Citation
Interspecific Competition as a Driver of Juvenile Spiny Lobster Abundance and Distribution
Permanent Link:

Material Information

Title:
Interspecific Competition as a Driver of Juvenile Spiny Lobster Abundance and Distribution
Creator:
Hart, John E
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (10 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Interdisciplinary Ecology
Committee Chair:
BEHRINGER,DONALD CHARLES,JR
Committee Co-Chair:
LINDBERG,WILLIAM J
Committee Members:
ST MARY,COLETTE MARIE
Graduation Date:
5/3/2014

Subjects

Subjects / Keywords:
Chemicals ( jstor )
Crabs ( jstor )
Ecology ( jstor )
Juvenile stages ( jstor )
Juveniles ( jstor )
Lobsters ( jstor )
Odors ( jstor )
Shelters ( jstor )
Species ( jstor )
Spiders ( jstor )
Interdisciplinary Ecology -- Dissertations, Academic -- UF
competition -- hard-bottom -- lobster -- shelter
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Interdisciplinary Ecology thesis, M.S.

Notes

Abstract:
Interspecific competition can strongly influence the population dynamics of competing species. However, the importance of competition and how to test for it has been vigorously debated. For economically important species such as the Caribbean spiny lobster (Panulirus argus) this is an important fishery management consideration. Anecdotal observations have pointed to an inverse relationship in abundance between juvenile stone crab (Menippe mercenaria) and juvenile spiny lobsters in the hard-bottom habitat of the Florida Keys. It is primarily during their vulnerable juvenile stages that these species have similar shelter requirements. I explored this relationship in mesocosm experiments to determine the competitively dominant species and field experiments to determine the effect of stone crab abundance on lobster abundance and distribution. My results showed that stone crabs are the dominant competitors regardless of the number of lobsters present, the presence of co-sheltering species such as the spider crab (Mithrax spinosissimus), or the order of introduction of competitors into the mesocosm. We also found that lobsters use chemical cues from stone crabs to detect and avoid them. Our manipulations of stone crab abundance demonstrated that increased stone crab abundance resulted in decreased lobster abundance and increased aggregation. The opposite occurred on stone crab removal sites. Our study suggests that stone crabs can limit the availability of shelter to lobsters, potentially increasing lobster mortality or driving them to emigrate from the area. If shelter is extremely limited or stone crab recruitment is high, competition may contribute to a population bottleneck for adult lobster populations that could reduce recruitment into the fishery. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2014.
Local:
Adviser: BEHRINGER,DONALD CHARLES,JR.
Local:
Co-adviser: LINDBERG,WILLIAM J.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-05-31
Statement of Responsibility:
by John E Hart.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
5/31/2015
Resource Identifier:
908645564 ( OCLC )
Classification:
LD1780 2014 ( lcc )
Downloads

This item has the following downloads:
Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EPML9G9QV_5LXDM7 INGEST_TIME 2014-10-03T22:49:29Z PACKAGE UFE0046833_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

Ecology ,95(1),2014,pp.68…77 2014bytheEcologicalSocietyofAmericaDelayedlifehistoryeffects,multilevelselection,andevolutionary trade-offsintheCaliforniatigersalamanderCHRISTOPHERA.SEARCY,1,5LEVIN.GRAY,2PETERC.TRENHAM,3ANDH.BRADLEYSHAFFER41DepartmentofEvolutionandEcologyandCenterforPopulationBiology,UniversityofCalifornia,OneShieldsAvenue,Davis, California95616USA2DepartmentofBiology,UniversityofNewMexico,Albuquerque,NewMexico87131USA3BiologyDepartment,WesternWashingtonUniversity,Bellingham,Washington98225USA4DepartmentofEcologyandEvolutionaryBiology,621CharlesE.YoungDriveSouthandLaKretzCenterforCalifornia ConservationScience,619CharlesE.YoungDriveSouth,UniversityofCalifornia,LosAngeles,California90095USAAbstract .Delayedlifehistoryeffects(DLHEs)occurwhen“tnessinonelifestageaffects “tnessinsubsequentlifestages.Giventheirbiphasiclifecycle,pond-breedingamphibians provideanaturalsystemforstudyingDLHEs,althoughtheseeffectsarenotrestrictedto specieswithbiphasiclifehistories.Inthisstudy,weusedmultiplemark…recapturetechniques enabledbyalargetrappingarraytomonitorcomponentsof“tnessandresultingDLHEsina populationoftheendangeredCaliforniatigersalamander( Ambystomacaliforniense ).We foundthatDLHEsareprominentacrossalllifestagetransitionsandthatthereisvariationin whetherselectionactsprimarilyattheindividualorcohortlevel.Wealsodemonstratedthat thereismorethananorderofmagnitudevariationinmeancohort“tness,providing tremendousvariationforDLHEstoactupon.Wedocumentedanevolutionarytrade-off betweenmassatemergenceanddateofemergence,whichmayplayaroleinmaintainingthe variationinmass(“tness)atemergence.Aliteraturereviewrevealedthatsuchhighlevelsof intercohortvariationoccurinmanyotherpond-breedingamphibians,andthatappropriately documentingthemagnitudeofintercohortvariationrequireslong-termstudies(roughlytwo populationturnovers).GiventheprofoundeffectthatDLHEscanhaveonpopulation dynamics,quantifyingintercohortvariationinmean“tnessandthelevel(s)atwhichselection actswillbeveryimportantfordevelopingaccuratemodelsofpopulationdynamics.Ingeneral, whendevelopingmodelsofpopulationdynamics,moreattentionshouldbepaidtovariation inmean“tnessandnotjustvariationintotalnumbers.Keywords: Ambystomacaliforniense; Californiatigersalamander;contextualanalysis;dateof emergence;intercohortvariation;JepsonPrairiePreserve,California,USA;massatemergence;pondbreedingamphibian;populationturnover;selectiongradient .INTRODUCTIONManytaxa,butmostnotablyholometabalousinsects andmanyamphibians,havecomplexlifecycles.Given thenumericaldominanceofsuchspeciesonearth,itis importanttounderstandhowtheirpopulationsare regulated.Inparticular,wewouldliketoknowwhether densitydependenceoccursinoneormanystagesand how“tnessinonestageaffects“tnessinsubsequent ones.Thislatterproblemconstitutesthegeneralpurview ofdelayedlifehistoryeffects(DLHEs;Beckermanetal. 2002),whichareknowntocreatelimitcyclesin populationsize(Leslie1959,ProutandMcChesney 1985).Such”uctuationshavebeenobservedinmany amphibianpopulations,buttheyareusuallyattributed toclimaticeffects(Pechmannetal.1991).Infact, DLHEsandclimaticvariationcanactsynergistically,as thepopulationoscillationscreatedbyDLHEscanbe entrainedbystochasticfactors,suchasclimatic”uctuation(Leslie1959).Moregenerally,thelessonsthatcan belearnedfromthestudyofDLHEsarenotapplicable solelytospecieswithcomplexlifecycles,astheyhave alsobeenfoundinhumans(LummaaandCluttonBrock2002)andothermammals(Albonetal.1987, Roseetal.1998). Althoughagreatdealoftheworkthatisrelevantto DLHEshasinvolvedamphibians,ithasnotbeen referredtoassuchintheliterature.Manyspeciesof amphibianshaveanaquaticlarvalstagefollowedbya terrestrialadultstage,andtheecologicalconsequences ofselectioninthelarvalstageonlaterterrestriallife provideanobviousquestionforlifehistorytheory.The mainstumblingblockinlearningaboutDLHEsin amphibianshasbeenlackofknowledgeaboutthe terrestrialadultstage,aconsequenceofthesecretive, oftenfossorial,natureofpost-metamorphicterrestrial amphibians.Breedingsitesaregenerallymuchsmaller thantheirassociatedterrestrialhabitat(Semlitschand Bodie2003,RittenhouseandSemlitsch2007),andare theoneconcentratedareathatmanymembersoftheManuscriptreceived20January2013;revised12June2013; accepted19June2013.CorrespondingEditor:W.D.Koenig.5E-mail:casearcy@ucdavis.edu 68

PAGE 2

populationcanbecountedontovisit,oftenata predictabletimeofyear.Thus,mostofwhatisknown aboutDLHEsinamphibianshasfocusedonthe relationshipbetweenthe“tnessofdispersingmetamorphsleavingthebreedingsiteandthe“tnessofthose sameindividualswhentheyreturninsubsequentyears assexuallymatureadults(Berven1990,Scott1994, Berven2009).Littleisknownaboutthe“tness relationshipsbetweenmetamorphsandjuveniles,betweenthemultipleyearsasterrestrialjuvenilesthat manyamphibiansrequiretomature,andbetween juvenilesandadults,andyet,suchstudiesarecriticalif wearetopiecetogetheracompleteanalysisofDLHEs innature. Here,weuseadisperseddriftfencearraytostudya populationoftheendangeredCaliforniatigersalamander( Ambystomacaliforniense ;CTS;seePlate1)during theentiresurface-activeseason.Thisallowsusto examine“tnessrelationshipsnotonlybetweenmetamorphsandadults,butalsoacrossthepoorlystudied interveningage-classtransitions.TheseDLHEsmaybe particularlyimportantinlightofthehugevariationthat weobservein“tness,notonlybetweenindividuals,but alsobetweenwholeannualcohorts.Studiesofother pond-breedingamphibianshavefoundvariationin meancohort“tness(Semlitschetal.1988,Scottetal. 2007,Berven2009),buttherehasbeenlittlediscussion oftheimpactofthatvariationonpopulationdynamics. Wealsousecontextualanalysis(HeislerandDamuth 1987)todeterminewhethervariationin“tnessisbetter modeledattheindividualorcohortlevel,becausethis willinformhowsubsequentmodelsofpopulation dynamicscanbestbedeveloped(TaylorandScott 1997). METHODSDatacollection OurstudywasconductedattheJepsonPrairie Preserve,SolanoCounty,California,USA,oneofthe bestremainingexamplesofnativeCaliforniaprairie (Ku ¨ chler1977).Themostprominentfeatureofthe preserveisOlcottLake,a33-haplayapoolthatservesas abreedingsiteforCTS.OlcottLakeisephemeraland “llstoamaximumdepthof ; 0.6m.Ourdriftfence arrayislocatedalongthenortheastshoreofOlcott Lake.Thefencesofthearrayareevenlydistributed acrosstheuplandlandscape,10…1000mfromthe shoreline.Theshorelinefenceiscontinuous,covering ; 17 % ofthepondshoreline(390m),withpitfalltraps every10m.Theremainingfencesare10mlongand separatedfromeachotherby90-mgaps.Pitfalltrapsare relativelylarge,3.78-Lbucketsthatcanaccommodate 20individuals,evenofthelargestsizeclass.Fora moredetaileddescriptionofthearrayseeSearcyand Shaffer(2011).Duringthefallof2010,asecond continuous840mlongdriftfencewasinstalled275m fromtheshoreline. ThepitfalltraparraywasoperatedMay2005…July 2011.Trapswereopeneveryrainynightbetween OctoberandMarchandforeverynightduringthe May…Julyemergenceperiod,andweremonitoreddaily (SearcyandShaffer2011).Werecordedthebodymass, traplocation,ageclass,andadigitalimageforeach salamandercaptured.Onaverage,CTSreachsexual maturityatfouryearsold;however,givenafastenough growthrate,somefemaleshavebeenknowntoreach sexualmaturityinthreeyearsandsomemalesintwo years(Trenhametal.2000).Weobservedafew individuals(bothmaleandfemale)thatweremature atoneyearold.Thelocationofeachdriftfencewas determinedusingaTripod200CRanger(TripodData Systems,Corvallis,Oregon,USA)withanaccuracyof 6 2.1m. Adultandjuvenilerecapturesweredeterminedusinga patternrecognitionprogramcustomdesignedforCTS (SearcyandShaffer2011).Metamorphslackdiscrete spots(spotstakeseveralmonthstofullycoalescepostmetamorphosis),sorecapturesweredeterminedusing eitherVIAT(visualimplantalphanumerictags)orVIE (visualimplantelastomer;NorthwestMarineTechnology,ShawIsland,Washington,USA)(Jerryetal.2001). In2005and2006,2335metamorphswereimplanted withVIATandin2010another1289metamorphswere implantedwithVIE(AppendixA). Analysis Throughoutthisstudy,weconducttwodifferenttypes ofanalyseswhenmakingcomparisonsamongthe“ve possibleage-classtransitions(metamorph…juvenile, metamorph…adult,juvenile …juvenile,juv enile…adult, adult…adult).Within-yearcomparisonscompareanimals toothermembersoftheirowncohort,askingwhether thereisasurvivaladvantagetobeinglargerthan animalscapturedinthesameyearfromthesameage class.Across-yearcomparisonspoolacrosscohorts, askingwhetherthereisanadvantagetobeinglarger thananimalsfromthesameageclass,regardlessofthe yearinwhichtheywerecaptured.Within-yearcomparisonsstatisticallyremovetheeffectsofgoodandbad years(currentpopulationdensity,rainfall,andother climacticvariablesbeingtheprimarydrivers),whereas across-yearcomparisonsincludethoseeffects. Inmanyofthefollowinganalyses,weuserecaptureas aproxyforsurvivaland,thus,“tness.Presumably,some (unknown)numberofanimalsthatwerenotrecaptured didsurvive,andaslongastherecapturedanimalswerea randomsubsetofthesurvivors,itshouldnotaffectour interpretations(AppendixB).Thealternativeistouse survivalestimatesfromMARK(WhiteandBurnham 1999).WeranMARKforourdata,andfoundthatthe survivalestimatesobtainedwereunrealisticallylow comparedtothosefromanotherCTSpopulationin MontereyCounty(P.C.Trenham, unpublisheddata ). Forexample,MARKprovidedadultsurvivalestimatesof0.24/yr(in2007)and0.28/yr(in2008),whichJanuary2014 69 AMPHIBIANDELAYEDLIFEHISTORYEFFECT

PAGE 3

areroughlyone-thirdofthosefromMontereyCounty. WebelievethatthisisbecauseourJepsondriftfences arealargelyopensystem,surroundingone-sixthofthe breedingpond,andthusmanyanimalsmigrateintoand outofourstudyarea.IntheCormack-Jolly-Seber model,onwhichourMARKanalysiswasbased, emigrantsarelumpedwithmortalities,thusdecreasing theestimatedsurvivalrate.Giventhis,wefeelthatinthe absenceofemigrationrates,estimatedmortalityschedulesfromMARKarebothbiasedandunrealistically lowforthissystem.Amoredetailedanalysisofthisissue willbepresentedelsewhere. WeusedstandardANOVAandregressiontotestfor selectiononmass.Thisstartedwith10ANOVAs lookingatindividual-levelselection,halfwithablocking termforyeartomeasurewithin-yearselectionandthe otherhalfwithoutablockingtermtomeasureacrossyearselection.Oneofeachtypewasusedforeachofthe “vebetween-andwithin-age-classtransitions.Wetested forcohort-levelselectiononmassbetweenthemetamorphandjuvenilestages,usingalinearregression betweenmeancohortmassatemergence(ME)and percentagerecapturedas“rst-yearjuveniles.Acontextualanalysis(HeislerandDamuth1987)wasusedtotest whetherselectiononmassbetweenthemetamorphand juvenilestageswasbettermodeledattheindividualor cohortlevel(i.e.,asafunctionofindividualmassorasa functionofmeancohortME). WenextexaminedtherelationshipbetweenMEand dateofemergence(DE).BecauseDEandMEare correlatedinCTS(Trenhametal.2000),weuseda correlatedselectionapproach(LandeandArnold1983) toinvestigateindividual-levelselectiononbothtraits. Thistestwasonlyrunforthetransitionbetween metamorphandjuvenilestages,asthatwasthe transitionmostlikelytobeaffectedbyDE.Wethen examinedhowthesetraitsaffectdistancetraveledasa metamorph.ThisanalysisutilizedanANCOVAwith DEandMEascovariatesandyearasaclassvariable. WeusedaseparateANCOVAtolookdirectlyatthe relationshipbetweenMEandDE,withMEasthe responsevariable,DEasthecovariate,andyearasthe classvariable.Toplacetheseselectionanalysesinthe contextofCTSlifehistory,weusedanANCOVAtotest therelationshipbetweenMEandmassasa“rst-year juvenile,withjuvenilemassastheresponsevariable,ME asthecovariate,andyearastheclassvariable.Wealso usedageneraladditivemodeltodescribetheaverage masstrajectoryforaCTS.Here,thenumberofdays sinceemergencewasusedtopredictfactorialchangein masssinceemergence. Becausetheimportanceofthesemass-dependent effectsdependscriticallyontheamountofstanding variationinME,wealsocollecteddataonthisvariation fromotherstudiesofpond-breedingamphibians.We identi“ed,tothebestofourknowledge,allstudiesthat includeddataonmeancohortmassacrossmultiple yearsatthesamebreedingsite.Wesummarizedthis variationastheratiobetweenthemeanmassofthe largestcohortandthemeanmassofthesmallestcohort, andplottedthisratioasafunctionofthenumberof annualcohortsavailableforthatsite.Wethen comparedthe“tofalinearvs.aquadraticregression toaskhowmanyyearsofdata,onaverage,areneeded toestimatethetotalvariationinMEforapopulation. Foramoredetaileddescriptionofthemethods,see AppendixB. RESULTSAnalysesacrossthreeyearsatOlcottLake Intotal,25344salamanderswerecaptured:4247 (spring2005),5582(2005…2006),1509(2006…2007), 1078(2007…2008),314(2008…2009),2448(2009…2010), and10166(2010…2011).WedetectedDLHEsacrossall lifestagetransitionsinCTSintheformofselectionfor increasedmass(Table1),althoughinsomecasesthis wasonlytrueoftheacross-yearandnotthewithin-year selection.Thisdifferencebetweenacross-andwithinyearselectionwasgreatestforthemetamorph-tojuveniletransition,wherewithin-yearselectioniseither neutralorinthedirectionofsmallermass,butacrossyearstronglyfavorslargermass.Thisdifferenceis drivenbystrongcohort-levelselectionforlargermass( P 0.03)thatoverwhelmsweakerindividual-levelselectionforsmallermass( P 0.17).Signi“cantcohort-level selectionforlargermasswasdetectablewithonlythree markedmetamorphcohortsduetotheexceptionally strongrelationshipbetweenrecapturerateandmean cohortME( R2 0.998;AppendixB:Fig.B.1).A contextualanalysis(HeislerandDamuth1987)revealed thatselectiononmassatemergenceacrossthe metamorph-to-juveniletransitionwasmodeledsigni“cantlybetteratthecohortlevelthanattheindividual level(multiplelinearregression; bI[coef“cientofpartial regressionontheindividualcharacter] 0.02, P 0.17; bC[coef“cientofpartialregressiononthecohortmean] 0.24, P 0.001). Althoughtherewasnoindividual-levelselectionfor largerbodysizeacrossthemetamorph-to-juvenile transition,therewassigni“cantindividual-levelselection forearlieremergence.Aselectionanalysisoncorrelated characters(LandeandArnold1983)revealedthat, withinyears,therewassigni“cantselectionforearlier emergenceandmarginallysigni“cantselectionfor smallermass(multiplelogisticregression;forDE, P 0.01;forME, P 0.07;foryear, P 0.001).Inaddition, metamorphsthatemergedearliertraveledfartherinthe metamorphstageandthushadtheopportunityto samplealargerareabeforeselectinganover-summering refugesite.AnANCOVArevealedthatmetamorphs thatemergedearliertraveledsigni“cantlyfarther, althoughtherewasalsosigni“cantinterannualvariation inthiseffect(ANCOVA;forDE, P 0.001;forME, P 0.97;foryear, P 0.001;forDE 3 year, P 0.001; forME 3 year, P 0.01).Whenanalyzedwithineach cohort,earlieremergencewasconsistentlyassociatedCHRISTOPHERA.SEARCYETAL. 70 Ecology,Vol.95,No.1

PAGE 4

withtravelingfarther,anditwasonlytheextentofthis effectthatvariedbyyear(AppendixB:TableB.1). Becausenoneofthesimpleeffectscontradictedthemain effect,weconcludethat,ingeneral,metamorphsthat emergedearliertraveledfarther. Althoughtherewasnowithin-yearselectionforlarger sizeasametamorph,therewasstrongwithin-year selectionforlargersizeasajuvenile(Table1).MEwasa strongpredictorofmasswhen“rstcapturedasa juvenile(Fig.1),implyingthatalthoughtherewasnot animmediateadvantagetolargesizeatemergence,there wasanadvantagetothislargersizelaterinlife.This madeMEaclassicexampleofaDLHE;therewasno immediateadvantagetoalargeME,butlargerMEwas selectedforthroughitspositivecorrelationwithmassas a“rst-yearjuvenile.Therewasalsostrongwithin-year selectionformetamorphstoemergebothearlyandata largesize,creatingapotentialevolutionarytrade-off, becauseearlyemergencenecessarilymeanstruncating theperiodofrapidlarvalgrowth.Thiswithin-year selectionresultedinanegativecorrelationbetweenME andDE(Fig.2),becausethemost“tlarvaeemerged bothearlyandlargeandtheleast“tlarvaeemerged bothlateandsmall. Fig.3showsthemasstrajectoryoftheaverageCTS, providingacontextforthesesometimescon”icting “tnesscomponents.MeanDEatJepsonPrairiewas10 June.Overthecourseofthatsummer,whentherewas selectionforearlyemergencebutnoneonME,the averagesalamanderlost36 % ofitsbodymass. Maximumlossregisteredon2October,justbeforethe wetseasonof“ciallybeginson26October(Ko ¨ ppen 1936).Assoonasthewetseasonbegan,salamanders wereconsideredone-year-oldjuvenilesandselection switchedtofavoringgreaterME.Thiscorrespondedtoa periodofrapidgrowthduringwhichtheaverage salamanderreturnedtoitsMEby5March.Growth stalledagain(2May…26September),againmatching closelythedryseason(17April…26October).Rapid growthcontinuedduringthesecondyearasajuvenile, suchthatby5Octoberofthethirdyear,theaverage salamanderhadreachedthemeanadultsize(23.43 6 0.17g;alldatareportedasmean 6 SE)andwas presumablyreadytobreed.Pastthispointinthethird year,wehaveverylittleinformation,andcanonlynote thatsalamandersappeartocontinuegrowingintotheir fourthyearandpossiblybeyond,probablyata deceleratingrate. Fromtheperspectiveofmodelingpopulationdynamics,theimportanceoftheseDLHEswilldependupon theamountofvariationinMEthatselectionhastoact upon.Overourstudy,therewassubstantialvariation bothinindividualME(range3.7…21.0g)andinmean cohortME(range7.1 6 0.1gto13.7 6 0.4g).Thus, regardlessoftheirsizeintermsofnumberofindividuals, somecohortswillcontributealmostnobreedingadults tothepopulation,whereasothercohortswillhavevery highsurvivorshiptomaturity. PatternsacrossCaliforniatigersalamandersites OlcottLakeisanenormousandsomewhatunusual breedingsiteintermsofitssize.Toexaminethe generalityofthevariationinmeancohortMEthatwe foundatOlcott,wesummarizevariationinMEfromtwo otherbreedingsitesoverarangeofyearsandhydroperiodsinFig.4.RoundPondisasmaller,butstillquite large,3-habreedingsite0.9kmfromOlcott;Blomquist Pondisamoretypical0.07-habreedingsiteinMonterey CountyadjacenttotheHastingsNaturalHistory Reservationthatwasstudiedwithasingledriftfence TABLE1.Effectofmass(mean 6 SE)onsurvivaloftheCaliforniatigersalamander( Ambystomacaliforniense )fromwithin-and across-yearanalysesof“veage-classtransitions. Age-class transition Massbeforeselection(g)Massafterselection(g)Masschange( % )WithinyearAcrossyear WithinyearAcrossyearWithinyearAcrossyearWithinyearAcrossyeardf P df P Metamorph tojuvenile 9.92 6 0.059.48 6 0.069.62 6 0.2110.89 6 0.29 3.014.93,24520.171,2454 0.001 Metamorph toadult 11.34 6 0.089.48 6 0.0613.08 6 1.0012.51 6 1.2215.432.13,23240.071,23260.01 Juvenileto juvenile 8.97 6 0.117.51 6 0.0811.56 6 0.7810.61 6 0.8528.841.23,1449 0.0011,1451 0.001 Juvenileto adult 8.98 6 0.117.51 6 0.0811.47 6 0.9711.25 6 1.1427.849.73,14390.011,1441 0.001 Adultto adult 22.86 6 0.2123.43 6 0.1724.43 6 0.8325.33 6 0.876.98.14,11130.061,11160.03 Notes: Thewithin-yearanalysisincludesablockingtermforyearsothatanimalsareonlycomparedtoindividualsinthesame ageclassthatwerecapturedinthesameyear.Thiscontrolsforthedifferentclimaticconditionsthatanimalsexperiencedin differentyears.Theacross-yearanalysiscomparesanimalstoallindividualsinthesameageclass,regardlessoftheyearinwhich theywerecaptured.Meanmassbeforeselectionisthemeanmassofallanimalsintheageclassbeforethetransition;meanmass afterselectionismeanmassintheageclassbeforethetransition,calculatedforthoseanimalsthatwererecapturedafterthe transition.Allestimatesofmeanmassareleast-squaremeanstakenfromtheANOVAmodels,whichexplainswhyvaluesare slightlydifferentdependingonthespeci“cmodel.Thelargerdiscrepancyinmeanmetamorphmassbeforeselectionforthewithinyearanalysisisduetothefactthat,forthemetamorph…adulttransition,alldatafromthe2006cohorthadtobedroppedbecause nomembersofthatcohorthadbeenrecapturedasadults.Masschangeexpressesthedifferencebetweenmeanmassbeforeand afterselectionasapercentage,andsigni“cancegivesthe P valuefortheANOVAassociatedwiththatmasschange. January2014 71 AMPHIBIANDELAYEDLIFEHISTORYEFFECT

PAGE 5

FIG.1.MassatemergencestronglypredictsjuvenilemassintheCaliforniatigersalamander( Ambystomacaliforniense ).Data fromallthreeyearswithmetamorphsmarked(2005,2006,and2010)areshown;factorsaremeancohortmassatemergence(ME), year,andtheirinteraction.Log-transformedmasswasoriginallymeasuredingrams. FIG.2.Duetocon”ictingselectionpressures,massatemergencedecreaseswithemergencedate.Dateisrepresentedasthe numberofdaysafter30April.Datafromallthreeyearswithmetamorphsmarked(2005,2006,and2010)areshown;factorsare dateofemergence(DE),year,andtheirinteraction.Log-transformedmasswasoriginallymeasuredingrams. CHRISTOPHERA.SEARCYETAL. 72 Ecology,Vol.95,No.1

PAGE 6

fornineyearsduringthe1990s(Trenhametal.2000). VarianceamongmeancohortMEisfairlysimilar betweenthethreesites:forBlomquist,SD 3.05g;for Olcott,SD 2.62g;forRound,SD 3.56g.This suggeststhatsubstantial“tnessvariationforDLHEsto actuponwillbecommonacrossCTSsites,regardlessof theirsizeandlocation.WealsonotethatOlcottLakeand RoundPondareincloseproximitytoeachotherand experienceidenticalclimaticconditions,yetthecorrelationinmeancohortMEbetweenthetwositesislow( r 0.27),indicatingthatthefactorscontrollingmeancohort MEarenotpurelyafunctionoflocalweatherconditions. FIG.3.MasstrajectoryforCaliforniatigersalamanders(solidline,mean;dashedlines,SE).Eachcirclerepresentsarecapture event.Thelinewas“tusingageneraladditivemodel.Numbersinparenthesesarecoordinatesforeachlocalminimumand maximumofthemodel.Valuesonthe x -axisaredaysafterdateofemergence(meandateofemergenceis10June); y valuesarethe percentageincreaseordecreaseinmassrelativetomassatemergence(meanmassatemergenceis10.77g).Thus,100onthe y axisrepresentsa100 % increaseinmasssinceemergence(i.e.,21.54gforasalamanderwiththemeanmassatemergence). FIG.4.MeancohortmassatemergenceforthreedifferentCaliforniatigersalamanderbreedingpondsacross7…9yearsof studyinCalifornia.Eachbreedingpondisrepresentedbyadifferentpattern:lightgray,BlomquistPond,HastingsNaturalHistory Reservation,MontereyCounty;mediumgray,OlcottLake,JepsonPrairie,SolanoCounty;black,RoundPond,JepsonPrairie, SolanoCounty. January2014 73 AMPHIBIANDELAYEDLIFEHISTORYEFFECT

PAGE 7

Patternsacrosspond-breedingamphibians Finally,toplacethevariationthatweobservedinME intoabroadercontext,wecompiledaliteraturereview ofstudiesofpond-breedingamphibiansthathave documentedmeancohortMEovermultipleyearsfrom thesamebreedingsite(AppendixB:TableB.2).We calculatedamong-cohortvariationastheratioofthe largesttothesmallestmeancohortME.Thevariation rangesfrom1.0to 5.0,andthevariationinmean cohortMEinCTS(2.8)fallsinthemiddleofthe variationobservedinotherspecies.Asexpected,the largest/smallestMEratioincreaseswiththenumberof cohortsthatareinvestigated( P 0.001;AppendixB: Fig.B.2).Basedontheavailabledata,thereisno indicationthattherangeofvariationissaturatingwith thenumberofmeasuredcohorts(thequadratictermis notsigni“cant, P 0.19),evenwiththelongestexisting studies(22and23annualcohorts). DISCUSSIONDelayedlifehistoryeffectsoccurwhenlifehistory traits(e.g.,survival)dependonbothcurrentand previousenvironments(Beckermanetal.2002).We havefoundpervasiveevidencefordelayedlifehistory effectsacrossthelifehistoryofCTS.Ateachstagein CTSslifecycle,greatermass,whichmustbegoverned inpartbycurrentandpreviousenvironments,leadsto highersurvivaltothenextstage(Table1).Previous studiesofpond-breedingamphibianshaveshownthat variationinMEaffects“tnessofbreedingadults (Berven1990,Scott1994,Berven2009;butseeBeck andCongdon1999,Boone2005,Gramapurohit2009 forcounterexamples).However,noneofthesestudies hasbeenabletopartitionthislarge-sizeadvantageat metamorphosisamongtheinterveninglifestagetransitions.Wefoundthatlargersizeisadvantageousatall lifehistorystagetransitions,althoughatthemetamorph-to-juveniletransition,thisselectionadvantageis bettermodeledatthecohortlevelthanattheindividual level.Themetamorph-to-juveniletransitionisparticularlynuanced,andinvolvesapotentialtrade-off betweenMEandDE,withearlierDEprovidingan immediate“tnessbene“ttometamorphs,whereas greaterMEonlyprovidesa“tnessadvantagelaterin life. Delayedlifehistoryeffects ModelsthatincorporateDLHEshaveproventomore accuratelyresembletruepopulationdynamicsinboth plantandinsectsystems(GinzburgandTaneyhill1994, Crone1997).Itisthusessentialthatweunderstandhow DLHEsoperateifwewanttobuildaccuratepopulation modelsforamphibiansandothertaxa.Thepervasive DLHEsthatwefoundinCTS,coupledwiththe substantialvariationinmeancohortME,clearlycould havealargeeffectonpopulationdynamics.In particular,themodelforacross-yearselectiononmass betweenthemetamorphandadultstagessuggeststhat withinasinglebreedingpond,terrestrialsurvivalofthe averagemetamorphintheheaviestcohort(14.9 6 0.3g) isupto18.8timesthatoftheaveragemetamorphinthe lightestcohort(5.3 6 0.1g).Althoughthisvariationin averagemetamorphqualityisnotasgreatasthe variationthathasbeendetectedinmetamorphquantity (whichrangesoverfourordersofmagnitude;seeSearcy etal.2013),itwillhavealargeeffectonpopulation dynamics. Unfortunately,westilldonotfullyunderstandwhich environmentalfactorsaregoverning”uctuationsin meanmetamorphmass.Noneoftheenvironmental factorsthatwetested(e.g.,numberofmetamorphs, numberofbreedingfemales,meantemperature,annual precipitation)wassigni“cantlycorrelatedwithmean cohortME.Whatwedounderstandisthetremendous levelofvariationinmeancohortMEinCTS populations,andthatsimilarlevelsofvariationare foundinvirtuallyallpond-breedingamphibiansthat havebeenstudiedtodate.Fig.B.2(AppendixB) suggestsboththattherangeofvariationinmeancohort MEisverylarge(5.42timesin Ambystomatalpoideum ; D.E.Scott, unpublisheddata )andthataccurately estimatingtherangeofvariationwillrequiremonitoring apopulationforaverylongtime(over20years). Averageturnoverperiodforthesespeciesis ; 10years (e.g.,GibbonsandSemlitsch1991,TaylorandScott 1997,Trenhametal.2000),andourliteraturereviewis consistentwithConnellandSousa(1983),whoconcludedthat,asagoodruleofthumb,populationsshouldbe followedfortwofullturnoverperiods(seealsoBlaustein etal.1994,PechmannandWilbur1994). Multilevelselection Ouranalysispointstoselectionactingatboththe individualandcohortlevel,andeachmaybeimportant contributorstothetotalselectiononapopulationover time.Otherstudieshaveshownthatthereisasigni“cant correlationbetweenmeancohortMEandsurvival (Berven1990,Scott1994),butdidnotutilizea contextualanalysis(HeislerandDamuth1987)to partitionthecohort-levelrelationshipintoindividuallevel(withineachcohort,ahigherpercentageoflarge individualssurvive)orcohort-level(arandomsampling fromwithineachcohortsurvive,butahigherpercentage arefromcohortswithalargermeansize)selection.Our contextualanalysisrevealsthatacrossthecritical metamorph-to-juveniletransition,selectionforlarger massisbetterrepresentedascohort-levelselection.This willplayanimportantroleinhowselectiononmassis modeledforapopulation,becausepredictionsmadeby modelsbasedoncohortvs.individuallevelselection divergeastheintrinsicrateofpopulationgrowth increases(TaylorandScott1997). Itisalsoimportanttokeepinmindthatthisisa purelycorrelationalanalysis.Itishardtoconceiveofa mechanismbywhichafocalmetamorphssurvival probabilitywouldbeincreasedbybeingsurroundedbyCHRISTOPHERA.SEARCYETAL. 74 Ecology,Vol.95,No.1

PAGE 8

otherlargemetamorphs.Itseemsmorelikelythatthere issomeenvironmentalfactor,orsuiteofcorrelated environmentalfactors,thatbothimprovesthequalityof theaquaticenvironment,therebyincreasingmean cohortME,andsimultaneouslyamelioratesmortality pressuresintheterrestrialenvironment,therebyincreasingsubsequentover-summersurvivorship.Onefactor thatcouldpotentially“llthisroleisthenumberof cohortmates,becausecompetitionwithotherindividualsinthesamecohortforbothaquaticpreyand terrestrialburrowswoulddecreasebothmeanMEand terrestrialsurvival.Althoughwedidnotdetecta signi“cantrelationshipbetweennumberofmetamorphs andmeancohortME(seeprecedingsection),and survivaloverthe“rstsummerwasbetterpredictedby meancohortME( R2 0.998)thanbynumberof metamorphs( R2 0.88),thisdoesnotprecludethe possibilitythatadatase twithmoreacross-year replicationwoulddetectasigni“cantcontributionof numberofmetamorphstobothoftheothermetrics. Whenconsideringthemeanmasstrajectorydepicted inFig.3,itisdif“culttobelievethatindividualselection onmassisnotoperatingacrossthemetamorph-tojuveniletransition.Duringthistransition,theaverage salamanderloses36 % ofitsmass.Largersalamanders presumablyhavelargerfatandwaterstores,andthus shouldbebetterabletotoleratethismassloss.For example,itishasbeenshownthatlargermassincreases timetodehydrationinadesert-adaptedanuran(NewmanandDunham1994),andshouldthereforebean importantcomponentofsurvivalovercentralCaliforniashot,rainlesssummer.However,itmaybethecase thatthethreatofdehydrationissoseverethatsuchan effectisoutweighedbytheover-summeringsitethata salamanderselects,becauseevenahigh-qualitysalamanderwilldieinalow-qualityretreat.Thismay explainwhyearlyemergence,whichiscorrelatedwith samplingalargerareaasametamorphandthushaving alargerrangeofover-summeringsitestochoosefrom,is suchanimportantcomponentof“tness. Optimalsizeatmetamorphosis Fig.2revealsanegativecorrelationbetweenMEand DE.IfthisrelationshipisdrivenbyselectiononME alone,aswasassumedinclassicmodelsofamphibian metamorphosis(Wilbur andCollins1973,Werner 1986),itrequiresthattheoptimalvalueofMEdecreases overtime.UndertheWilburandCollins(1973)model, thiswouldrequirethebodysizeatwhichlarvalgrowth rateslowstodecreaseovertime,asthisdecreasinglarval growthrateisthepostulatedtriggerinitiatingmetamorphosis.Thiscouldbethecaseiflargerlarvaerequirea higherrateofpreycapturethansmalleronesto maintainthesamesize-speci“cgrowthrate.Giventhat CTSaresit-and-waitpredatorsthatswallowtheirprey whole,thepro“tabilityofpreymaybegovernedsolely bysearchtime,whichislinearlycorrelatedwithprey density.Preydensitymaywelldecreaselinearlyoverthe courseoftheemergenceperiod,consistentwithalinear decreaseinoptimalME.Fromtheperspectiveofthe PLATE1.AdultCaliforniatigersalamander( Ambystomacaliforniense )attheJepsonPrairiePreserve,SolanoCounty, California,USA.AdultandjuvenileCaliforniatigersalamandersareonlysurface-activeintheterrestrialhabitatonrainynights duringthefallandwinter.Individualstravelhundredsofmetersbetweentheirbreedingpondandrefugesites(rodentburrows), wheretheyareprotectedfromCaliforniashot,drysummers.Photocredit:VideOhlin. January2014 75 AMPHIBIANDELAYEDLIFEHISTORYEFFECT

PAGE 9

Werner(1986)model,earlymetamorphosisatalarger sizecanbeexplainediftheaquatic l / g (mortality/ growth)curvemovesupwardoverthecourseofthe emergenceperiod,shiftingitsintersectionwiththe terrestrial l / g curvetoalowermass.Thisupward movementoftheaquatic l / g curvecouldoccurifeither l increasesor g decreases.Bothoftheseseemplausible, since l presumablyincreasesdramaticallyasponddryingreducesthevolumeofwater,and g maydecrease asthedensityofaquaticpreydecreases.However,a simplerexplanationforthenegativecorrelationbetween MEandDEresultsifweconsiderthatbothtraitsare underselectionsimultaneously,andthatthereisatradeoffbetweenthem. Evolutionarytrade-offs Agrawaletal.(2010:245)de“neamulti-traittrade-off asoccurringwhentwoormoretraits,whichareboth underdirectionalselectiontoincrease,sharealimiting resource.Ouranalysesdemonstratethat,inCTS,both MEandDEareunderdirectionalselection(MEfor largersizeandDEforearliertime).Wepositthatthe sharedlimitingresourceistime.Clearly,individualsthat emergefromthepondearlieraresacri“cingtimeinthe aquaticenvironment.Aslongasextratimeinthe aquaticenvironmentwouldalsoleadtogreaterME, thenthelimitingresourceisshared.Althoughthe“rst metamorphsemergeinlateMay,otherlarvaeremainin theaquaticenvironmentforanothermonth.Ifthereare enoughresourcestosustainthegrowthoftheselateremergingmetamorphs,thenthereiseveryexpectation thattheearly-emergingmetamorphs,whichmusthave hadahigherrateofresourceacquisitiontoobtaintheir earlysizeadvantage,wouldcontinuetogrowifthey remainedintheaquaticenvironment. Wepostulatethatanevolutionarytrade-offbetween MEandDEcaneasilyexplainthenegativecorrelation betweenthesetwocriticallifehistoryfeatures.Clearly, theoptimalstrategyinthissituationistoemergeboth earlyandlarge.Larvaewithahighrateofresource acquisitionmaybeclosetothisoptimumandhavethe luxuryofmakingsmallsacri“cesinbothMEandDEas partoftheirtrade-off.Larvaewithlowerratesof resourceacquisitionwillalsobeforcedintoatrade-off betweenDEandME,andduetotheirslowgrowthrate, willemergewithanevenlowerME,despitemakinga largersacri“ceinDE.Anadvantagetoearlyemergence seemstobeacommonthemeacrossamphibianspecies (Smith1987,Semlitschetal.1988,AltweggandReyer 2003),suggestingthatthistrade-offmaybecommon, andmaymaintainvariationinbothmetrics(Bervenand Gill1983)withinandacrosscohorts. Conclusions Previousworkhasshownthat:(1)iftheyexist, DLHEswillplayanimportantroleinpopulation dynamics(Leslie1959);(2)thatDLHEsarecommon inpond-breedingamphibians(Berven1990,Scott1994, Berven2009);and(3)thatthereissubstantialintercohortvariationinmetamorphqualityforDLHEsto actupon(Semlitschetal.1988,Scottetal.2007,Berven 2009).Althoughthisintercohortvariationinmetamorphqualityhasbeendocumented,itspotential impactonpopulationdynamicsanditspartitioning amongterrestriallifehistoryphaseshavereceivedvery littleattention.WedemonstratethatinCTS,intercohort variationinaveragemetamorphqualitysubstantially impactspopulationdynamics.Wehavealsoshownthat accuratelydocumentingintercohortvariationinmetamorphqualityrequiresdataforatleasttwocomplete populationturnovers,thatDLHEsareprominentacross alllifestagetransitions,andthatthereisvariationin whetherselectionisprimarilyattheindividualorcohort level.Alloftheseempiricalobservationsaffecthow populationmodelsshouldbedeveloped(Taylorand Scott1997).GiventhatDLHEsarecommonacross diversetaxa,notjustpond-breedingamphibians(Rose etal.1998,Beckermanetal.2002,Lummaaand Clutton-Brock2002),ourobservationsontheimportanceofvariationinqualityaswellasquantityandhow itinteractswithDLHEsarewidelyapplicabletothe understandingofpopulationdynamics.ACKNOWLEDGMENTSWethankK.PopeandD.Scottforcontributingdataforthe meta-analysis,A.Clause,J.Ersan,E.Gabbai-Saldate,B. Johnson,andM.Starkeyfortremendoushelpatthedrift fences,andE.Cole,R.Hartman,S.Lawler,J.Rose,andtwo anonymousreviewersforgreatlyimprovingthemanuscript. ThisworkwasconductedunderFederalFishandWildlife permitTE094642-0andwasfundedbygrantsfromtheBureau ofReclamation,theCaliforniaDepartmentofTransportation, theNationalScienceFoundation,theSolanoCountyWater Agency,UC…Davis,andtheUCNaturalReserveSystem. LITERATURECITEDAgrawal,A.,J.K.Conner,andS.Rasmann.2010.Tradeoffs andnegativecorrelationsinevolutionaryecology.Pages243… 268 in M.A.Bell,D.J.Futuyma,W.F.Eanes,andJ.S. Levinton,editors.EvolutionsinceDarwin,the“rst150years. SinauerAssociates,Sunderland,Massachusetts,USA. Albon,S.D.,T.H.Clutton-Brock,andF.E.Guinness.1987. Earlydevelopmentandpopulationdynamicsinreddeer.2. Density-independenteffectsandcohortvariation.Journalof AnimalEcology56:69…81. Altwegg,R.,andH.Reyer.2003.Patternsofnaturalselection onsizeatmetamorphosisinwaterfrogs.Evolution57:872… 882. Beck,C.W.,andJ.D.Congdon.1999.Effectsofindividual variationinageandsizeatmetamorphosisongrowthand survivorshipofsoutherntoad( Bufoterrestris )metamorphs. CanadianJournalofZoology77:944…951. Beckerman,A.,T.G.Benton,E.Ranta,V.Kaitala,andP. Lundberg.2002.Populationdynamicconsequencesof delayedlife-historyeffects.TrendsinEcologyandEvolution 17:263…269. Berven,K.A.1990.Factorsaffectingpopulation”uctuationsin larvalandadultstagesofthewoodfrog( Ranasylvatica ). Ecology71:1599…1608. Berven,K.A.2009.Density-dependenceintheterrestrialstage ofwoodfrogs:evidencefroma21-yearpopulationstudy. Copeia2009:328…338. CHRISTOPHERA.SEARCYETAL. 76 Ecology,Vol.95,No.1

PAGE 10

Berven,K.A.,andD.E.Gill.1983.Interpretinggeographic variationinlife-historytraits.AmericanZoology23:85…97. Blaustein,A.R.,D.B.Wake,andW.P.Sousa.1994. Amphibiandeclines:judging stability,persistence,and susceptibilityofpopulationstolocalandglobalextinctions. ConservationBiology8:60…71. Boone,M.D.2005.Juvenilefrogscompensateforsmall metamorphsizewithterrestrialgrowth:overcomingthe effectsoflarvaldensityandinsecticideexposure.Journalof Herpetology39:416…423. Connell,J.H.,andW.P.Sousa.1983.Ontheevidenceneeded tojudgeecologicalstabilit yorpersistence.American Naturalist121:789…824. Crone,E.E.1997.Parentalenvironmentaleffectsandcyclical dynamicsinplantpopulations.AmericanNaturalist150: 708…729. Gibbons,J.W.,andR.D.Semlitsch.1991.Guidetothereptiles andamphibiansoftheSavannahRiverSite.Universityof GeorgiaPress,Athens,Georgia,USA. Ginzburg,L.R.,andD.E.Taneyhill.1994.Populationcycles offorestLepidoptera:amaternaleffecthypothesis.Journal ofAnimalEcology63:79…92. Gramapurohit,N.P.2009.Catch-upgrowthduringjuvenilelife cancompensateforthesmallmetamorphicsizein Euphlyctis cyanophlyctis .CurrentScience97:1243…1246. Heisler,I.L.,andJ.Damuth.1987.Amethodforanalyzing selectioninhierarchicallystructuredpopulations.American Naturalist130:582…602. Jerry,D.R.,T.Stewart,I.W.Purvis,andL.R.Piper.2001. Evaluationofvisualimplantelastomerandalphanumeric internaltagsasamethodtoidentifyjuvenilesofthe freshwatercray“sh, Cheraxdestructor .Aquaculture193: 149…154. Ko ¨ ppen,W.1936.Dasgeographischesystemderklimate.Pages 1…44 in W.Ko ¨ ppenandR.Geiger,editors.Handbuchder Klimatologie.VerlagvonGebru ¨ derBorntraeger,Berlin, Germany. Ku ¨ chler,A.W.1977.Themapofthenaturalvegetationof California.UniversityofKansas,Lawrence,Kansas,USA. Lande,R.,andS.J.Arnold.1983.Themeasurementof selectiononcorrelatedcharacters.Evolution37:1210…1226. Leslie,P.H.1959.Thepropertiesofacertainlagtypeof populationgrowthandthein”uenceofanexternalrandom factoronanumberofsuchpopulations.Physiological Zoology32:151…159. Lummaa,V.,andT.Clutton-Brock.2002.Earlydevelopment, survivalandreproductioninhumans.TrendsinEcologyand Evolution17:141…147. Newman,R.A.,andA.E.Dunham.1994.Sizeatmetamorphosisandwaterlossinadesertanuran( Scaphiopuscouchii ). Copeia1994:372…381. Pechmann,J.H.K.,D.E.Scott,R.D.Semlitsch,J.P. Caldwell,L.J.Vitt,andJ.W.Gibbons.1991.Declining amphibianpopulations:theproblemofseparatinghuman impactsfromnatural”uctuations.Science253:892…895. Pechmann,J.H.K.,andH.M.Wilbur.1994.Puttingdeclining amphibianpopulationsinperspective:natural”uctuations andhumanimpacts.Herpetologica50:65…84. Prout,T.,andF.McChesney.1985.Competitionamong immaturesaffectstheiradultfertility:populationdynamics. AmericanNaturalist126:521…558. Rittenhouse,T.A.G.,andR.D.Semlitsch.2007.Distribution ofamphibiansinterrestrialhabitatsurroundingwetlands. Wetlands27:153…161. Rose,K.E.,T.H.Clutton-Brock,andF.E.Guinness.1998. Cohortvariationinmalesurvivalandlifetimebreeding successinreddeer.JournalofAnimalEcology67:979…986. Scott,D.E.1994.Theeffectoflarvaldensityonadult demographictraitsin Ambystomaopacum .Ecology75:1383… 1396. Scott,D.E.,E.D.Casey,M.F.Donovan,andT.K.Lynch. 2007.Amphibianlipidlevelsatmetamorphosiscorrelateto post-metamorphicterrestrialsurvival.Oecologia153:521… 532. Searcy,C.A.,E.Gabbai-Saldate,andH.B.Shaffer.2013. Microhabitatuseandmigrationdistanceofanendangered grasslandamphibian.BiologicalConservation158:80…87. Searcy,C.A.,andH.B.Shaffer.2011.Determiningthe migrationdistanceofavagilevernalpoolspecialist:how muchlandisrequiredforconservationofCaliforniatiger salamanders?Pages73…87 in D.G.AlexanderandR.A. Schlising,editors.Researchandrecoveryinvernalpool landscapes.StudiesfromtheHerbarium,Number16. CaliforniaStateUniversity,Chico,California,USA. Semlitsch,R.D.,andJ.R.Bodie.2003.Biologicalcriteriafor bufferzonesaroundwetlandsandriparianhabitatsfor amphibiansandreptiles.ConservationBiology17:1219… 1228. Semlitsch,R.D.,D.E.Scott,andJ.H.K.Pechmann.1988. Timeandsizeatmetamorphosisrelatedtoadult“tnessin Ambystomatalpoideum .Ecology69:184…192. Smith,D.C.1987.Adultrecruitmentinchorusfrogs:effectsof sizeanddateatmetamorphosis.Ecology68:344…350. Taylor,B.E.,andD.E.Scott.1997.Effectsoflarvaldensity dependenceonpopulationdynamicsof Ambystomaopacum Herpetologica53:132…145. Trenham,P.C.,H.B.Shaffer,W.D.Koenig,andM.R. Stromberg.2000.Lifehistoryanddemographicvariationin theCaliforniatigersalamander( Ambystomacaliforniense ). Copeia2000:365…377.Werner,E.E.1986.Amphibianmetamorphosis:growthrate, predationrisk,andtheoptimalsizeattransformation. AmericanNaturalist128:319…341. White,G.C.,andK.P.Burnham.1999.ProgramMARK: survivalestimationfrompopulationsofmarkedanimals. BirdStudy46Supplement:120…138. Wilbur,H.M.,andJ.P.Collins.1973.Ecologicalaspectsof amphibianmetamorphosis:nonnormaldistributionsof competitiveabilityre”ectselectionforfacultativemetamorphosis.Science182:1305…1314.SUPPLEMENTALMATERIALAppendixA Descriptionofmethodsfordeterminingwhichcaptureeventsarerecaptures( EcologicalArchives E095-007-A1 ). AppendixB DetaileddescriptionofregressionandANOVAanalysesmentionedinthetextwithsupplementaltablesand“gures,plus additionalanalysesprovidingevidencethatrecapturedanimalsarearandomsubsetofthosethatsurvived( EcologicalArchives E095-007-A2 ). January2014 77 AMPHIBIANDELAYEDLIFEHISTORYEFFECT



PAGE 9

Panulirus argus

PAGE 10

Panulirus argus Menippe mercenaria Mithrax spinosissimus

PAGE 14

Tribolium Balanus balanoides Chthamalus stellatus C. stellatus

PAGE 16

Panulirus argus Menippe mercenaria

PAGE 17

Pagurus acadianus Pagurus pubescens P. acadianus, P. pubescens Pagurus longicarpus Pagurus pollicaris P. argus P. argus Laurencia

PAGE 18

M. mercenaria Mithrax spinosissimus Octopus spp. Opsanus tao

PAGE 19

Octopus Sargassum fluitans Crassostrea virginica

PAGE 21

Mesocosm Shelter Competition Experiment Animal Collection Experimental Setup

PAGE 22

Experimental Procedure

PAGE 23

Statistical Analysis

PAGE 24

Chemosensory Driven Shelter Selection Experimental Setup Mithrax spinosissimus

PAGE 25

Experimental Procedure Statistical Analysis

PAGE 26

Hard-bottom habitat surveys Survey Locations Statistical Analysis

PAGE 27

Stone Crab Density Manipulations Experimental Design Experimental Procedure

PAGE 28

Statistical Analysis

PAGE 30

Mesocosm Shelter Competition Experiments Influence of Chemical Cues on Shelter Choice

PAGE 31

Hard-Bottom Habitat Surveys

PAGE 32

Stone Crab Density Manipulations

PAGE 35

df P

PAGE 36

Within-subjects effects Between subjects effects

PAGE 37

F P F P

PAGE 39

*

PAGE 43

Procambarus clarkii Procambarus zonangulus Homarus americanus

PAGE 46

P. pollicaris P. longicarpus, P. pollicaris P. longicarpus

PAGE 49

Panulirus argus Panulirus argus Menippe mercenaria Panulirus argus

PAGE 50

menippe Panulirus argus Panulirus argus Panulirus argus Octopus briareus Panulirus argus

PAGE 51

Panulirus argus Chthamalus stellatus

PAGE 52

Panulirus argus Ficedula Albicollis

PAGE 53

Panulirus argus Panulirus argus Gobiodon histrio Panulirus argus,

PAGE 54

Menippe mercenaria Panulirus argus. Microtus arvalis Menippe

PAGE 55

Laurencia Panulirus argus Tribolium confusum Tribolium castaneum Tribolium Menippe mercenaria

PAGE 56

The biology of symbiosis Menippe mercenaria

PAGE 57

Thalassia testudinum ex