland.
JournalofFishBiology,37,935–944.
Carss, D.N. Parkinson, S.G.(1996) Errors associated with otter Lutra lutra faecal analysis. I. Assessinggeneral diet fromspraints. JournalofZoology,London,238,301–317.
Chanin, P.(1981)Thedietoftheotteranditsrelation withthe feral mink intwo areas ofSouthwest England. ActaTheri- ologica,26,83–95.
Clevenger,A.P.(1993)Pinemarten(MartesmartesLinne´,1758) comparativefeedingecologyinaislandandmainland popu- lationofSpain.Zeitschriftfu¨rSa¨ugetierkunde,58,212–224.
Correia, A.M. (2001) Seasonal and interspecific evaluation of
predation by mammals and birds on the introduced red swamp crayfish Procambarusclarkii (Crustacea,Cambari- dae) inafreshwater marsh (Portugal). JournalofZoology, London,255,533–541.
Delibes,M.(1975) Somecharacteristicfeatures ofpredationin the Iberian Mediterranean ecosystem. XIICongresso da UniaˆoInternacionaldosBiologistasdaCac¸a,pp.31–36.
Delibes,M.Adria´n,M.I.(1987)Effectsofcrayfishintroduc- tiononotter Lutralutra foodintheDon˜anaNationalPark, SWSpain.BiologicalConservation,42,153–159.
Emberger,L.,Gaussen,H.,Kassas,M.Philippis,A.de(1963) Carte bioclimatique de la zoneme´diterrane´enne (Etude e´cologiquedelazoneme´diterrane´enne).UNESCO-FAO,Paris. Erlinge,S.(1967)Foodhabitsofthefish-otterLutralutraL.in
SouthSwedishhabitats.Viltervy,4,371–447.
Erlinge,S.(1968) Foodstudiesoncaptiveotters LutralutraL.
Oikos,19,259–270.
Erlinge,S.(1969)FoodhabitsoftheotterLutralutraL.andthe mink MustelavisonSchreber in a trout water in southern Sweden.Oikos,20,1–7.
Erlinge, S.(1986) Specialistsand generalists among mustelids.
Lutra,29,5–11.
Futuyma,D.J.Moreno,G.(1988)Theevolutionofecological specialization.AnnualReviewofEcologyandSystematics,
19,207–233.
Gasith, A & Resh, V.H. (1999) Streams in Mediterranean climateregions– abioticinfluencesandbioticresponsesto predictableseasonal events. AnnualReviewofEcologyand Systematics,30,51–81.
Geidezis,L.(1998) What dootters (Lutralutra)feedinacarp
pond area in Saxony, Eastern Germany. Boku-reportson
WildlifeResearchandGameManagement,14,65–72. Gittleman, J.L. Purvis, A. (1998) Body-size and species-
richnessincarnivores andprimates. ProceedingsoftheRoyal
SocietyofLondon.SeriesB,BiologicalSciences, 265, 113–
119.
Gormally, M.J.Fairley,J.S.(1982)FoodofottersLutralutra inafreshwater loughand anadjacent brackish loughinthe WestofIreland. JournalofZoology,London,197,313–321. Goszczyn´ski, J., Je˛drzejewska, B.Je˛drzejewski, W. (2000) Dietcompositionofbadgers(Melesmeles)inapristineforest and rural habitatsof Poland compared to other European
populations.JournalofZoology,London,250,495–505.
Hawkins, B.A. (2001) Ecology’s oldest pattern?. Trendsin
EcologyandEvolution,16,470.
Henshilwood,D.A. (1981) Status and diet ofthe otter (Lutra lutra) at Bosherston lakes, Dyfed. Otters,ecology and conservation.(edsC.F.Mason and S.M.Macdonald).Cam- bridgeUniversityPress,Cambridge.
Herrera,C.M. (1974) Trophic diversity oftheBarnOwl Tyto albaincontinentalWestern Europe. OrnisScandinavica,5,
181–191.
Hulbert, S.H.(1984)Pseudoreplicationandthedesignof ecological field experiments. EcologicalMonographs, 54,
187–211.
Iriarte, A., Franklin, W.L., Johnson, W.E. Redford, K.H. (1990) Biogeographic variation offoodhabits andbodysize oftheAmerican puma. Oecologia,85,185–190.
Je˛drzejewska, B.,Sidorovich,V.E.,Pikulik,M.M. Je˛drzejew- ski,W.(2001) Feedinghabits oftheotter andtheAmerican mink in Bialowieza Primeval Forest (Poland) compared to other Eurasian populations.Ecography,24,165–180.
Jacobsen, L.Hansen, H.M. (1996) Analysisofotter (Lutra lutra) spraints: Part 1:Comparisonofmethods to estimate preyproportions;Part 2:Estimationofthesizeofpreyfish. JournalofZoology,London,238,167–180.
King,C.M.(1991)Bodysize-preysizerelationshipsinEuropean stoats Mustelaerminea:atest case. HolarcticEcology,14,
173–185.
King, C.M. Moody, J.E. (1982) The biology of the stoat (Mustelaerminea)intheNationalParksofNewZealand III. Morphometricvariation inrelation togrowth, geographical distributionand colonisation.NewZealand Journalof Zoology,9,81–102.
Knollseisen, M. & Kranz, A. (1998) Influence of different spraint sampling methods ontheresultsofotter dietstudies. Boku-reportsonWildlife Researchand GameManagement,
1,37–45.
Korpimaki,E.Marti, C.D. (1995) Geographicaltrends in trophic characteristics of mammal-eating and bird-eating raptors inEuropeandNorthAmerica.Auk,112,1004–1023. Kruuk, H. (1995) Wild otters.Predation andpopulations.
Oxford UniversityPress,Oxford.
Kyne,M.J., Smal,C.M. And Fairley, J.S.(1989) The food of otters Lutra lutrainthe Irish Midlandsan dacomparison with that ofthemink Mustelavison inthesameregion. ProceedingsoftheRoyalIrishAcademy,89,33–46.
Lanszki,J.Ko¨rmendi, S.(1996) Otter dietinrelation tofish availability inafishpondinHungary.ActaTheriologica,41,
127–136.
Libois, R.(1995) Re´gimeettactique alimentaires delaloutre
(Lutra lutra) en France: synthe`se.Cahiersd’Ethologie, 15,
251–274.
Lode´,T. (1989) Laloutre enLoire-Atlantique.Note pre´limi- nairesurl’evolutionre´centedespopulations,l’alimentationet lecomportementpre´dateur deLutralutra,L.1758. Bulletin delaSocietedeScienceNaturelldel’OuestdelaFrance,11,
69–76.
Lo´pez-Nieves, P.Hernando,J.A.(1984) Food habits ofthe otter in central Sierra Morena (Co´rdoba, Spain). Acta Theriologica,29,383–401.
MacArthur,R.H. Pianka, E.R.(1966) On optimal useofa patchy environment.AmericanNaturalist,100,603–609. Magalha˜es,M.F.,Batalha,D.C.Collares-Pereira,M.J.(2002)
Gradientsinstream fishassemblagesacrossaMediterranean
landscape: contributionsonenvironmentalfactorsandspatial structure. FreshwaterBiology,47,1015–1031.
Mart´ın, R.,Rodrı´guez, A.Delibes,M.(1995) Localfeeding specialization bybadgers (Melesmeles)inaMediterranean environment.Oecologia,101,45–50.
Mason, C.F. Macdonald,S.M. (1986) Otters,ecologyand conservation.Cambridge UniversityPress,Cambridge.
Mooney, H.A.(1981)Primaryproduction inMediterranean region. EcosystemsoftheWorld.Mediterraneantypeshrub- lands(edsP.diCastri, D.W. Goodwalland R.L. Specht). Elsevier,Amsterdam.