Latitudinal variation in sexual dimorphism in life-history traits of a freshwater fish

Ecology and Evolution 2017; 7: 665–673   | 665 www.ecolevol.org Received:22August2016 | Revised:24October2016 | Accepted:5November2016 DOI:10.1002/ece3.2658 ORIGINAL RESEARCH Latitudinal variation in sexual dimorphism in lifehistory traits of a freshwater fish Satu Estlander1 | Kimmo K. Kahilainen1 | Jukka Horppila1 | Mikko Olin1 | Martti Rask2 | Jan Kubečka3 | Jiří Peterka3 | Milan Říha3 | Hannu Huuskonen4 | Leena Nurminen1 ThisisanopenaccessarticleunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsuse,distributionandreproductioninanymedium, providedtheoriginalworkisproperlycited. © 2016 The Authors. Ecology and EvolutionpublishedbyJohnWiley&SonsLtd. 1DepartmentofEnvironmentalSciences/ AquaticSciences,UniversityofHelsinki, Helsinki,Finland 2NaturalResourcesInstituteFinland, Jyväskylä,Finland 3BiologicalCentre,AcademyofSciencesof theCzechRepublic,HydrobiologicalInstitute, ČeskéBudějovice,CzechRepublic 4DepartmentofBiology,UniversityofEastern Finland,Joensuu,Finland Correspondence SatuEstlander,DepartmentofEnvironmental Sciences/AquaticSciences,Universityof Helsinki,Helsinki,Finland. Email:[email protected] Funding information UniversityofHelsinki;R.ErikSerlachius Foundation;BergsrådetBrorSerlachius Foundation Abstract Sexualdimorphismiscommonacrosstheanimalkingdom,butthecontributionofenvironmentalfactorsshapingdifferencesbetweenthesexesremainscontroversial.In ectotherms,life-historytraitsareknowntocorrelatewithlatitude,butsex-specific responsesarenotwellunderstood.Weanalyzedlife-historytraitvariationbetween thesexesofEuropeanperch(Perca fluviatilisL.),acommonfreshwaterfishdisplaying largerfemalesize,byemployingawidelatitudinalgradient.Weexpectedtofindsex- dependentlatitudinalvariationinlife-historyvariables:lengthatage,lengthincrement,andsizeatmaturity,withfemalesshowingconsistentlyhighervaluesthanmales atalllatitudes.Wefurtheranticipatedthatthisgenderdifferencewouldprogressively decreasewiththeincreasinglyharshenvironmentalconditionstowardhigherlatitude. Wehypothesizedthatgrowthandlengthincrementwoulddecreaseandsize/ageat maturitywouldincreaseathigherlatitudes.Ourresultsconfirmedfemale-biasedsexualsizedimorphismatalllatitudesandthemagnitudeofsexualdimorphismdiminished with increase in latitude. Growth of both sexes decreased with increase in latitude,andthefemalelatitudinalclinesweresteeperthanthoseofmales.Hence,we challengetwopredominantecologicalrules(Rensch’sandBergmann’srules)thatdescribecommonlarge-scalepatternsofbodysizevariation.Ourdatademonstratethat thesetworulesarenotuniversallyapplicableinectothermsorfemale-biasedspecies. Ourstudyhighlightstheimportanceofsex-specificdifferencesinlife-historytraits alongalatitudinalgradient,withevidentimplicationsforawiderangeofstudiesfrom individualtoecosystemslevel. KEYWORDS Bergmann’srule,growth,perch,Rensch’srule,sex,sexualmaturity 1 | INTRODUCTION Sizedifferencebetweenthesexesisacommonphenomenonamong animals,butpronouncedinter-andintraspecificvariationexistsinthe magnitudeofsexualsizedimorphism(SSD)(Blanckenhorn,Stillwell, Young, Fox, & Ashton, 2006; Cox, Barrett, & John-Alder, 2008). Several theories have attempted to explain the variation in SSD by different factors including sex-dependent differences in sexual 666 | ESTLANDER ET AL. selection,vulnerability to predators, niche segregation, and parentalinvestment(Abouheif&Fairbairn,1997;Rennieetal.,2008).The regulationofSSDiscomplexbecauseeachofthesefactorsmayconstrainoramplifythedegreeofdimorphism(Shine,1989),anditisstill unclearhowthesedifferentfactorsdeterminethevariationinsexual dimorphism(Mandikietal.,2004;Young,2005).OneprominentmacroecologicalpatternistheRensch’srule(Rensch,1950),whichstates thatthemagnitudeofSSDtendstoincreasewithincreaseinbody sizewhenmalesarethelargersexandtodecreasewithincreasein sizewhenfemalesarelarger(Fairbairn,1997). Consequently,male bodysizevariesmorethanfemalebodysize,irrespectiveofwhich sexislarger.Rensch’sruleholdsforavarietyofanimaltaxa,forexample,insects,reptiles,birds,andmammals(Blanckenhorn,Meier,& Teder,2007;Fairbairn,1997).However,themechanismsunderlying Rensch’sruleremainobscure,andtheruleappearstobemoreconsistentintaxawithmale-biasedSSDthanintaxawithfemale-biased SSD(Webb&Freckleton,2007). A general assumption is that response of sexes is similar to changesinenvironment,butsomestudieshaveshowndifferential sensitivity of males and females to environmental factors such as temperature(Fairbairn,2005),thuspotentiallypromotingvariation in SSD. When environmental conditions improve, the sex that is moresensitivemayachieveoptimalsizemorereadilythaninpoorer conditionsresultinginanincreaseinSSD(Vedder,Dekker,Visser, &Dijkstra,2005),hence,consistentorcountertoRensch’sruledependingwhichsexismoresensitive.Forexample,largeindividuals requiremorefoodtoattainlargersizeandtomaintainbodyfunctions (Blanckenhorn,1998),andarelikelymoresensitivetothermalvariationviatheirhighermetabolicrates(Pörtner&Peck,2010).Thus, thisfollowsthatchangesintheenvironment,suchaslatitudinalvariation,mayhaveagreaterimpactonthelargersexinSSD-displaying species. WhileRensch’sruleexplainstherelationshipbetweenbodysize andextentofSSD,anotherwell-knownecologicalrule,Bergmann’s rule (Bergmann, 1848; ref. in James, 1970) describes geographical sizevariation. In addition, as Rensch’s rule depends on body sizevariation,ithasbeensuggestedthatBergmann’s(orconverse Bergmann’s) rule may relate to sexual size differences and their putative selective causes (Blanckenhorn etal., 2006). Bergmann’s rulestatesthatthebodysizeofawidelydistributedanimalclade increaseswithlatitude.WhilethedirectapplicabilityofBergmann’s ruleisunestablished(Blanckenhornetal.,2006; Meiri,2011),the consensus is that latitudinal body size variation is evidently connectedtotemperature(Blanckenhorn&Demont,2004).Theusual explanationtenderedforBergmann’sruleisthatlargeanimalsexpend lessenergyforthermoregulation,becauseoftheirsmallsurface-to- volumeratio,andtherefore,largerindividualsizeisfavoredincolder climates.Bergman’srulewasinitiallyformulatedforendothermicanimals,anditsextrapolationtoectothermsiscontroversial(Ashton& Feldman,2003).Infact,oppositeclinesinbodysizes(i.e.,converse Bergmann’s rule) are common in many ectotherms, such as frogs andsalamanders(Adams&Church,2008;Miaudetal.,2001),and inseveralfishspecies,bodysizedecreasestowardthepoles(New, Hulme,&Jones,1999;Vázquez& Stevens,2004). Inectotherms, suchasfish,thetemperature-associatedshortergrowingseasonat higherlatitudesmaylimitbodysize(Blanckenhorn&Demont,2004). However, Bergmann’s rule for fish is still relatively under studied (Rypel,2014). Intermsofgrowthorbodysizeplasticity,fishareaninterestinggroupbecausefishdisplayallometricgrowth,whichenablesa faster response to changing environmental conditions relative to manyendothermicanimals(Arnold,Ruf,&Kuntz,2006;Wootton, 2012).Temperature is the most important environmentalvariable governingmetabolicactivity(Brown,Gillooly,Allen,Savage,&West, 2004)andinducesconsiderablephenotypicplasticityinbodysize of ectothermic animals (Angilletta & Dunham, 2003). Generally, growthoffishincreaseswithincreaseintemperaturetoaspecies- specific optimum value, decreasing thereafter (Wootton, 2012). Optimaltemperatureforgrowthmaychangewithageandsize,as juvenilesgenerallypreferhighertemperaturesthanadults(Pedersen &Jobling,1989). Teleost fish species display predominantly female-biased SSD (Webb&Freckleton,2007).Asthegonadsizeoffemalesgenerallyincreasesmorerapidlywithsizethanthatofmales(Henderson,Trivedi, &Collins,2000),largefemalesareespeciallyimportantinpopulation- levelreproduction(Olinetal.,2012;Venturellietal.,2010).According tothefecundityadvantagehypothesis(Darwin,1871;Shine,1978), female-biasedSSDisduetoselectionfavoringalargebodysizeto ensurehigherreproductivesuccess,whichalsoleadstotheinverseof Rensch’srule(Fairbairn,1997). Here,weanalyzethelatitudinalvariationinsexualdimorphism inlife-historytraits inEuropeanperch(Perca fluviatilis L.) byusing a comprehensive field data from core distribution (50ºN) to the northerndistributionlimit(69ºN).Perchisoneofthemostcommon freshwater fish species across Europe (38–69°N), inhabiting lentic habitatsfrompondstothelargestlakes(Kottelat&Freyhof,2007). Itisacool-waterspring-spawningspecieswithanoptimumgrowing temperatureofca.23°C(Mélard,Kestemont,&Grignard,1996)anda maximumlengthof60cm,butmoretypicallyattainingalengthofup to25cmdependingonlakeandpopulationtype(Kottelat&Freyhof, 2007).Thegenerallife-historytraitsofpercharewellknownanddocumentedinseveralpapers (Heibo,Magnhagen,&Vøllestad,2005; LeCren,1951;Thorpe,1977).Perchdisplaysfemale-biasedsexual dimorphism in size, growth, and maturation (Heibo & Magnhagen, 2005;Mélardetal.,1996),butthesex-specificlatitudinalpatterns arelargelyunknown. Weexpectedtoconfirmsexualdimorphismofgrowthandmaturityofperchatallstudiedlatitudes.Astheenergydemandoffemale perchishigherthanthatofmales(Malison,Best,Kayes,&Amundson, 1985),inadditiontothehighersensitivityoffemalestothermalvariation(Estlanderetal.,2015),weexpectedfemalestoshowasteeper latitudinalvariationingrowth.Thus,thisfollowsthatsex-specificdifferencesin growthand sizewoulddecreasetowardhigherlatitude andwould produce a pattern of SSD contrary to the prediction of Rensch’srule.Finally,possibleexplanationsforobservedpatternsof SSDarediscussed. | 667 ESTLANDER ET AL. 2 | MATERIALS AND METHODS 2.1 | Sampling Datawerecollectedfrom25wildperchpopulationsalongalatitudinal gradient(50°–69°)withmultimeshgillnets(EuropeanStandardGillnet SamplingEN14757;meshrange5.25–60mm)during2000–2012at theendofthegrowingseason(August-September)(Table1).Bothlittoralandpelagialweresampledtoassessthe putativepresenceof divergentperchmorphs(Svanbäck&Eklöv,2003).However,wedid notfindsignsofperchpopulationdivergence.Thedataincluded2736 individuals:1139males(42%)and1597females(58%)(Table1).At latitudes from 69 to 50°N, the annual averagewater temperatures increasefrom2to9°C(Straškraba,1980)andthelengthofthegrowing season from 110 to 190days (Rötzer & Chmielewski, 2001). Latitude, longitude, altitude, lake surface area, and total phosphorusconcentrationweremeasuredfromallstudylakes.Waterquality parameters,suchaswatertransparency,thatis,Secchidepth(averageofalllakes,3.6m±2.2SD)andpH(6.9±0.3m)varied,butwere notstatisticallysignificantbetweenlatitudes(p<.05).Allthesampled lakesaremultispecies,andspeciesdiversityandfishdensityvaryboth betweenlakesandlatitudes. 2.2 | Length increment, maturity, and sexual size dimorphism Sex,totallength(accuracy1mm),andweight(0.1g)weremeasured, andoperculawerecleanedforageandback-calculatedgrowthdeterminations(Bagenal&Tesch,1978).Thelengthorageatmaturity datawasnotavailableforallofthelakesstudied,butlengthandage atmaturityareknowntocorrelatepositivelywithlatitudeinperch (Heibo,2003;Heiboetal.,2005).Asthedataanalyzedhereshowed asimilarpattern,ageatmaturitywasestimatedaccordingtoHeibo (2003)bylinearregression:ageatmaturity=−1.2+0.04×latitude; R2=.59,p < .005,andthesewereusedtoassessthelengthatmaturityfromcurrentdata.Theback-calculatedgrowthofperchwasdeterminedfromtheotolithoroperculumboneforeachindividualusing theMonastyrskymethod:nonlinearrelationshipbetweentheotolith/ operculumradiusandtotallengthofthefish(Bagenal&Tesch,1978): Li = (Si/Sc)b × Lc, where Li, Si=lengthoffishatformationofi:thradiusorradiusatage i; Lc, Sc=lengthoffishorradiusatthetimeofcapture;andb =growth coefficienti.e.theslopeoftherelationshipbetweenotolith/operculumradiusandlength. The between-sex and latitudinal differences in the annual (a) lengthincrementsand(b)lengthatagewereanalyzedwithanalysis ofvarianceforrepeatedmeasures(ANOVAR),anddatausedwererestrictedtoagegroups1–6(n=2004),becauseolderfishwererareor absentfromsouthernpopulations(50°).IntheANOVARmodels,sex (twolevels)andlatitude(fourlevels)wereconsideredasfixedfactors. Mauchly’stestindicatedthattheassumptionofsphericityhadbeen violatedinbothmodels,(a) x2 (14) = 5699.44 ,p > .05;(b) x2 (14) = 332.44 , p > .05,andtherefore,degreesoffreedomwerecorrected(e.g.,Field, TABLE1 Summaryoflatitudinalperchmeansizeandage(±SD)datausedinanalyses(CZE=CzechRepublic,SFI=SouthernFinland,CFI=CentralFinland,NFI=NorthernFinland).Lake characteristicvaluesrepresentthemeanvaluesperlatitude,witharangeinparentheses Population origin Lake information Fish data metrics Latitude (ºN) Longitude (ºE) Altitude (m a.s.l.) Populations (n) Lake size (ha) Mean depth (m) TotP (μg L−1) Perch Individuals (n) Female (%) Total length (mm) Weight(g) Age (years) 50 16 240 4(CZE) 120(60˗250) 17(14˗23) 44(10˗80) 668 58 180±2 96±5 4±0.6 60 25 130 6(SFI) 270(20˗700) 5(3˗6) 21(5˗40) 516 60 150±2 51±3 3±0.9 63 29 150 6(CFI) 230(20˗470) 4(4˗5) 10(5˗14) 350 62 160±3 63±4 4±0.1 69 26 200 9(NFI) 13660(350˗104300) 8(3˗14) 7(4˗21) 1202 56 193±2 109±3 6±0.5 668 | ESTLANDER ET AL. 2013)usingGreenhouse–Geisserestimatesofsphericity(a)(p = .280); (b)(p = .872). Ratiosforsexualdimorphismindices(SDI)werecalculatedusing themethodofGibbons(1992): (A/B)−1, whereAisthemeansizeofthelargestsexandBisthemeansize ofthesmallestsex.Sex-specificmeansizesforSDIcalculationwere weighted by number of individuals within age groups (2–10years) whenlake-specificsubsamplesrepresentedtruelengthandagedistributionineachlake;thus,oneSDIvalueperpopulation(intotal25) wascalculated.Stepwisemultipleregressionswithforwardselection ofvariableswereusedtoidentifythemostimportantenvironmentalvariablesexplainingthevariationinSDIs.Environmentalvariables (latitude, longitude, altitude, lake surface area, and total phosphorusconcentration)wereenteredinthemultipleregressionanalysis, if p < .05. The statistical analyses were performed using IBM SPSS StatisticsforWindows,version21.0(IBMCorp.,Armonk,NY,USA). 3 | RESULTS 3.1 | Growth Thelatitudinalvariationsoflengthatage(ANOVAR;F[3,924]=6.92, p < .001)andlengthincrement(ANOVAR;F[3,924]=2.10,p = .010)of perchwere sex-dependent (Figure1).Overall, the growthof perch decreased with increase in latitude (ANOVAR; F[3,924] =131.07, p < .001),and thelengthat ageof femaleswaslargerthanthat of malesatalllatitudes(ANOVAR;F[1,924]=36.48,p < .001)(Figure1). However,thelengthatageandlengthincrementdifferencebetween sexesvaried,dependingonlatitudeandage(Figure1;Table2).Atlatitude50°N,thelengthatageforfemaleswaslargerthanformalesinall agegroups(ANOVAR;F[5,384]=14.99,p < .001)(Table2),andtheannuallengthincrementwashigherinfemalesthaninmales(ANOVAR; F[1,84] =14.38,p < .001),despitenosex-dependentdifferencesinage groups2 and 6(Table2; Figure1). Atlatitude 60°N, femaleswere overalllargerthanmales(ANOVAR;F[5,76] =5.54,p = .012),indicated bysex-dependentdifferencesinagegroups2–6(Table2;Figure1). No significant sex dependency on annual length increments in the latitude60ºNincrementwasdetectedwhenpoolingallagegroups (ANOVAR;F[1,76]=3.02,p = .086);however,asex-dependentdifferencewasobservedinagegroups4and5(Table2),withgrowthof femalesbeingfasterthanthatofmales(Figure1).Atlatitude63°N, femaleswerelarger(ANOVAR;F[5,106]=4.72,p = .02)andgrewfaster (ANOVAR;F[1,106]=4.57,p = .035)thanmaleswhenallagegroups werepooled,indicatedbysex-dependentdifferencesinlengthatage andlengthincrementintheolderagegroups(groups4–6)(Table2; Figure1).Whenallagegroupswerepooledatthenorthernmostlatitude69°N, females werelarger (ANOVAR; F[5,658]=5.26,p = .012) and grew faster (ANOVAR; F[1,658]=4.09, p = .043) than males (Figure1),indicatedbysex-dependentdifferencesinlengthatagein agegroups5and6(Table2)andtheannuallengthincrementinage groups1,3,5,and6(Table2).Ingeneral,theannuallengthincrement FIGURE1 Averageannualtotallengthincrements(top)andaveragetotallengthatage(bottom)withstandarddeviation(±SD)offemale (opencircles)andmale(blacktriangles)perchatlatitudes50–69°N Age (years) | 669 ESTLANDER ET AL. showed disparate latitudinal clines; the southernmost populations hadthefastestgrowthintheearlyyearsoflifeandthenorthernmost populationsinlateryears(Figure2a,b).Additionally,inthefirstyear oflife,agreaterproportionofoverallgrowthwasattainedbyfemales relativetomales,whereasinolderindividuals,theoppositetrendwas observed(Figure2a,b). Whenpoolingallagegroupsandbothsexes,themaximumlength of perch decreased with increase in latitude (420, 365, 344, and 332mmatlatitudes50,60,63,and69°N,respectively).Theaverage lengthvariedsignificantlybetweenlatitudes(ANOVA,F[3,2736] =95.59, p < .001),butwithoutaclearlatitudinaltrend(Table1). 3.2 | Longevity and maturity Theaverageagevariedsignificantlybetweenalllatitudes(ANOVA, F [3,2736]=177.53; p = .001), as the oldest fish were found at latitude 69°N and youngest at latitude 60°N (Table1). Males were 1.2–1.5years younger than females at latitudes 60, 63, and 69°N (ANOVA;F[3,2728]=17.69,p < .001),butatlatitude50°N,theaverage ageoffemalesandmaleswasthesame.Theageatmaturityincreased from 2 to 5years with increase in latitude, and the corresponding lengthatmaturityincreasedca.50mmforbothmales(126–180mm) andfemales(138–190mm)(Table3). 3.3 | Sexual size dimorphism Inthestepwisemultipleregressionmodel,otherenvironmentalfactorsfailedtoentertheregressionequationoncelatitudewasincluded (Table4).Thedegreeofsexualsizedimorphismofperchdecreased withincreaseinlatitude,andlatitudesignificantlypredictedSD indices(Figure3)(R2 =.84;F = 93.33;p < .001),indicatingthatthesizedifferencebetweensexeswasinverselyassociatedwithlatitude. 4 | DISCUSSION Asexpected,allperchpopulationsstudiedexhibitedsignificantsexual dimorphismingrowth,size,andmaturity,withfemalesgrowinglarger andmaturinglaterthanmales.Growthofbothsexesdecreasedand thelengthatmaturityincreasedwithlatitude,butlatitudinaltrends were generally steeper in females than in males. Accordingly, the magnitude of SSD diminished in concert with increase in latitude, suggesting stronger sensitivity of females to latitudinal variation, becausefemalebodysizeshowedanincreasedplasticityrelativeto males.Thus,perchdidnotfollowRensch’sruleinthepresentstudy, butshowedtheexactconversepattern.Incontrast,ourresultsare consistent with the conception that growth response can be sex- specific to environmental conditions (Bonduriansky, 2007; Stillwell, Blanckenhorn,Teder,Davidowitz,&Fox,2010).Inaddition,studies thatdescribetheinverseofRensch’srule(e.g.,Fairbairn,1997)suggest that SSD results from fecundity selection favor larger female size.Thislikelyholdsalsoforperch,becausethefecundityofperch increaseswithfemalebodysize(Olinetal.,2012). Several factors inducing gender-specific differences in growth havebeenproposed,includingenergyallocation,risk-taking,andvulnerabilitytopredatorsandparentalinvestment(Rennieetal.,2008). Morerecently,sex-specificdifferencesofperchhavebeenobserved ingutmicrobiotalinkedtodistinctdietarypreferences(Bolnicketal., 2014)andmaythushavefurtherimplicationsofenergyroutingand individualmetabolism.Duetothedifferentdemandsforenergyacquisition,malesandfemalesmayhavevariablestrategiesfortrade-offs between food acquisition and prevailing environmental conditions (Holtby & Healey, 1990). In optimal environmental conditions (e.g., clear water for foraging, optimal temperature, low predation pressure),femalesinvestinactivefeedingtoensuresomaticgrowthand latergonadosomaticgrowth,whereasmalesdonotneedtoinvestas muchinfeedingandfastgrowth,asspermislessenergy-demanding toproduce(Rennieetal.,2008).Therefore,changesinoptimalfeeding conditionshavethemostpronouncedeffectsonthemostactivefeeders,which are often females (Estlander etal., 2015; Horppila etal., 2011),whereasmalesneedtogrowonlytosizeatsexualmaturity. Consequently,contradictingwithRensch’srulewhichimpliesthatsexualselectionisthemaindrivingforceunderlyingSSD(Fairbairn,2005), wesuggestthatdifferentforcesbeyondsexualselection,suchassex- specificresponsestovariationinenvironmentalconditions,maybe alsoresponsibleforshapingSSDpatternsinperch. InadditiontolatitudinalvariationinSSD,perchdisplayedanoveralldecreasinggrowthintermsoflengthincrementwithincreasein latitudeirrespectiveofsex.ThisisinlinewithHeiboetal.(2005),who suggestedthatperchfollowtheconversepatternofBergmann’srule TABLE2 p-valuesfromrepeatedmeasurementsanalysisofvarianceinbetween-sexcomparisonsofannualtotallengthincrementsand lengthatageofperchatlatitudesof50–69°N.Significantvalues(p<.05)areinbold Age (years) and sample size Latitude (°N) 50° 60° 63° 69° 50° 60° 63° 69° Annuallengthincrements Lengthatspecificage 1(n=292) 0.0323 0.6873 0.1878 0.0158 0.0323 0.6873 0.1878 0.2050 2(n=293) 0.2492 0.4996 0.6628 0.2121 0.0224 <0.0001 0.2193 0.6220 3(n=316) <0.0001 0.1538 0.2625 0.0473 0.0007 <0.0001 0.1484 0.2850 4(n=485) 0.0009 0.0274 0.0220 0.1588 <0.0001 <0.0001 0.0436 0.1190 5(n=350) 0.0026 0.0213 0.0232 0.0190 <0.0001 0.0010 0.0383 0.0406 6(n=260) 0.0735 0.1568 0.0063 0.0058 <0.0001 0.0290 0.0132 0.0105 670 | ESTLANDER ET AL. ingrowth.Manyanimaltaxa,suchasbirdsingeneral,haveastrongintraspecifictendencytowardlargerbodysizesathigherlatitudesandin coolerenvironments(Ashton,2002),incontrasttomanyectotherms (Angilletta,Steury,&Sears,2004).Corroboratingourresults,Heibo etal.(2005)foundmanylife-historyvariables,suchaslengthatage andlengthincrementtodecreaseandageatmaturitytoincreasewith latitude.Thisisattributedtothelatitudinalclineintemperatureand durationofthegrowingseason.Alsosupportingourresults,theasymptoticbodylengthdidnotsimilarlycorrelatewithlatitude.Inour data,thegrowthofperchdecreasedwhiletheaveragelengthandage increasedalonglatitude,suggestinga greaterlongevityofnorthern populations.Largesizecanbetheresultofgreaterlongevity,ifmortalityislowevenwithrelativelyslowgrowth(Angillettaetal.,2004). Incoolerclimatesathigherlatitudes,fishmayinvestmoreinsomatic growthtoreachalargersizeattheexpenseofgonadgrowth,suggestingatrade-offbetweenindividualenergyallocations.Year-class strengthofperchpopulationsatdistribution limitis alsoknownto behighlydependentontemperature(Hayden,Harrod,&Kahilainen, 2014;Tolonen,Lappalainen,&Pulliainen,2003)thatmayalsopromote subsequentgrowthofsingleyearclasstolargesizeinmultispecies communities.Ingeneral,perchpopulationsconsistoflargersizedindividualsinhigherlatitudes(Jeppesenetal.,2010),suggestingthatthe population-levelshifttopiscivorymaybemorefrequentthaninlower latitudesormerelyreflectslowertemperature-relatedmetaboliccosts andthushigherlongevityinnorth.Also,otherbioticfactors,notconsideredhere,suchasavailablefoodresources,intraspecificcompetitionandinterspecificcompetitionaffectgrowthofperch.However, latitude(temperature,durationofthegrowingseason,productivity) directlyandindirectlyregulatesseveralabioticandbioticfactorsand thereforepotentiallyalsoaffectthetrophicinteractionsbetweenspecies(Jeppesenetal.,2010).Accordingly,ourresultssuggestthatin additiontosex-specificsensitivitytoenvironment,variationinSSD couldresultfromsex-specificdifferencesinlongevity,agestructure, ordifferencesindiet.Furtherfieldandexperimentalstudiescombiningsex-specificdietary,sizestructure,andlife-historytraitdataare neededtoassessthesepatterns. SomestudieshavesuggestedthatdifferenttimingofmaturitybetweenthesexesmayberesponsibleforthelevelofSSD exhibited byaspecies(Fairbairn,1990;Gibbons&Lovich,1990).Thesestudies implythatjuvenilegrowthratesbetweenthesexesaresimilar,and theearliermaturingsexremainssmallerthanthelatermaturingsex (Badyaev,2002).Inthisstudy,malesmaturedearlierthanfemales,a patterncommoninfish,asfemalesincreasetheirfecunditywithsize, butreproductivesuccessinmalesisnotassize-dependent(Stearns, 1992).However,Blanckenhornetal.(2007)suggestedinastudywith FIGURE2 Proportionofoverallgrowth (totallengthinterment)offemale(open circles)andmale(blacktriangles)perchin thefirstyear(a)andfifthyear(b) 0 10 20 30 40 50 60 63 69 (a) 0 5 10 15 50 60 63 69 (b) Proportion of overall growth Latitude (°N) TABLE3 Ageatmaturity(±SD)atdifferentlatitudesestimated accordingtoHeibo(2003)bylinearregressionandthecorresponding totallengthsfortheseagesanalyzedfromdata Latitude (°N) Age at maturity (years) Length at maturity (mm) Males Females 50° 2±0.18 131±17 138±21 60° 3±0.21 126±21 154±24 63° 4±0.27 142±20 164±25 69° 5±0.18 180±38 190±41 TABLE4 Stepwisemultipleregressionmodel(factorsincluded when p<.05)forsexualsizedimorphism.Theonlyfactorselected waslatitude βt p Latitude 0.731 13.561 .0001 Longitude −0.018 −0.142 .889 Altitude −0.146 −1.614 .122 Lakesize −0.149 −1.466 .158 Totalphosphorus −0.214 −1.762 .093 FIGURE3 Latitude-specificdegreeofsexualsizedimorphism inperchpopulations(SDI=−0.01×latitude+0.731,R2=.84, p<.0001) 0.0 0.1 0.2 0.3 45 50 55 60 65 70 SDI Latitude (°N) | 671 ESTLANDER ET AL. insectsthatSSDismorelikelyrelatedtodifferentialgrowthratesbetweenthesexesthatmaybedifferentlyconstrainedbygrowthconditionswhenattainingtheiroptimalbodysizesandthelargersexshows strongerresponsetoareductioninenvironmentalquality.Theresults ofourresearchsupportthissuggestion,becauseperchshowedasex- dependentdifferenceinsizeand back-calculatedgrowthalreadyat thejuvenilestage,asfemalesappearedtoinvestmoreingrowthinthe earlyyearsoflife.Fastergrowingfisharealsomorelikelytoshiftto piscivorythatmaybemoreimportantforfemalesbenefitingonlarger maturitysizemorethanmales.Thiscouldbeanimportantmechanism tounderstandbimodalsizestructureofperchpopulationsaswellas preyfishcommunities,butremainstobeevaluatedinexperimental andfieldstudies.Therefore,wearguethatgrowthrate,ratherthan timingofmaturity,maybeamoresignificantfactorbehindSSDvariationinperch.Itmustbenoted,however,thatitisdifficulttorank thesetwofactorsinorderofimportance,astheselife-historytraitsare highlycorrelated(Stearns,1992). Thereisalinearrelationshipwithlatitudeandtemperature,and risingtemperatureacceleratesgrowthandearliermaturity(Berrigan &Charnov,1994;Heiboetal.,2005),demonstratedalsointhisstudy, asthematurationlengthandageatmaturityincreasedwithlatitude inboth sexes.Typically,delayedmaturationprovidesa benefit,becausefecundityincreaseswithbodysize(Roff,2002;Stearns,1992). AccordingtoHeiboetal.(2005),themaximumreproductivelifespan, thatis,higherlongevityincreaseswithlatitudeinperch,butreproductiveinvestment(measuredasrelativegonadmass)ineachspawning seasondecreaseswithlatitude.Suchlife-historystrategyisbeneficial, ifmortalityislow,thatis,reproductivelifespanofbothsexesislong. Similar findings are suggested also in coho salmon (Oncorhynchus kisutch)astheeggnumberincreases,buttheindividualeggsizeand thetotaleggbiomassdecreasewithlatitude,andthus,atthepopulationlevel,thegameticeffortmaybeconstantwithlatitude(Tamate& Maekawa,2006). Overall,theregulationofSSDisacomplicatedissueandwidely acceptedecologicalrulessuchasRensch’sandBergmann’srulesdescribingthepatternsinbodysizearenotstraightforwardlyapplicable infishspecieswithfemale-biasedSSD.Forinstance,eveniffecundity selectionwouldbetheultimatecausebehindtheSSDoffish,several environmentalfactorsmightregulatethemagnitudeofSSD,suchas latitudethatpotentiallyregulatesacomplexmixofenvironmentaland ecologicalfactors.Moreover,sensitivitytothesefactorsmayvarybetweensexes.Inaddition,itisdifficulttodistinguishwhetherphenotypicchangesingrowthorsize/agestructurealongalatitudegradient arearesultofadaptiveevolutionorphenotypicplasticityoramixture ofboth(e.g.,Kuparinen&Merilä,2007),andthus,moreresearchis needed,suchascommongardenexperiments,tobetterunderstand both the phenotypic and genetic relationships between SSD and growth.Ourresearchdemonstrateslatitudinalpopulation-levelvariationinthemagnitudeofSSDbasedongrowthrate,supportingthe predictionsofpreviousexperimentalandregionalstudiesbyFontaine, Gardeur,Kestemont,andGeorges(1997),Horppilaetal.(2011),and Estlanderetal.(2015),allsuggestingthatenvironmentalfactorslimitingoverallgrowthmaydecreasethemagnitudeofSSD.Thisstudy alsohighlightstheimportanceofsex-specificresponsedifferencesto environmentalvariablesinregulatingpatternsofallometrybetween thesexesinfish.Ingeneral,understandingthecausesbehindbody sizevariationisparticularlyimportantinfishasitisrelatedtofecundity and survival. Our results of growth and sexual maturity of an abundantfishinEuropeanlakessuggestthatsexhasanimportantrole indetermininglife-historytraits,butmayhaveimplicationsonindividualmetabolism,predator–preyrelationships,andsizestructuring offishpopulationsinlakes.Weconcludethatfollow-upstudiesfrom individualtoecosystemlevelscaleareneededtoassesspotentially holisticconsequencesofsexualsizedimorphism. ACKNOWLEDGMENTS The study was financially supported by three-year research grants fromtheUniversityofHelsinki,theR.ErikSerlachiusFoundation,and theBergsrådetBrorSerlachiusFoundation.Wethankallofthepeople whoparticipatedinfieldsamplingandlaboratoryanalysesoffish. CONFLICT OF INTEREST Nonedeclared. REFERENCES Abouheif,E.,&Fairbairn,D.J.(1997).Acomparativeanalysisofallometry for sexual size dimorphism: Assessing Rensch’s rule. American Naturalist,149,540–562. Adams,D.C.,&Church,J.O.(2008).AmphibiansdonotfollowBergmann’s rule. Evolution,62,413–420. Angilletta,M.J.Jr,&Dunham,A.E.(2003).Thetemperature-sizerulein ectotherms: Simple evolutionary explanations may not be general. American Naturalist,162,332–342. Angilletta,M.J.,Steury,T.D.,&Sears,M.W.(2004).Temperature,growth rate,andbodysizeinectotherms:Fittingpiecesofalife-historypuzzle. Integrative and Comparative Biology,44,498–509. Arnold, W., Ruf,T., & Kuntz, R. (2006). Seasonal adjustment of energy budgetinalargewildmammal,thePrzewalskihorse(Equus ferus przewalskii) II. Energy expenditure. Journal of Experimental Biology, 209, 4566–4573. Ashton,K.G.(2002).Patternsofwithin-speciesbodysizevariationofbirds: StrongevidenceforBergmann’srule.Global Ecology and Biogeography, 11,505–523. Ashton,K.G.,&Feldman,C.R.(2003).Bergmann’sruleinnonavianreptiles: Turtles follow it, lizards and snakes reverse it. Evolution, 57, 1151–1163. Badyaev,A.V.(2002).Growingapart:Anontogeneticperspectiveonthe evolutionofsexualsizedimorphism.Trends in Ecology & Evolution,17, 369–378. Bagenal,T.B.,&Tesch,F.W.(1978).Ageandgrowth.InT.B.Bagenal(Ed.), Methods for Assessment of Fish Production in Fresh Waters (pp. 101– 136.).Oxford:BlackwellScientificPublication. Bergmann, C. (1848). Über die Verhältnisse der Wärmeökonomie der ThierezuihrerGrösse.Gollinger studien,3,595–708. Berrigan,D.,&Charnov,E.L.(1994).Reactionnormsforageandsizeat maturityinresponsetotemperature:Apuzzleforlifehistorians.Oikos, 3,474–478. Blanckenhorn, W. U. (1998). Adaptive phenotypic plasticity in growth development and body size in the yellow dung fly. Evolution, 52, 1394–1407. 672 | ESTLANDER ET AL. Blanckenhorn, W. U., & Demont, M. (2004). Bergmann and converse Bergmannlatitudinalclinesinarthropods:Twoendsofacontinuum? Integrative and Comparative Biology,44,413–424. Blanckenhorn, W. U., Meier, R., &Teder,T. (2007). Rensch’s rule in insects: Patterns among and within species. In D. J. Fairbairn, W. U. Blanckenhorn&T.Szekely(Eds.),Sex, Size and Gender Roles Evolutionary Studies of Sexual Size Dimorphism(pp.60–70.).Oxford:UniversityPress. Blanckenhorn,W.U.,Stillwell,R.C.,Young,K.A.,Fox,C.W.,&Ashton,K.G. (2006).WhenRenschmeetsBergmann:Doessexualsizedimorphism changesystematicallywithlatitude?Evolution,60,2004–2011. Bolnick,D.I.,Snowberg,L.K.,Hirsch,P.E.,Lauber,C.L.,Org,E.,Parks, B.,…Svanbäck,R.(2014).Individualdiethassex-dependenteffects onvertebrategutmicrobiota.Nature Communications,5,doi:10.1038/ ncomms5500 Bonduriansky, R. (2007). The evolution of condition-dependent sexual dimorphism.American Naturalist,169,9–19. Brown,J.H.,Gillooly,J.F.,Allen,A.P.,Savage,V.M.,&West,G.B.(2004). Towardametabolictheoryofecology.Ecology,85,1771–1789. Cox,R.M.,Barrett,M.M.,&John-Alder,H.B.(2008).Effectsoffoodrestrictionongrowth,energyallocation,andsexualsizedimorphismin Yarrow’sspinylizard,Sceloporus jarrovii. Canadian Journal of Zoology,86, 268–276. Darwin,C.R.(1871).Thedescentofman,andselectioninrelationtosex. London.JohnMurray. Estlander,S.,Nurminen,L.,Mrkvička,T.,Olin,M.,Rask,M.,&Lehtonen,H. (2015).Sex-dependentresponsesofperchtochangesinwaterclarity andtemperature.Ecology of Freshwater Fish,24,544–552. Fairbairn,D.J.(1990).Factorsinfluencingsexualsizedimorphismintemperatewaterstriders.American Naturalist,136,61–86. Fairbairn,D.J.(1997).Allometryforsexualsizedimorphism:Patternand processinthecoevolutionofbodysizeinmalesandfemales.Annual Review of Ecology and Systematics,28,659–687. Fairbairn,D.J.(2005).Allometryforsexualsizedimorphism:Testingtwo hypotheses for Rensch’s rule in the water strider Aquarius remigis. American Naturalist,166,69–84. Field,A. (2013). Discovering Statistics Using IBM SPSS Statistics. London: Sage. Fontaine,P.,Gardeur,J.N.,Kestemont,P.,&Georges,A.(1997).Influence offeedinglevelongrowth,intraspecificweightvariabilityandsexual growthdimorphismofEurasianperchPerca fluviatilisL.rearedinarecirculationsystem.Aquaculture,157,1–9. Gibbons,J.W.(1992).Areviewoftechniquesforquantifyingsexualsize dimorphism.Growth, Development & Aging,56,269–281. Gibbons,J.W.,&Lovich,J.E.(1990).Sexualdimorphisminturtleswith emphasis on the slider turtle (Trachemys scripta). Herpetological Monographs,4,1–29. Hayden,B.,Harrod,C.,&Kahilainen,K.K.(2014).Lakemorphometryand resource polymorphism determine niche segregation between cool andcold-water-adaptedfish.Ecology,95,538–552. Heibo, E. (2003). Life-history variation and age at maturity in Eurasian perch(Perca fluviatilisL.).Doctoralthesis,UniversityofUmeå,Sweden. Heibo,E.C.,&Magnhagen,C.(2005).Variationinageandsizeatmaturity inperch(Perca fluviatilisL.),comparedacrosslakeswithdifferentpredationrisk.Ecology of Freshwater Fish,14,344–351. Heibo,E.,Magnhagen,C.,&Vøllestad,L.A.(2005).Latitudinalvariationin life-historytraitsinEurasianperch.Ecology,86,3377–3386. Henderson,B.A.,Trivedi,T.,&Collins,N.(2000).Annualcycleofenergy allocationtogrowthandreproductionofyellowperch.Journal of Fish Biology,57,122–133. Holtby,L.B.,&Healey,M.C.(1990).Sex-specificlifehistorytacticsand risk-takingincohosalmon.Ecology,71,678–690. Horppila,J.,Estlander,S.,Olin,M.,Pihlajamäki,J.,Vinni,M.,&Nurminen, L.(2011).Gender-dependenteffectsofwaterqualityandconspecific densityonthefeedingrateoffish-factorsbehindsexualgrowthdimorphism.Oikos,120,855–861. James,F.C.(1970).Geographicsizevariationinbirdsanditsrelationshipto climate.Ecology,51,365–390. Jeppesen, E., Meerhoff, M., Holmgren, K., Gonzalez-Bergonzoni, I., Teixera-deMello,F.,Declerck,S.A.J.,...Conde-Porcuna,J.M.(2010). Impactsofclimatewarmingonlakefishcommunitystructureandpotentialeffectsonecosystemfunction.Hydrobiologia,646,73–90. Kottelat,M.,&Freyhof,J.(2007).Handbook of European Freshwater Fishes. Cornol,Switzerland:PublicationsKottelat. Kuparinen, A., & Merilä, J. (2007). Detecting and managing fisheries- inducedevolution.Trends in Ecology & Evolution,22,652–659. LeCren,E.D.(1951).Thelength-weightrelationshipandseasonalcycle ingonadweightandconditionintheperch(Perca fluviatilis).Journal of Animal Ecology,2,201–219. Malison,J.,Best,C.,Kayes,T.,&Amundson,C.(1985).Hormonalgrowth promotionandevidencefora size-relateddifferenceinresponseto estradiol17β in yellow perch (Perca flavescens). Canadian Journal of Fisheries and Aquatic Sciences,42,1627–1633. Mandiki,S.N.M.,Houbart,M.,Babiak,I.,Vandeloise,E.,Gardeur,J.N.,& Kestemont,P.(2004).Aresexsteroidsinvolvedinthesexualgrowth dimorphism in Eurasian perch juveniles? Physiology & Behavior, 80, 603–609. Meiri,S.(2011).Bergmann’sRule-what’sinaname?Global Ecology and Biogeography,20,203–207. Mélard,C.,Kestemont,P.,&Grignard,J.C.(1996).IntensivecultureofjuvenileandadultEurasianperch(P. fluviatilis):Effectofmajorbioticand abioticfactorsongrowth.Journal of Applied Ichthyology,12,175–180. Miaud,C.,Andreone,F.,Ribéron,A.,DeMichelis,S.,Clima,V.,Castanet,J., …Guyétant,R.C.(2001).Variationsinage,sizeatmaturityandgestationdurationamongtwoneighbouringpopulationsofthealpinesalamander(Salamandra lanzai).Journal of Zoology,254,251–260. New,M.,Hulme,M.,&Jones,P.(1999).Representingtwentieth-century space-timeclimatevariability.PartI:Developmentofa1961-90mean monthlyterrestrialclimatology.Journal of Climate,12,829–856. Olin,M.,Jutila,J.,Lehtonen,H.,Vinni,M.,Ruuhijärvi,J.,Estlander,S.,… Lappalainen,J.(2012).Importanceofmaternalsizeonthereproductivesuccessofperch,Perca fluviatilis,insmallforestlakes:Implications for fisheries management. Fisheries Management and Ecology, 19, 363–374. Pedersen,T.,&Jobling,M.(1989).Growthratesoflarge,sexuallymature cod,Gadus morhua,inrelationtoconditionandtemperatureduringan annualcycle.Aquaculture,81,161–168. Pörtner,H.O.,&Peck,M.(2010).Climatechangeimpactsonfishandfisheries:Towardsacauseandeffectunderstanding.Journal of Fish Biology, 77,1745–1779. Rennie,M.D.,Purchase,C. F., Lester, N., Collins,N.C., Shuter,B.J.,& Abrams,P.A.(2008).Lazymales?Bioenergeticdifferencesinenergy acquisitionandmetabolismhelptoexplainsexualsizedimorphismin percids.Journal of Animal Ecology,77,916–926. Rensch,B.(1950).DieAbhängigkeitderrelativenSexualdifferenzvonder Körpergrösse.Bonner Zoologische Beiträge,1,58–69. Roff,D.A.(2002).Lifehistoryevolution.SinauerAssociates. Rötzer, T., & Chmielewski, F. M. (2001). Phenological maps of Europe. Climate Research,18,249–257. Rypel,A.L.(2014).Thecold-waterconnection:Bergmann’sruleinNorth Americanfreshwaterfishes.The American Naturalist,183,147–156. Shine, R. (1978). Sexual size dimorphism and male combat in snakes. Oecologia,33,269–277. Shine,R.(1989).Ecologicalcausesfortheevolutionofsexualdimorphism: Areviewoftheevidence.Quarterly Review of Biology,64,419–461. Stearns, S. C. (1992). The Evolution of Life Histories. Oxford: Oxford UniversityPress. Stillwell,R. C.,Blanckenhorn,W. U.,Teder,T.,Davidowitz,G.,&Fox,C. W.(2010).Sexdifferencesinphenotypicplasticityaffectvariationin sexualsizedimorphismininsects:Fromphysiologytoevolution.Annual Review of Entomology,55,227–245. | 673 ESTLANDER ET AL. Straškraba,M.(1980).Theeffectsofphysicalvariablesonfreshwaterproduction:Analysesbasedonmodels.InE.D.LeCren,&R.H.McConnell (Eds.), Functioning of Freshwater Ecosystems (pp. 13–84). Cambridge: CambridgeUniversityPress. Svanbäck,R.,&Eklöv,P.(2003).Morphologydependentforagingefficiency inperch:Atrade-offforecologicalspecialization?Oikos,102,273–284. Tamate,T.,&Maekawa,K.(2006).Latitudinalvariationinsexualsizedimorphismofsea-runMasusalmon,Oncorhynchus masou. Evolution,60, 196–201. Thorpe, J. E. (1977). Morphology, physiology, behavior, and ecology of Perca fluviatilis L. and P. flavescens Mitchill. Journal of the Fisheries Research Board of Canada,34,1504–1514. Tolonen,A.,Lappalainen,J.,&Pulliainen,E.(2003).Seasonalgrowthand yearclassstrengthvariationsofperchnearthenorthernlimitsofits distributionrange.Journal of Fish Biology,63,176–186. Vázquez,D.P.,&Stevens,R.D.(2004).Thelatitudinalgradientinniche breadth:Conceptsandevidence.American Naturalist,164,E1–E19. Vedder,O.,Dekker,A.,Visser,G.,&Dijkstra,C.(2005).Sex-specificenergy requirementsinnestlingsofanextremelysexuallysizedimorphicbird, the European sparrow hawk (Accipiter nisus). Behavioral Ecology and Sociobiology,58,429–436. Venturelli,P.A.,Murphy,C.,Shuter,B.J.,Johnston,T.A.,VanCoeverden deGrootP.J.,Boag,P.T.,…Leggett,W.C.(2010).Maternalinfluences onpopulationdynamics:Evidencefromanexploitedfreshwaterfish. Ecology,91,2003–2012. Webb,T.J.,&Freckleton,R.P.(2007).Onlyhalfright:Specieswithfemale- biasedsexualsizedimorphismconsistentlybreakRensch’srule.PLoS One,2,e897. Wootton, R. (2012). Ecology of teleost fishes. Springer Science and BusinessMedia Young,K.A.(2005).Life-historyvariationandallometryforsexualsizedimorphisminPacificsalmonandtrout.Proceedings of the Royal Society of London B: Biological Sciences,272,167–172. How to cite this article:Estlander,S.,Kahilainen,K.K., Horppila,J.,Olin,M.,Rask,M.,Kubečka,J.,Peterka,J.,Říha,M., Huuskonen,H.,andNurminen,L.(2017),Latitudinalvariation insexualdimorphisminlife-historytraitsofafreshwaterfish. EcologyandEvolution,7:665–673.doi:10.1002/ece3.2658