Citation
Journal of cave and karst studies

Material Information

Title:
Journal of cave and karst studies
Series Title:
Journal of Cave & Karst Studies
Alternate Title:
Continues NSS bulletin (OCLC: 2087737)
Creator:
National Speleological Society
Publisher:
National Speleological Society
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Geology ( local )
Genre:
serial ( sobekcm )
Location:
United States

Notes

Restriction:
Open Access - Permission by Publisher
Original Version:
Vol. 71, no. 1 (2009)

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
K26-02052 ( USFLDC DOI )
k26.2052 ( USFLDC Handle )
11254 ( karstportal - original NodeID )
0146-9517 ( ISSN )

USFLDC Membership

Aggregations:
Karst Information Portal

Postcard Information

Format:
Serial

Downloads

This item is only available as the following downloads:

Cover and Table of Contents ( .pdf )

Editorial Journal of Cave and Karst StudiesUse of FSC-Certified and Recycled Paper Malcolm S. Field ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )

Article Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Dupérré, Donald J. Buckle, and Julian J. Lewis ( .pdf )


Full Text

PAGE 1

THELEGENDOFCARBONDIOXIDEHEAVINESS G IOVANNI B ADINO Dip.FisicaGenerale,Universita `diTorino,ViaGiuria,1,Torino,Italy,badino@to.infn.it Abstract: Thefalselegendofcarbondioxidetrapsresultingfromtheweightofcarbo n dioxidegasisdisproved.Inspiteofwater-vaporlightnessincomparison withair,no water-vaportrapexistsoncaveceilings.Infact,undergroundatmospher eswithspecific compositionsarenotrelatedtogravity,buttotheabsenceofanyairmovem entaround thegassources.Theprocessofdoublediffusionofoxygenandcarbondioxi deduring organiccompounddecompositioninstillairisshowntobesignificant.Th is phenomenoncanformatmospheresthataredeadlyduetooxygendeficiencie sand poisonousbecauseofexcesscarbondioxide.Carbondioxidestoragebehav eslikealiquid andcanfloworcanbepoured,ascoldaircan,butthesearetypicaltransien tprocesses withnorelationtoacave’sfoulairformation. I NTRODUCTION Averycommonopinionamongcavers(andnotonly cavers,see(Al-Azmi,2008))isthatdensegasestendto accumulateindepressions,andespecially,atthebottomof caves.ItiswidelyacceptedthatCO 2 accumulatesatthe bottomofshafts;thisconcepthasbeeninfrequently discussed,butoftenrepeatedfrompapertopaper.Inthis paperweshowthatthisconceptisinfactfalseandmaybe regardedasanundergroundlegend(James,2003;Cigna, 2008).Theaimofthispaperistoprovideaquantitative assessmentanddetailsofgasentrapmentprocesses. G AS D ENSITIES Afirstindicationthatthosewhousetheconceptofgas densityforgasentrapmentaresimplyrepeatingit,without anyunderstandingoftheprocesses,isthetypicalstatement ‘‘theheaviergasaccumulates.…’’Obviously10kgof nitrogenareheavierthan2kgofcarbondioxide,oreven of1kgofradon.Carbondioxideisnotheavierthan oxygenorwatervapor,butitisdenserthanthosegases. Eveninpaperswherethetermdensityisused,the downwardincreaseofcarbondioxideconcentrationis qualifiedasnormal.Renault(1972)wrotethatthis phenomenonispredictablefromthelawsofphysicsand thennoticedthatinversegradientsarenotsoeasily explained.Infact,thephenomenaaremuchmorecomplex. Inaddition,thenormalCO 2 gradientthatcouldbe attributedtodensitydifferencesisundetectableonthe scaleoftypicalcavedimensions(seeEquation(A3)in AppendixA). Airiscomposedofmanydifferentgases,eachonewith adifferentmolarmass.Consideringanisothermalmassof agas(temperature T 0 ,molarmass M mol density r g ),its densityreallydependson M mol r g M mol P RT 0 1 where R 8.3142Jmol 1 K 1 isthegasconstant. Table1showsthat,dependingonmolarmass,gas densitychangessignificantlywithwatervapor,with methaneandhydrogenbeinglessdensethanair,while carbondioxideandradonaremuchdenser.Ifstratification reallydoesoccur,accordingtothemeaningadoptedby thosewhostatethatheaviergasesaccumulateindepressions,thenwewouldliveinacarbondioxideatmosphere justafewmetersabovesealevel,inoxygenatthetopof mountains,innitrogenabovethemountains,andfinally,in watervapor(andrain)inthestratosphere.TheDeadSea andtheCaspianSeawouldexistinpureradonatmosphereswhereashydrogenwouldbeconcentratedinthe ionosphere(incidentally,whereitactuallyis,butdueto reasonsotherthanstratification).Thisisanunrealistic scenario. Sododenserfluidssinkintolessdensefluids?Inthe caseofliquidstheansweriscomplex.Itisnecessarytotake intoaccountmanyeffectsconnectedwiththeinteractions ofthemoleculesinvolved.Forgases,thebehaviorismuch simplerifitisassumedthatthegasmoleculesdonot interactwitheachother(i.e.thatthegasmoleculesbehave inanidealway).Eachairmoleculeisfreetodiffusein everydirectionanditiseasytocalculatetheatmospheric structureatequilibrium. G RAVITATIONAL S TRATIFICATION Letuscalculatethepressureanddensityofagasat constanttemperature T 0 stratifiedinthegravitationalfield ( g 9.8ms 2 )onaflatsurfaceat z 0.Thehydrostatic equilibriumimposesapressure P ( z )variationwithaltitude z (positiveupward)as dP P M mol g RT 0 dz 2 Integratingbetweenthesurface(z 0, P P 0 )and(z, P (z)) yields P P 0 exp M mol g RT 0 z 3 G.Badino–Thelegendofcarbondioxideheaviness. JournalofCaveandKarstStudies, v.71,no.1,p.100–107. 100 N JournalofCaveandKarstStudies, April2009

PAGE 2

Theresultisanexponentialpressuredecreasewith altitude.Ouratmosphereisnotisothermal,obviously,but thisequationcanbeusedtodescribetheupperpartsofthe EarthÂ’satmosphere,from10to80kmwheretheair temperatureisquiteconstant( 220K).Itisnotcorrectto usethisformulaforthelowestportionoftheEarthÂ’s atmospherewherewelive,buttheequationstilldescribesthe possibilityofchemicalgasstratificationincavesverywell becauseundergroundthetemperatureislocallyconstant. Therearesomedetailsabouttheequilibriumofsuchan atmospherewhichareinterestingbecausetheytouchonthe thermalstratificationofacaveÂ’satmosphere,butthe equilibriumdetailsarenotimportantforourdiscussion. TheterminbracketsinEquation(3)mustbe dimensionless,whichintroducesthelengthscale L z for gasstratificationinagravitationalfield L z RT 0 M mol g 4 andallowscalculationofthelengthscale L z at T 0 288K. Atmospheresofpuregasesinequilibriumstratifyexponentially;meaningthateachaltituderiseof L z causesa relativereductionbyafactor e 1 0.36ofpressure.For instance,inpureoxygen,wewouldhavetorise7.6kmto reducethepressureby63%.Inpurehydrogen,therise wouldhavetobe120km,inradon1.1km,andsoon. ItisthenpossibletocalculatethenaturalCO 2 gradient enrichmentwithdepth(seeAppendixA).Thereal atmosphereisagasmixture,butgasesareindependentof eachother.Gasescollaboratetocreateafinaltotal pressure,butthepartialpressureofeachgasbehavesas iftheothersdidnotexist. Hence,theatmospherecanbeconsiderednotonlya mixtureofdifferentgases,butalsoamixtureofdifferent atmospheres,witheachatmospherecomposedofpuregas. Table2showsthatgaseshaveatendencytoseparatefrom eachotheraccordingtoaltitudedifferencesofmany kilometers.Athigheraltitudes,theairwouldbeenriched bygaseswithsmallmolarmassesanddepletedofthosewith greatermolarmasses.Asamatteroffact,however,this stratificationdoesnotexistbecauseintherealatmosphere thestrongverticalmixinginitslowerpart(calledthe homosphere,upto100km)preventssuchseparation,and createsaquiteuniformchemicalcomposition. Chemicalstratificationappearsonlyintheextreme upperlayers(heterosphere),wheretheEarthÂ’satmosphere isarrangedintofourshells,thelowerdominatedby molecularnitrogen,thesecondbyatomicoxygen,thethird byhelium,andfinallybyhydrogenatoms(Lutgens,1998). Ifwecanceltheverticalmixingandconsideranideal perfectlycalmatmosphere,wecoulddevelopdifferencesin chemicalcomposition,butonlywithlargealtitudedifferencesbecausethegasesÂ’lengthscalesarearoundten kilometers.Table2showsthat1kmabovesealevel,the differentgaseshaveessentiallythesamepartialpressure, butinthisimaginarysituationitwouldbedifficultto detectthedifferentmolarweightsofthepressures.The chemicalstratificationwouldbenoticeableonlyat z 10km(Table2). Stratificationcouldexist,butonlyinanon-mixing atmospherewithkilometersofdrop,whereasonthescale ofafewmeters,itisimpossibletodetectanychemical variation,evenbymeasuringradonconcentrations.So, watervaporandmethanedonotconcentrateoncave ceilings,butneitherdoescarbondioxideorradon concentrateoncavefloorsunlessverticalcavedimensions areontheorderofmanykilometers.Thestratification idea,repeatedthousandsoftimesfromonecavingbookto another,isabsolutelyfalse. T HE C ARBON D IOXIDE T RAPS Thereisaproblemwiththesupposednonstratification ofcarbondioxide,however,becausethestratificationeffect doesactuallyappeartoexist.OurworkintheTropics findingcavesfilledwithcarbondioxideresultedinthe developmentofspecializedshaftequipmentfordescending intoholeswhereencounteringdeadlyatmosphereswas likely(Antonini,1998). Table1.Molarmassanddensityofcommongases. Gas M mol (10 3 kgmol 1 ) r g at SP T 288K (kgm 3 ) Nitrogen281.19 Oxygen321.36 Air28.91.22 Hydrogen20.08 Methane100.42 WaterVapour180.76 CarbonDioxide381.61 Radon2229.40 Table2.Comparisonofpuregasatmosphereinneutral equilibrium. Gas Molar Mass, M mol (10 3 kg mol 1 ) Scale Length, L z (km) Pressureat z 1km P ( z )/ P 0 Pressureat z 10km P ( z )/ P 0 Nitrogen28870.890.32 Oxygen32760.880.27 Air289850.890.31 Hydrogen21200.990.92 Methane10240.960.66 Water Vapor181360.930.48 Carbon Dioxide38640.860.21 Radon222110.400.000113 G.B ADINO JournalofCaveandKarstStudies, April2009 N 101

PAGE 3

Carbondioxidetrapsdoexist,buttheirclassical interpretationisfalse.Theshaftswerefilledwithcarbon dioxidenotbecauseitisadensegas,butbecausecarbon dioxideisproducedatthebottomoftheshaftinan absolutelycalmatmosphere.Thereasonforstratificationis notgravitation,butdiffusionofgasingas.Gastrapscould alsoexistintheabsenceofgravity. Itisworthnotingthatthefalseconceptofgas stratificationdependsonthefactthatthegasispotentially deadly.Ifattheshaftbottomthereisalake,the surroundingairisfilledwithwatervapor,whichmeans thattherelativehumidityis100%.Thishighhumidityis notaresultoftheheavinessofwatervapor(whichis actuallylessdensethanair),butismerelyduetothe presenceofwater.Thewatervaporisconcentratedatthe watersurface. Inexactlythesameway,carbondioxideisconcentrated atthecarbondioxidesourcesandinthesamewaythegases ofupperatmosphereareconcentratedattheirsource.In theionosphere,theXandUVsolarradiationsproduce atomicoxygenandhydrogenandtheyaccumulatethere. Thisisthereasonwhyradonaccumulatesincellarsandin lowerfloorsandwhymethanetrapsaresetatmineconduit ceilingswherethermalstratification(hotair)creates relativelyquietbubblesformethaneconcentration. Carbondioxideisgeneratedessentiallyfromthe oxidationoforganicsubstances,ismuchdenserthanthe surroundingair,andtendstoaccumulateatthelowestcave levelsaswateroftendoes(althoughwatercanpotentially flowawaywhereasdeadorganicsubstancescannot).So, carbondioxideandwatervaportendtoaccumulateinthe depressionsthatareoftenhumidandsometimesenriched todeadlylevelswithcarbondioxide. Worse,butoftenneglected,arehypoxicconditions. Carbondioxideisdangerousathighconcentrations,but oxygen-pooratmospheres,independentofthepresenceof othergases,arealsodeadly.Eachcarbondioxidemolecule isderivedfromthereactionofacarbonatomwithan oxygenmolecule.Therefore,ingeneral,nearacarbon dioxidesourcewecanfindadeadlypresenceofcarbon dioxideaswellasadeadlyabsenceofoxygen. ACO 2 trapthusrepresentsatwo-foldphenomenon. Fororganiccompoundsontheleft,wehavetocalculate theoxygenfluxfromrighttoleftandthecarbondioxide fluxfromlefttoright. G AS D IFFUSIONIN G ASES Itisquiteeasytomodelcarbondioxidetrapformation, butitisnecessarytofirstdiscusswhyandhowagasdiffuses inacertaindirection.Gasmoleculesarequitefreetomove andtheydosocontinuously.Forasurfaceinspace,the moleculesflowthroughitinthetwopossibledirections.If theconcentration c 1 ofamoleculepervolumeunitnearone surfacesideisthesameas c 2 neartheothersurfaceside,the netfluxisthesameandnonetgastransferthroughthe surfaceoccurs.Inthisinstance,itiscommonlystated, incorrectly,thatnodiffusionoccurswheninfactthetwo diffusionscanceleachother.However,if c 1 isgreaterthan c 2 thenmoremoleculeswillflowfromsideonethanwillflow fromsidetwoasaresultofconcentrationgradients(Fig.1). Thisprocessisusuallydescribedasthegasdiffusingthrough thesurfacewithanintensityproportionaltotheconcentrationdifferenceandisknownasFickÂ’slaw. Inordertoquantifythegasdiffusionprocess,itis necessarytostatethatthetotalflux F (kilogramsper secondpersquaremeter)oversomedistance D z that separatestwogasvolumeswith c 1 and c 2 gasconcentrations(kilogramspercubicmeter)isgivenby F D g c 1 c 2 D z 5 where D g isthegascoefficientofdiffusion,whichgenerally dependsonthegasviscosity g g anddensity r g as D g f g g r g 6 where f isafactoroforderunity.FromEquation(6)we caneasilyobtainthe D g dependenceonpressure P and temperature T because D g 1 P D g T 3 2 7 Diffusionprocessesare,ingeneral,verycomplex,and onlyself-diffusion(moleculediffusivityinagasofidentical molecules)isusuallydescribed.Therealcase,inwhichone gasdiffusesinsideanother,requiresconsiderationof differentmoleculesizes,asymmetries,masses,andrepulsive forcesbetweenmolecules.Adetailedreviewofthe processesmaybefoundinJost(1952). Changesinthediffusioncoefficient D g arequitesmall. Table3,adaptedfromJost(1952),liststhisparameterat Figure1. T HELEGENDOFCARBONDIOXIDEHEAVINESS 102 N JournalofCaveandKarstStudies, April2009

PAGE 4

standardtemperatureandpressure(STP).Itshouldbenoted thatEquation(5)statesthatdiffusiondoesnotdepend,at firstorder,ongravity.Theupwardordownwarddiffusionof agasinanother,heavierorlightergas,hasthesameintensity becausethebuoyancytermisnegligible. Theequationsdescribingdiffusionprocessesarethesame asthosethatdescribeconductivethermaltransfer,transient time-dependentprocesses(Fourierequation),andstationary processes(Laplaceequation).Thermal-transferprocesses alsodescribeadiffusionprocessknownasthermal-energy diffusion(Badino,2005).Theseequationsarenoteasily solvedexactly,especiallyfortransientconditions(Nashchokin,1979),becauseevenapparentlysimpleboundary conditionscancausesignificantdifficulties. Nevertheless,itispossibletodemonstratethatminimal carbondioxideproductioninanon-mixingatmosphere cancreatedeadlyconcentrationsclosetothesource. Considerareservoir S connectedtoafreeatmosphereby alayer L of D z thicknessofnon-mixingairwitha continuouscarbondioxideproductionin S .Assumingin Equation(5) c 1 1.61kgm 3 and c 2 0thatisapuregas in S andnothingattheothersideof L ,weobtainthe maximumgasfluxwhichcanbeevacuatedbydiffusion through L F max D CO 2 1 61 D z 2 2 10 5 D z 8 Iftheproducedfluxisgreaterthan F max ,thecarbon dioxideconcentrationin S tendsasymptoticallyto saturationandalsoaccumulates L wheretheconcentration isnotlinearlydecreasingandistimeindependent.A perfectlytoxicatmosphereofalmostpurecarbondioxideis obtainedbecauseofaverysmallfluxandverycalm conditionsatthebottom(oratthetop)ofthecave. Forexample,consideraCO 2 source S extendedona flatsurfacewhichcanbeeitherhorizontalorvertical. Supposethatthesourceisabletorelease0.01kgofCO 2 perdaypersquaremeter( F 10 7 kgm 2 s 1 )thatis evacuatedbydiffusioninaquietatmosphere L infrontof S overadistanceof D z 10mwherethegasmeetsthefree atmosphere.AssumingaCO 2 concentration c 2 0 (actuallytheaverageconcentrationis383ppmvand quicklyincreasing)yields c 1 F D z D g 10 7 10 1 4 10 5 & 0 07 9 Inthiscase,givensuchaminimalflux,thecarbon dioxideconcentrationatequilibriumandnearthesource reachesroughly5%involume,avaluethatisalready dangerous.The F max isnow2.2 3 10 6 Increasingtheflux F 10 5 kgm 2 s 1 onehundred timesstrongerthaninthepreviousexample,theequation wouldyieldameaninglessvalueof7kgm 3 .Thereason forthisisthattheassumptionofalineardecreasein z of concentrationatequilibriumcannotbesatisfiedfor transferofsuchhighfluxesoversuchlongpathsand F F max .Fortheseconditions,atthefirstfewmetersin frontof S ,theCO 2 concentrationwouldbeextremelyhigh, andonlynearthefreeatmospherewouldtheconcentration decreasealmostlinearlytozero. Theweight(density)ofcarbondioxidedoesnotmatter. Exactlythesameapproachcanbeusedforoxygenor nitrogen(ormethane,incoalmines)sourcestoobtain almostpuregasatmospheresnearthesource.Thesame couldbesaidforawater-vaporsource,butthisgasisfar fromperfectatroomtemperature(i.e.,itsbehavioris highlycomplicatedbyprocesseslikesaturation,condensation,enthalpyreleases,eddies,etc.).Theproblemofwatervapordiffusioninstableatmospheresisverycomplex. ThekeypointincreatingaCO 2 trapundergroundisthe thermaluniformityofcaveatmospheresthathampersair movementsandwhichcancreatetrapsofquietcoldair.No heavygasstratifiationoccurs,butthermalstratifiation does.EkandGewelt(1985)includesseveralincorrect commentsontheroleofgravityandnormalgradient increasingdownward,buttheimportanceofthermal stratifiationiscorrectlynoted.EkandGeweltalsonote theapparentlysurprisingfactthat‘‘…inasingleroomor gallerythe p CO 2 isfrequentlyhigherneartheroofthan neartheground.’’ D OUBLE D IFFUSION Therearemanystudiesexplainingcarbondioxide enrichmentincaves(see,forexampleJames,2003and referencestherein).Themain,andmostdangerous,reason forcarbondioxideenrichmentincavesistheoxidationof organiccompoundsinmotionlessatmospheres(Ekand Gewelt,1985).Thisleadstoanotherundergroundlegend: thatcarbondioxideisabletoextinguishflames. Oxidationanddegradationofanylongorganicmolecule causesanentropyincreaseinonedirection.Eachcarbon atomandeachhydrogenatominanorganicwastewillbind tooxygentoflowawayandreturnaswood,skin,milk,etc. Inopen-air,theprocessisgenerallyveryfastandishelpedby countlesstypesofcreatures(fungi,bacteria,birds,jackals, humanbeings,etc.)thatliveoffthedegradationofthese deadstructures,andareabletoquicklydistributeitto Table3.Diffusioncoefficientsofcommongases. GasinGas D g (m 2 s 1 ) O 2 inO 2 1.89 3 10 5 N 2 inN 2 1.98 3 10 5 CO 2 inCO 2 1.04 3 10 5 O 2 inAir1.78 3 10 5 CO 2 inAir1.38 3 10 5 H 2 OinAir2.36 3 10 5 G.B ADINO JournalofCaveandKarstStudies, April2009 N 103

PAGE 5

surroundinglife.However,whentheorganiccompounds stayinaverystablesituation,farfromthedegradation workers,asinacave,onlytheSecondLawofThermodynamicscanwork,anditworksslowlybydiffusion,byslow airdraughts,andbyminimaltemperaturedifferences. Ingeneral,acarbondioxidesourcedoesnotexist,but someorganicresidues(i.e.,vegetablematter)areincontact withair.Itispossibletoseethatif(1)theoxygenand carbondioxidejustdiffuseinthesamesurrounding atmospherewithoutanydraughtand(2)theoxidation rateisproportionaltotheoxygenconcentration(see AppendixB).Thereisthenacorrespondencebetweenthe oxygenandcarbondioxideconcentrationsaroundthe organicstorageasshownby c CO 2 ,0 1 170 28 c O 2 ,0 10 Calling F id theoxidationrateof S infreeatmosphere whentheoxygenconcentrationis c O,1 0.28kgm 3 ,the actualoxidationrateisslowerduetotheloweroxygen concentration F F id 1 1 F id D z 0 28 D O 2 e F id 11 Thereductionfactor e dependsonthegeometryof diffusingatmosphere.Itisdirectlyconnectedwiththe oxygenconcentrationnearthedeposit,givenby c O,0 0 28 e 12 Returningtothepreviousexampleandassumingthat thefluxused,10 7 ,is F id ,itispossibletocalculatenew valuesthatresultfromtheeffectofdoublediffusiontogive F 0 85 F id c O 2 ,0 0 85 c O 2 ,1 c CO 2 ,0 0 06 13 Theoxidationfluxisreducedincomparisontofree atmosphere,andcorrespondingly,theconcentrationsof oxygenandcarbondioxidearereduced.Forextreme conditionsandif F id isveryhigh, F cannotincreaseover themaximumvaluegivenby: F & 5 0 10 6 D z 14 withconcentrations: c O 2 ,0 & 0 15 c CO 2 ,0 & 0 35 16 Thismeansthatinthecaseofoxidativeprocesses,the maximumcarbondioxideconcentrationaroundthesource isapproximately21%involumeasoxygeninfreeair. Thesearetypicalconditionsinwhichaflamewill extinguish,notbecauseofthehighcarbondioxide concentration(whichdoesnottakepartincombustion, asdoesnotnitrogen),butbecauseoflowoxygenpartial pressure.Flamesburnonlyifthereisasufficientoxygen concentration.Infact,ifcarbondioxidedoesnotcome fromoxidation,asitusuallydoes,butfromothersources likereactionofsulphuricacidwithcarbonatesorvolcanic gases,theatmospherescanbeverydangerousandthe flamesburnverywell(Mucedda,1998). E NTHALPY R ELEASEAND T RAP S TABILITY Thelogicofaverystableatmospherearound S isan issue.Oxidationalwaysinvolvesenthalpyreleasesand temperatureincreasescausingconvectiveprocessesthat transferairmuchmoreefficientlythandoesdiffusion. Thereactionoforganicsubstanceswithoxygen producestypically3 3 10 7 Jkg 1 .Aspreviouslydemonstrated,thereactionof1kgofmatterreleases3.1kgof carbondioxide,withaconsequentenergyreleaseof10MJ kg 1 ofcarbondioxide.Foraflux F (kilogramsofcarbon dioxidepersquaremeterpersecond)wethenhave I O 2 & 10 7 F Wm 2 17 Watermoleculesthatareproducedduringthereaction formagas,andingeneral,havetoreleasethevaporization enthalpy L w whereitcondenseson S oronfartherwallsto obtain: I w 1 3 3 1 FL w 10 6 F Wm 2 18 where I w issignificantlysmallerthan I O 2 Thefluxassumedinthepreviousexamples( F 10 7 kgm 2 s 1 )releasesanenergyfluxapproximately 1Wm 2 .Thisenergyfluxcanincreasethetemperatureof airandthenforcegasmigrationmoreefficientlythan diffusiontotransfercarbondioxideandoxygen,butthe rateoftemperatureincreaseandgasevacuationdepends ontheshapeofthesystem.Forexample,asimilarenergy releaseisabletoincreaseinaday100kgofairby1K,or 100kgofmoistvegetablesofabout0.2K,butitcanonly evaporate0.04kgofwater.Ingeneral,thisisaverysmall energyreleaseanditseffectisdependentonthesystem shape,suchasorganicdepositorientation,reactivityand depth,waterpresenceinthedeposit,andheatexchanges.If thetemperaturedifferencesbetweentheorganicdeposit andthecavearearound0.1–0.5K,theyforceair convectionsofafewcentimeterspersecond,whichare muchmoreefficientthandiffusiontoevacuatesuchsmall fluxesofcarbondioxide. Thesituationthatappearsmostfavorabletousethis energytotriggerconvectivemovementsiswhen S isona verticalwall.Theheatedairformeddiesinfrontof S ,and T HELEGENDOFCARBONDIOXIDEHEAVINESS 104 N JournalofCaveandKarstStudies, April2009

PAGE 6

atthesametime,thewaterflowsawayfromittoreduceits coolingrole. Theleastfavorablesituationuseofthisenergyoccurs when S isontheceiling(or,morereasonably,inaclosed ascendingcavebranch).Theenergyreleasedcreatesa thermalsedimentationthattrapsairinthehotbubbleand onlydiffusioncanevacuategasesfromthere,evenifthe entranceisquitelarge.Theescaperoutefromasimilartrap isdownwardanditisthenquiteeasytogetawayfromit withlittleeffortandwithouttheneedforintensebreathing. Thepresenceofwaterthusallowsforarelatively isothermalreactionandcontributestosystemclosure, becausethetemperaturedifferencesarenotsufficientto createeddies.Inanycase,itappearsthatextremelylowair fluxesaresufficienttopreventtheformationofcarbon dioxide-richoroxygen-pooratmosphereinsidecaves.In actuality,theseconditionsarequiterare. T RANSIENT C ONDITIONS Theprecedingdiscussioncoveredsteady-statesituationsinwhichtheparametersdonotdependontimeand quietconditions.Forthosesystems,itwasshownthateven whenessentiallyatequilibrium,itispossibletocreate exoticmicro-atmospheres. Asuddengasemissionofonecubickilometerofcarbon dioxidefromNyosLakeisnotaprocessatequilibrium (Sigvaldason,1989;Tazieff,1989;Evansetal.,1993),andit isverysimilartothecoldairbubblethatfallsonourlegs whenweopenarefrigerator.Thesearetypicaltransient situations.Acarbondioxide-filledcupisverysimilartoa hotstone,whichisgoingtocool,slowly,accordingtoa similardiffusionlaw. Ahotstoneisnotstableinthelongtermbecauseitisin astateofdisequilibriumandwilleventuallyreachafinal stablestateoverarelativelylongperiodoftime.Acarbon dioxidereservoirbehavesinasimilarway.Overalongtime scale,itisunstableandtendstodiffuseintotheatmosphere withinanaltitudeofadozenkilometers.Onamuch shortertimescalethanthetimescaleofequilibriumdrift, thegeneralbehaviorchanges.TheFourier(notthe Laplace)equationhastobeusedandourapproachto studytheasymptoticstatedoesnotdescribetheprocess. Thismeansthatifweproduce,insomeway,thefilledcup, itsgaswillremainthereforsomeperiodoftime,likethe coldairinasupermarketfreezer,butthesituationis unstable.Justlikethefreezersituationisusuallystabilized bycontinuousair-cooling,thecarbondioxidetrapcanbe stabilizedbyagassourceresultinginareturntostationary physics.However,ifthesesourcesareabsent,thesystems evolvetothemaximumentropystate,withoneprogressing touniformtemperature,andtheotherprogressingto completegasmixing.Inthiscase,gaswillthendiffuseaway tofilltheEarth’satmosphereveryslowly,butitcanbe pouredlikealiquidorflowalongagalleryfloor. C ONCLUSIONS Confusionbetweenstationaryandtransientconditions hascreatedafalseundergroundlegendofgasentrapment, whichobscuresrecognitionofthetrueprocessesthat producecarbondioxide,methane,andradontrapsincaves andmines.Thebasicconceptshavebeenfurtherconfused bythefactthatifcarbondioxideisproducedbyoxidation nearitssource,thereisnotonlyhighcarbondioxide concentration,butalsoaverylowoxygenconcentration, whichleadstotheoccurrenceofflameextinctionsand similarevidencesofpooratmospheres. Thetrapsareessentiallyduetoaccumulationneara source(whatevertheorigin)inmotionlessatmospheres.The up-downgradientsaregenerallydueto(1)preferredpointof organicaccumulationand(2)airthermalstratifiationthat createsamotionlesstrapofcoldorwarmair. Structure,periodicity,andintensityoftrapsdependon organicmatterinflow,thermalstratifiation,andshapeof thecavity. R EFERENCES Al-Azmi,D.,Abu-Shady,A.I.,Sayed,A.M.,andAl-Zayed,Y.,2008, IndoorRadoninKuwait:HealthPhysics,v.94,p.49–56. Antonini,G.,andBadino,G.,1998,TecnicheSpecialiediAutosoccorso: Erga,Genova,288p. Badino,G.,2005,UndergroundDrainageSystemsandGeothermalFlux: ActaCarsologica,v.34,no.2,p.277–316. Cigna,A.,andBadino,G.,2008,IndoorRadoninKuwait:Comment: HealthPhysics,v.95,no.2,p.255–256. Ek,C.,andGewelt,M.,1985,Carbon-dioxideincaveatmospheres:Earth SurfaceProcessesandLandforms,v.10,no.2,p.173–187. Evans,W.C.,Kling,G.W.,Tuttle,M.L.,Tanyileke,G.,andWhite,L.D., 1993,GasBuildupinLakeNyos,Cameroon:TheRechargeProcess anditsConsequences:AppliedGeochemistry,v.8,p.207–221. James,J.,2003,CarbonDioxide-EnrichedCaves, in Guun,J.,ed., EncyclopediaofCaveandKarstScience,London,FitzroyDearbon, p.183–184. Jost,W.,1952,DiffusioninSolids,Liquids,Gases,NewYork,Academic Press,558p. Lutgens,F.,andTarbuck,E.,1998,TheAtmosphere,7thEdition:N.J., Prentice-Hall,544p. Mucedda,M.,1998,GrotteepozzidellaNurradiSassari:Bollettino GruppoSpeleologicoSassarese,v.17,p.10–26. Nashchokin,V.,1979,EngineeringThermodynamicsandHeatTransfer, Moscow,MIRPublisher,572p. Renault,P.,1972,Legazdescavernes:Decouverte,v.2443,p.12–18. Sigvaldason,G.E.,1989,ConclusionsandrecommendationsofInternationalconferenceonLakeNyosdisaster:JournalofVolcanologyand GeothermalResearch,v.39,p.97–109. Tazieff,H.,1989,MechanismsoftheNyosCarbonDioxideDisaster: JournalofVolcanologyandGeothermalResearch,v.39,p.109–116. A PPENDIX A T HE N ATURAL C ARBON D IOXIDE G RADIENT Theairpressureinanisothermalatmosphere,inneutral equilibrium,atheight h [km]asgivenbyEquation(5)and Table2,leadsto P air P 0 exp h 8 5 A1 G.B ADINO JournalofCaveandKarstStudies, April2009 N 105

PAGE 7

Similarly,inapureCO 2 atmospherethepressurebehaves accordingto P CO 2 P 0 exp h 6 4 A2 Theratiobetween P CO 2 and P air inthepreviousequations givestheidealcarbondioxideimpoverishmentwith altitudesothatwecanthenobtain: P CO 2 P air exp h 8 5 h 6 4 exp h 26 A3 Thescalelengthofreductionofcarbondioxideconcentrationduetobuoyancyisthenaround26km.Forthese conditionsandwithacarbondioxideconcentrationof3% atthefloor,itisnecessarythattheceilingbelocatedat 10.4kmofaltitudetoreach2%concentration.Thisisin factthetruenaturalCO 2 gradientunderidealconditions. A PPENDIX B D OUBLE D IFFUSION Usinganorganicdeposit S andsomemass M per squaremeter,thechemicalcompositionoforganicmatter isveryroughlyC n H 2n .Putrefaction(oxidation)ofthese moleculeseventuallyis 3 2 O 2 CH 2 [ CO 2 H 2 O B1 whichintermofreactingmassesgives 3 4 M O 2 M [ 3 1 M CO 2 1 3 M H 2 O B2 Volumetrically,thismeansthatputrefactionofone kilogramofvegetablemassneedstheoxygeninabout12 cubicmetersofairandemitstwocubicmetersofcarbon dioxide.Mostimportantly,itneedsanoxygensupply. Itisreasonabletoassumethattheoxidationrateofa compoundisproportionaltothelocaloxygenconcentrationin S .ThereforetheproducedfluxofCO 2 isgivenby: F K O 2 c O 2 ,0 B3 where c O 2 ,0 istheoxygenconcentrationnearthecompounds. Theproportionalitydependsonexposedsurface,compoundtype,reactivity,temperature,presenceofbacteria, totalmassof S ,etc.Calculationof K O isbasicallyimpossible, butsome K O doesexist.Wecanassumethatthefree atmospherethatfeedsoxygento S isalsothesame atmospherethatevacuatescarbondioxide(itisnotan obviousassumption,butingeneral,itistrue),sowecanalso assumethatthefreeatmosphereisatadistance D z from S Atequilibrium,systemparametersdonotdependontime (thesystemisstationary)andtherateofCO 2 molecules evacuationmustequalexactlytheincomingoxygenmolecules,aswellasthecarbondioxideproduction. Theconcentrations c CO 2 ,0 and c O 2 ,0 arethegas concentrationsin S ,andtheconcentrations c CO 2 ,1 and c O 2 ,1 arethoseinthefreeatmosphere.Wemustalso accounttheforthedifferentmassesinvolved,and assumingthatthefluxisnotsostrongastodestroythe lineardistribution,wehave F CO 2 D CO 2 c CO 2 ,0 c CO 2 ,1 D z B4 forthediffusedcarbondioxidefluxfrom S to z 0,and F O 2 D O 2 c O 2 ,0 c O 2 ,1 D z B5 forthefluxofdiffusedoxygento S .Themassbalance showninEquation(B2)gives F O 2 3 4 3 1 F CO 2 B6 Atsealevelwecanassumethat c O 2 ,1 0 28kgm 3 c CO 2 ,1 0kgm 3 B7 sothatwecanthenobtainarelationbetweenthecarbon dioxideandoxygenin S D O 2 0 28 c O 2 ,0 D z 3 4 3 1 D CO 2 c CO 2 ,0 D z B8 Then,withnumericalvalues c CO 2 ,0 1 170 28 c O 2 ,0 B9 setting F id theoxidationrateof S infreeatmosphere(i.e., whentheoxygenconcentrationis c O 2 ,1 0 28kgm 3 ).We expectthattherealoxidationratetobeslowerduetothe loweroxygenconcentrationwhichisproportionalto K O 2 F O 2 K O 2 c O 2 ,0 F id c O 2 ,0 c O 2 ,1 B10 Combiningandrearranging F F id 1 1 F id D z 0 28 D O 2 e F id B11 c O 2 ,0 c O 2 ,1 F F id 0 28 e B12 Thecarbondioxideconcentrationin S issimply c CO 2 ,0 F D z D CO 2 e F id D z D CO 2 B13 T HELEGENDOFCARBONDIOXIDEHEAVINESS 106 N JournalofCaveandKarstStudies, April2009

PAGE 8

Thatis,carbondioxideconcentrationisalsoreducedbya factor e whencomparedwiththeconcentrationthatwe wouldobtaininthecaseofsinglediffusion. Wefoundareductionfactor e thattakesintoaccount thelimitedtransferpossibilitiesduetothedoublediffusion whichshowedthattheactualfluxratio F/F id onlydepends ontheoxygentransfercapacityoflocalatmosphere.If D O 2 D zisverylarge(thatis,thegasesdiffuseveryeasily)in comparisonwith F id ,then e 1and F isnear F id asitmust be.Theequationsthenreducetotheconditionsforsingle diffusion.So,thelimitingbehaviorofdoublediffusion needonlybeconsideredwhendealingwithhighreactivity and,correspondingly,withhighCO 2 concentrations. Itiseasytoshowthatfortheseconditions,oxidation ratetendstozero,thereisnooxygennearthesource (hypoxicatmosphere),andflamessmother. Equation(B11)canberewrittenas F 1 1 F id D z 0 28 D O 2 B14 andif F id ishigh, F reachesitsmaximumvalueaccordingto F & 5 0 10 6 D z B15 c O 2 ,0 & 0 B16 c CO 2 ,0 & 0 35 B17 whichmeansthatinthecaseofoxidativeprocesses,the maximumcarbondioxideconcentrationaroundthesource isapproximately21%involume. G.B ADINO JournalofCaveandKarstStudies, April2009 N 107



PAGE 1

MONKSEAL( MONACHUSMONACHUS )BONESINBEL TORRENTECAVE(CENTRAL-EASTSARDINIA)ANDTHEIR PALEOGEOGRAPHICALSIGNIFICANCE J O D E W AELE 1 ,G EORGE A.B ROOK 2 AND A NKE O ERTEL 3 Abstract: Fragmentsofmonksealbones( Monachusmonachus )discovered7–12m belowwaterlevelinBelTorrenteCave(central-eastSardinia)in2004hav ebeenAMS radiocarbondated.Thebones,probablyofdifferentindividuals,haveca libratedages rangingfrom5000–6500calendaryearsB.P.andallowreconstructionofth e paleogeographyofthecaveandthesurroundingareaduringthistimeperio d.Monk sealslivinginlargenumbersalongtheSardiniancoastusedthecaveforsh elterandto givebirthtotheirpups.Thelowersealevelofthemid-Holocene,combined withcave morphology,allowedthemtoreachfarintothemaintunnelofthecave.Thel arge numberofbonesfoundofapproximatelythesameageseemstoindicatethatt hemonk sealsusedcaveseithertoshelterfromstormwavesortoescapefromnatura lpredators duringperiodswhenhumandisturbanceofthecoastwasminor.Thiscouldsu ggestthe monksealshadotherpredatorstheywerealsotryingtoavoid. I NTRODUCTION Duringthesummerof2004,scubadiversexploringBel TorrenteCave,oneofthemostinterestingsubmarinekarst resurgencesintheGulfofOrosei,central-eastSardinia, discoveredseveralskeletonsofmonkseals( Monachus monachus )inanunderwaterpassage.Theskeletonswere 750mfromthecaveentranceand8–12mbelowthewater surface(Sgualdini,2004).Ageomorphicstudyofthecave andAMSradiocarbondatingofsomemonksealfinger andtoeboneswereundertakeninanattemptto reconstructtheenvironmentalconditionsatthetimethis remarkableconcentrationofsealbonesaccumulatedin whatarenowsubmergedpassages. M ONK S EAL B IOGEOGRAPHY Recentgeneticstudiessuggestthatmonkseals(genus Monachus )originatedintheTethysregionduringthe Tortonianage(ca12Ma),andsincehaveoccupiedthe temperatewatersoftheMediterranean(Mediterranean monkseal, Monachusmonachus ).Theythenspreadfrom easttowesttotheCaribbeanfirst(Caribbeanmonkseal, Monachustropicalis ,nowextinct),andthentothePacific Ocean(Hawaiianmonkseal, Monachusschauinslandi endemictotheHawaiianIslands)(Fyleretal.,2005). Intherecentpast,Mediterraneanmonksealswere presentalongcoastsfromtheBlackSeathroughtheentire MediterraneantotheAtlanticshoresofMoroccoand reachingasfarsouthasGambiaandwestwardstothe Azores(Johnsonetal.,2008).Monksealswereoften mentionedduringtheGreekandRomanPeriodsas occurringalongrockyshorelinesandalsoonbeaches. Sinceancienttimes,theanimalwashuntedforitsskin, meat,fat,andoil,butitwasonlyinRomantimesthatthe sealpopulationwasseriouslydepleted.Therewasapartial recoveryinnumbersafterthefalloftheRomanEmpire, butmonksealswereagainendangeredduringtheMiddle Ageswheretheysoughtshelteralonginaccessiblecoasts andofteninseacaves,someonlywithunderwater entrances.TheinaccessiblecoastsofSardiniamusthave beenidealplacesforimportantpopulationsofmonkseals tosettle(BarehamandFurreddu,1975).Thevastterritory formerlyoccupiedbymonksealswasrapidlylimitedbythe increasinguseandoccupationofcoastalareasbyhumans. Consequently,theanimalhasalmostcompletelydisappearedfromFrance,Italy,Spain,Egypt,Israel,and Lebanon.Althoughtherearestillsporadicsightingsof monksealsalongsomepartsofthesecoasts,theredonot appeartobepermanentpopulations(Johnsonetal.,2008). Today,themajormonksealpopulationsarefound alongtheCaboBlancopeninsula(WesternSaharaMauritania)(SamaranchandGonzale `z,2000;Aguilaret al.,2007;Borrelletal.,2007),theDesertasIslandsof Madeiraarchipelago(Karamandlidisetal.,2004;Pireset al.,2007),theMediterraneancoastbetweenMoroccoand Algeria(Borrelletal.,1997),theCilicianbasininTurkey (Gucuetal.,2004),andinCyprusandtheGreekIslands (Dendrinosetal.,2007a;2007b).Monksealsarestill occasionallysightedalongtheSardiniancoast,butthelast permanentresidentsdatebacktoatleast30yearsago. BeforeWorldWarIImonksealswereregularlyhuntedby localfishermen,butduringthe1950stherewerestilltensof sealsalongthecoast(Altara,1995;Johnson,1998).This numbercontinuedtodecreaseduetohunting,butalso 1 DipartimentodiScienzedellaTerraeGeologico-Ambientali,University of Bologna,ViaZamboni67–40127Bologna,ITALY.E-mail:jo.dewaele@unibo .it 2 DepartmentofGeography,UniversityofGeorgia,AthensGA30602,U.S.A.E mail:gabrook@uga.edu 3 Erentrudisstr.19/11,A5020,Salzburg,AUSTRIA.E-mail:anke.oertel@g mx.at J.DeWaele,G.A.Brook,andA.Oertel–Monkseal( Monachusmonachus )bonesinBelTorrenteCave(central-eastSardinia)andtheir paleogeographicalsignificance. JournalofCaveandKarstStudies, v.71,no.1,p.16–23. 16 N JournalofCaveandKarstStudies, April2009

PAGE 2

becauseofincreasedtourism,withthefamousBueMarino Caveopeningforvisitsin1960(Ariscietal.,2000). Tourismisoneofthemostimportantdisturbancesinkarst areasinthecentral-easternpartofSardiniaandmonkseals havebeenamongthefirsttosuffer(DeWaele,2008).The lastmonksealreportedintheBueMarinoCavewaskilled byafishermanin1970,andabouttenindividualswereseen attheGrottadelFico,afewkilometerssouthofBue Marino,intheearly1970s(BarehamandFurreddu,1975). B EL T ORRENTE C AVE E XPLORATION TheBelTorrenteCaveislocated0.5kmnorthofCala Sisine(Fig.1).Thecavewasdiscoveredandexploredby JochenHasenmayerinthe1970sandthefirst500mwas surveyedinthe1990s(Fancelloetal.,2000;Morlockand Mahler,1995).Cavedivingexpeditionsin2003,2004,and 2006exploredandmappedthecavetomorethan3km. Thesidebranchwiththelargestnumberofmonkseal boneswasdiscoveredinthesummerbytwocavedivers (LucaSgualdiniandEnricoSeddone)workingforthe divingclubatSantaMariaNavarrese(Sgualdini,2004). Thecavewassurveyedwithawrist-heldcompass. Distancesweredeterminedusingtagsonthesafetyline spacedat5-meter-intervals.Depthwasmeasuredwithboth analog-anddigital-depthgauges.Atsurveypoints, distancestothecavefloorandroofwereestimatedwith anaccuracyofabout1m.Overallprecisionofthecave planandprofileisaround1%. M ORPHOLOGY BelTorrenteCaveischaracterisedbya5–20mwide tunnelwithanaverageheightof5mandadepthof12m (Fig.2).Thecaveextendstothesouthwestforthefirst 550mandthereareseveralair-filledpassagesseparatedby shortsumps.Thenthepassagehasa22-m-deepsump (SifoneCentraleorCentralSump)thatallowsaccessto Figure1.AerialphotographoftheBelTorrenteCavearea.Cavepassagesar eshowninblack;theellipsemarkstheRamodel Buearea. J.D E W AELE ,G.A.B ROOK AND A.O ERTEL JournalofCaveandKarstStudies, April2009 N 17

PAGE 3

anotherair-filledchamberwhereadeepandonlypartially exploredsumpstartsandaby-passgivesaccesstoaseries ofair-filledgalleries.BeforetheSifoneCentrale,thereare twosidepassagestotheleft.Thefirstsidepassageleadsto theSpiaggiadelBue(oxbeach),wherebonesofmonkseal havebeenfoundonthesandyfloorat3–4mdepthand otherremainsofsmallervertebratesinseveralplacesonthe rockyfloorapproximately1mabovesealevel.These boneshavenotbeensampledanddated. Thesecondsidepassage,theRamodelBue(oxgallery) (OertelandPatzner,2007;Sgualdini,2004),isentirely underwateranddepartsfromthe SifoneCentrale at10m belowsealevel.A3-m-widetunnelleadstothesouthandis 7–22-m-deepwiththeshallowersection( 7mmeasuredat bottomofthegallery)located50–100mfromthemain tunnelandischaracterizedbyalargeflowstoneentering fromabove(Fig.3). Thefloorsofthemaintunnelandthesidepassagesare coveredwithsandsandgravelscontainingbothlimestone andgranitefragmentswithfewfinesedimentssothateven afterdivershavepassedthroughthemwaterinthe passagesremainsrelativelyclear.Thelackoffinesediments isrelatedtotheregularflushingofthecavebyfreshwater floods.Duringnormalconditions,thedischargeof freshwaterthroughthecaveisonlytensoflitersper secondsothatthewatercurrentishardlynoticeable.Near theentrance,andupto200–400minsidethecave (dependingonseaandclimateconditions),thereisa haloclineat1–2mofwaterdepth(OertelandPatzner, 2007).Freshwaterformsa‘‘surfaceblanket’’overbrackish andseawater.Afterheavyrains,themaintunnelisflooded entirelybyfreshwaterandflowvelocitiesareupto2ms 1 (MorlockandMahler,1995).Thesefloodstransportclastic deposits(includingfinesediments)fromthecaveand erode/corrodethewallsofthetunnel.Asaresult,thefloor, ceiling,andwallsdisplaytypicalphreaticerosionand corrosionfeatures.Inseveralplaces,speleothems(flowstones,stalagmites,andstalactites)arepresentabovewater andalsoseveralmetersbelowpresentsealevel.Thesehave beenintensivelycorrodedanderodedbyfloodwaters belowsealevelandalsouptoatleastonemeterabovesea level. ThemorphologyoftheBelTorrenteCavegenerally resemblesthatofthenearbyBueMarinoCave,exceptthat thepassagesofBelTorrentearemainlyunderwater(De WaeleandForti,2003).Thisdifferencemaybedueto neotectonicactivitythatresultedinthesouthwardstilting oftheTyrrheniantidalnotch,datedto125,000yearsB.P. andranginginheightbetween10.5ma.s.l.atCalaGonone and7.7ma.s.l.atSantaMariaNavarrese(Antoniolietal., 1999).Thisslighttiltingcouldberesponsibleforthe altitudedifferencebetweentheBelTorrenteandBue Marinocaves(DeWaele,2004;FortiandRossi,1991).If true,theBelTorrenteCavesystempredatesthetilting,and thereisevidencesuggestingthatthemainperiodofcave formationwasmorethan3Ma.Oneconvincingpieceof evidenceisPlio-Pleistocenebasalts,datedbetween2–3Ma (Savellietal.,1979)thatfillkarstconduitsoftheBue Marinomaingallery,indicatingakarstphaseolderthan thisvolcanicactivity,whichisthoughttobeofMioPlioceneage(DeWaele,2004;Mahler,1979). DuringtheQuaternary,changesinsealevelresultedin periodicdryingandfloodingofcavesalongthecoast.The mostrecentdryingepisodewas22–18kaB.P.whensea leveldroppedapproximately125meters.Fromrecent studies,especiallyoncavestalagmites,postglacialsealevel hadalreadyrisento6–10mbelowpresentbyabout6.5ky B.P.(Antoniolietal.,2004),thusleavingmostoftheBel Torrentegalleriesabovewater.Asaresult,5–6kaBel TorrentemayhaveresembledthepresentBueMarino Cave,withanundergroundriverflowingoutofthe mountainsandeasilyaccessibleforatleast550meters. SealevelcontinuedtoriseinthemidtolateHolocene reaching0.5–1mbelowsealevel2kyB.P.duringRoman times. T HE S EAL C EMETERY Severalmonksealskeletonswerefoundintheshallow partoftheRamodelBuepassage,50–100mfromthemain gallery(720–790mfromtheentrance).Bonesandskullsof atleastfivemonksealshavebeenfoundatdepthsof8– Figure2.PlanofBelTorrenteCave.Theellipsedefinesthe areafromwhichmonksealboneswereobtained. M ONKSEAL ( M ONACHUSMONACHUS ) BONESIN B EL T ORRENTE C AVE ( CENTRAL EAST S ARDINIA ) ANDTHEIRPALEOGEOGRAPHICALSIGNIFICANCE 18 N JournalofCaveandKarstStudies, April2009

PAGE 4

12m,restingonthesandyfloororfalleninfissuresorholes alongthewalls(Figs.3and4).Thecavediverswhoexplored thepassagereportseeingthewatersurfaceinthisareaso thattherecouldbeanair-filledchamberabovetheflooded passage.Althoughonlyfiveskullshavebeencounted,more couldbeburiedbeneathsand,trappedinnichesalongthe walls,orinapossibleair-filledchamberabove. S EAL B ONE A GES S AMPLING Foursamplesofsmallfingerand/ortoeboneswere collectedfromskeletalmaterial20–70mfromtheentrance oftheRamodelBuebranchpassage(720–790mfromcave entrance),atdepthsof7.6–12m(Table1andFig.4). Figure3.FiveskullsofmonksealdiscoveredintheRamodelBue:A.Skullon asandyfloorinthecenterofthepassage(cave diverforscale);B.Skullandboneswithablackcoatingdepositedinafiss ureonthetunnelwalls;C.Jawwithblackcoating andsomespinalbonesonasandyfloorofasideniche;D.Smallblackenedjaw lyingonbarerock;otherbonescanbeseenin theback;E.Jawandboneswithblackcoatingonthebarerocksurfaceinalat eralalcove. J.D E W AELE ,G.A.B ROOK AND A.O ERTEL JournalofCaveandKarstStudies, April2009 N 19

PAGE 5

Smallerboneswereselectedforstudyasthesewerelarge enoughtocontainenoughbonecollagenfordating,which allowedleavingtheskullsandlargerbonestoremain intact.Whencollected,thefragmentswerelabelledandput inplasticbagstogetherwiththewater.Allofthebone fragmentshadadarkbrownpatina,andalthough composedofdenserbonematerial,wererelativelyfragile. Inthelaboratory,sampleswerelefttodryforseveralweeks andoftenlostconsistency. R ADIOCARBON D ATING T ECHNIQUES Todeterminetheagesofthemonksealbones,bone apatite(bioapatite)andbonecollagenweredated.The boneswerecleanedbyabrasionandwashedusingan ultrasonicbath.Thecrushedbonewastreatedwithdiluted 1Naceticacidtoremovesurface-absorbedandsecondary carbonates.Periodicevacuationensuredthatevolved carbondioxidewasremovedfromtheinteriorofthe samplefragments,andthatfreshacidwasallowedtoreach eventheinteriormicro-surfaces.Thechemicallycleaned samplewasthenreactedundervacuumwith1NHClto dissolvethebonemineralandreleasecarbondioxidefrom bioapatite. Thecrushedbonewasthentreatedwith1NHClat4 u C for24hours.Theresiduewasfiltered,rinsedwith deionizedwater,andunderslightlyacidconditions(pH 3)heatedat80 u Cfor6hourstodissolvecollagenand leavehumicsubstancesintheprecipitate.Thecollagen solutionwasthenfilteredtoisolatepurecollagenanddried out.Thepurifiedcollagenwascombustedat575 u Cinan evacuated,sealedPyrexampouleinthepresenceofCuO. Theresultingcarbondioxidewascryogenicallypurified fromtheothercombustionproductsandcatalytically convertedtographiteusingthemethodofVogeletal. (1984).GraphiteC 14 /C 13 ratiosweremeasuredusingthe 0.5MeVacceleratormassspectrometerattheCenterfor Figure4.MonksealbonesintheRamodelBue:(A)Ribandvertebraonbareroc konthesideofthepassage;(B)Depositof longandshortbonesinalateralfissure;(C)Smallbone,probablytoe,ina sandyfissure;(D)Vertebraandotherbonesina lateralalcove. M ONKSEAL ( M ONACHUSMONACHUS ) BONESIN B EL T ORRENTE C AVE ( CENTRAL EAST S ARDINIA ) ANDTHEIRPALEOGEOGRAPHICALSIGNIFICANCE 20 N JournalofCaveandKarstStudies, April2009

PAGE 6

AppliedIsotopeStudiesattheUniversityofGeorgia.The sampleratioswerecomparedtotheratiomeasuredfrom theOxalicAcidIstandard(NBSSRM4990).SampleC 13 / C 12 ratiosweremeasuredseparatelyusingastableisotope ratiomassspectrometerandexpressedas d 13 Cwithrespect toPDB,withanerroroflessthan0.1 % .The d 13 Cofthe bonecollagenvariedbetween 0.4and 2.4 % 6 0.1 % relativetothePDBstandard,whileboneapatitevaried between 7.2 % and 7.5 % 6 0.1 % .Thesevalueswere subsequentlyusedtocalculatecorrectionsforisotope fractionation. Thequoteduncalibrateddatesareinradiocarbonyears before1950(yearsBP),usingthe 14 Chalf-lifeof5568years (Table2).Theerrorisquotedasonestandarddeviation andreflectsbothstatisticalandexperimentalerrors.The dateshavebeencorrectedforisotopicfractionation assumingthatthesamplesoriginallyhada d 13 Ccompositionof 25 % .TheagesshowninTable2werecalibrated usingOxCalversion3.9(Ramsey,1995,2001)andthe calibrationcurveofStuiveretal.(1998). R ESULTS SamplesBandFweredatedusingbothcollagenand bio-apatiteforcomparison.Inbothsamplesthebio-apatite agesareseveralhundredyearsolderthanthecollagenages presumablybecauseoftheincorporationofold,dead carbonduringaccumulationorbecauseoflatercontamination.Becausethecaveisaspring,dischargingground watercontainssignificantquantitiesofoldcarbonthat couldexplainthisobservation.Thecollagenagesare consideredmorereliable.CollagensamplesCandFare statisticallyofthesameage(6447 6 106calyrB.P.and 6698 6 150calyrB.P.)asaresamplesBandD(5124 6 211calyrB.P.and4896 6 194calyrB.P.).Thismeans thatthesamplesrecoveredcouldhavecomefromtwo individuals,onedyingaround6500calyrB.P.andthe otheraround5000calyrB.P.,orfromseveraldifferent sealsthatdiedatthesetimes. D ISCUSSION Basedontheagesofthebones,andassumingthatthe sealscouldnothaveclimbedtoledgesinthecavemuch abovewaterlevel,sealevelwasatmost10mlowerthan presentlevelbyca.6.5ka.Infact,sealevelrecordsforthe Tyrrhenianshowaltitudesbetween6and10mbelow presentatthistime(Antoniolietal.,2004).AtAlghero(NSardinia),Neolithicburialsdatedtoaround7kaB.P.have beenfoundinthefinalsumpofGrottaVerde8–10m belowpresentsealevel(Antoniolietal.,1994). ThelongitudinalprofileoftheBelTorrenteCave (precision 1m)showsthatwhensealevelwas6mlower thantoday,themonksealswouldprobablyhavebeenable toenterthefirst500metersofthecave(Fig.5).Thiswould havegiventhemaccesstoSpiaggiadelBue.Beyondthis, thedeepcentralsumpreaching22mdepthandcompletely submerged6kamayhavebeenasignificantobstacletothe seals.However,theRamodelBuegallery,withaninitial sectionoflimiteddepthandthentwosumpsaround15m deep,mayhavebeenpartlyaccessible.Infact,6ka SpiaggiadelBueandthefirstshallowsectionofRamo delBue,500mand750–800mfromtheentrance, respectively,mayhavebeenspecialrestingplacesformonk sealsandfemalesgivingbirthonthesandybeaches alongsidetheundergroundriver.Supportingthisconclusionareobservationsofsimilarbehaviorbymonkseals Table2.AMSradiocarbonagesonsealbonecollagenandbioapatite. SampleID UGACAIS ID a LibbyAgewith BackgroundSubtracted d 13 C LibbyAgewith d 13 CCorrection CalibratedAgesin cal.yrBC(95.4%) CalibratedAge(cal yrbeforeAD2000) BR01879-B4957 6 50 7.535098 6 503989–37745881 6 107 BR01879-C4308 6 54 11.124421 6 543335–29135124 6 211 CR01880-C5501 6 57 11.195613 6 574553–43416447 6 106 DR01881-C4192 6 53 12.394293 6 533090–27024896 6 194 FR01882-B6798 6 55 7.226942 6 555978–57247851 6 127 FR01882-C5739 6 59 10.445857 6 594848–45486698 6 150 a B bioapatite,C collagen. Table1.Locationanddescriptionofthebonesamples. SampleDistancefromEntrance(m)Depth(m)Description B7707.6Biggerbone(10cm)foundinsandinthepassage C7609.5Smallbone(finger?)foundonrightsideofpassageinsmallsand filledcleft D72012Smallbone(finger?)foundonthesandinmiddleofpassage F7609Smallbonefoundinmiddleofpassageonthesandbetweenrocks J.D E W AELE ,G.A.B ROOK AND A.O ERTEL JournalofCaveandKarstStudies, April2009 N 21

PAGE 7

thatusedBueMarinoCave.AccordingtoJohnson,these sealsshelteredorgavebirthalmost1kmfromtheentrance tothiscave(Johnson,1998). C ONCLUSIONS IthasbeensuggestedthatmonksealsintheMediterraneansoughtoutcavesasrefugesfromseawavesduring heavystorms,humaninterference,andkilling.Ouranalysis ofsealbonesfromBelTorrenteCavesuggestthateven 6.5ka,whenhumanpressureswererelativelylowby modernstandards,monksealswereusingcavesasrefuges. Theelevationofthebonesindicatesthatbythistimesea levelwasalreadywithin10mofthepresentposition.The morphologyofBelTorrenteCaveconfirmsthatinthemid Holoceneitwasacoastalcavewithanundergroundriver, andmonksealswouldhavebeenabletopenetrateabout 800mwithoutencounteringseveredifficultiessuchasdeep sumps.Ourdatarevealthatmonkseals,eveninperiodsof lowhumandisturbance,hadthehabitofusingcoastal caves,penetratingasfaras800minside.Thissuggeststhat 6.5kahumanswerenottheonlypredatorsofmonkseals. A CKNOWLEDGEMENTS Theauthorswouldliketothankthemanycaversand cavediverswhoexploredandsurveyedtheBelTorrente systemanddocumentedthemonksealcemetery,especially Ju ¨rgenBohnert,KarstenGessert,HerbertJantschke, SalvatoreBusche,PeterdeCoster,AndreasKu ¨cha,Enrico Seddone,andLucaSgualdini.Radiocarbondatingwas performedattheCenterforAppliedIsotopeStudies, UniversityofGeorgia.WeadditionallythankJu ¨rgen Bohnert,KarstenGessert,AnkeOertel,andEnrico SeddoneforthephotographsshowninFigures3and4. ThanksalsototheCentroNauticaSubNavarresefor technicalsupportduringexplorationofthecave.Finally twoanonymousreviewersarethankedfortheirvaluable comments. R EFERENCES Aguilar,A.,Cappozzo,L.H.,Gazo,M.,Pastor,T.,Forcada,J.,and Grau,E.,2007,Lactationandmother-pupbehaviourinthe Mediterraneanmonkseal Monachusmonachus :anunusualpattern foraphocid:JournaloftheMarineBiologicalAssociationofthe UnitedKingdom,v.87,p.93–99. Altara,E.,1995,LaFocaMonaca:Sottoterra,v.101,p.43–54. Antonioli,F.,Bard,E.,Potter,E.K.,Silenzi,S.,andImprota,S.,2004, 215-kahistoryofsea-leveloscillationsfrommarineandcontinental layersinArgentarolacavespeleothems(Italy):GlobalandPlanetary Change,v.43,no.1–2,p.57–78. Antonioli,F.,Ferranti,L.,andLoSchiavo,F.,1994,Thesubmerged neolithicburialsofthegrottaVerdeatCapoCaccia(Sardinia,Italy): ImplicationfortheHolocenesea-levelrise:Memoriedescrittivedella CartaGeologicad’Italia,v.52,p.329–336. Antonioli,F.,Silenzi,S.,Vittori,E.,andVillani,C.,1999,Sealevel changesandtectonicmobility:precisemeasurementsinthree coastlinesofItalyconsideredstableduringthelast125ky:Physics andChemistryoftheEarth(A),v.24,no.4,p.337–342. Arisci,A.,DeWaele,J.,andDiGregorio,F.,2000,Naturalandscientific valenceoftheGulfofOroseiCoast(central-eastSardinia)andits carryingcapacitywithparticularregardtothepocket-beaches: PeriodicumBiologorum,v.102,no.suppl.1,p.595–603. Bareham,J.R.,andFurreddu,A.,1975,Observationsontheuseof grottosbyMediterraneanmonkseals( Monachusmonachus ):Journal ofZoology,v.175,p.291–298. Borrell,A.,Aguilar,A.,andPastor,T.,1997,Organochlorinepollutant levelsinMediterraneanmonksealsfromtheWesternMediterranean andtheSaharacoast:MarinePollutionBulletin,v.34,no.7, p.505–510. Borrell,A.,Cantos,G.,Aguilar,A.,Androukaki,E.,andDendrinos,P., 2007,Concentrationsandpatternsoforganochlorinepesticidesand PCBsinMediterraneanmonkseals( Monachusmonachus )from WesternSaharaandGreece:ScienceoftheTotalEnvironment, v.381,p.316–325. DeWaele,J.,2004,Geomorphologicevolutionofacoastalkarst:theGulf ofOrosei(Central-EastSardinia,Italy):ActaCarsologica,v.33, no.2,p.37–54. Figure5.LongitudinalprofileofBelTorrenteCaveshowingaccessibilit ytodayand6kyB.P.whensealevelwasmuchlower. M ONKSEAL ( M ONACHUSMONACHUS ) BONESIN B EL T ORRENTE C AVE ( CENTRAL EAST S ARDINIA ) ANDTHEIRPALEOGEOGRAPHICALSIGNIFICANCE 22 N JournalofCaveandKarstStudies, April2009

PAGE 8

DeWaele,J.,2008,EvaluatingdisturbanceonMediterraneankarstareas: theexampleofSardinia(Italy):EnvironmentalGeology,(inprint). DeWaele,J.,andForti,P.,2003,Estuarisotterranei, in Cicogna,F.,Nike Bianchi,C.,Ferrari,G.,andForti,P.,eds.,GrotteMarine: cinquant’annidiricercainItalia:Rapallo,MinisteroperlaDifesa dell’Ambiente,p.91–104. Dendrinos,P.,Karamandlidis,A.A.,Androukaki,E.,andMcConnell, B.J.,2007a,Divingdevelopmentandbehaviorofarehabilitated Mediterraneanmonkseal( Monachusmonachus ):MarineMammal Science,v.23,no.2,p.387–397. Dendrinos,P.,Tounta,E.,Karamandlidis,A.A.,Legakis,A.,and Kolomatas,S.,2007b,Avideosurveillancesystemformonitoring theendangeredMediterraneanmonkseal( Monachusmonachus ): AquaticMammals,v.33,no.2,p.179–184. Fancello,L.,Fileccia,A.,andMazzoli,M.,2000,LaGrottadelBel Torrente:Speleologia,v.43,p.67–69. Forti,P.,andRossi,G.,1991,Idrogeologiaedevoluzionecarsicadella CoduladiLuna(Sardegna):AttieMemoriedellaCommissione‘‘E. Boegan’’,v.30,p.53–79. Fyler,C.A.,Reeder,T.W.,Berta,A.,Antonelis,G.,Aguilar,A.,and Androukaki,E.,2005,Historicalbiogeographyandphylogenyof monachineseals(Pinnipedia:Phocidae)basedonmitochondrialand nuclearDNAdata:JournalofBiogeography,v.32,p.1267–1279. Gucu,A.C.,Gucu,G.,andOrek,H.,2004,Habitatuseandpreliminary demographicevaluationofthecriticallyendangeredMediterranean monkseal( Monachusmonachus )intheCilicianBasin(Eastern Mediterranean):BiologicalConservation,v.116,p.417–431. Johnson,W.M.,1998,MonksealmythsinSardinia:TheMonachus Guardian,v.1,no.1,p.1–8. Johnson,W.M.,Karamandlidis,A.A.,Dendrinos,P.,Fe `rnandezde Larrinoa,P.,Gazo,M.,Gonzale z,L.M.,Guclusoy,H.,Pires,R., andSchnellmann,M.,2008,MediterraneanMonkSeal:www. monachus-guardian.org. Karamandlidis,A.A.,Pires,R.,CarinaSilva,N.,andCostaNeves,H., 2004,Theavailabilityofrestingandpuppinghabitatforthecritically endangeredMediterraneanmonkseal Monachusmonachus inthe archipelagoofMadeira:Oryx,v.38,no.2,p.180–185. Mahler,A.,1979,VerkarstungderKarbonatgebieteamGolfodiOrosei (Sardinien):GeologischerPalaeontologischerMitteilungenInnsbruck v.7,no.8–9,p.1–49. Morlock,W.,andMahler,A.,1995,LaGrottadelBelTorrente:lapiu ` importanterisorgenzacarsicadelcomplessocalcareodelGolfodi Orosei:SardegnaSpeleologica,v.8,p.35–36. Oertel,A.,andPatzner,R.A.,2007,Thebiologyandecologyofa submarinecave:theGrottadelBelTorrente(Central-EastSardegna, Italy):MarineEcology,v.28,no.suppl.1,p.60–65. Pires,R.,CostaNeves,H.,andKaramandlidis,A.A.,2007,Activity patternsoftheMediterraneanMonkSeal( Monachusmonachus )inthe ArchipelagoofMadeira:AquaticMammals,v.33,no.3,p.327–336. Ramsey,C.B.,1995,Radiocarboncalibrationandanalysisofstratigraphy:theOxCalProgram:Radiocarbon,v.37,no.2,p.425–430. Ramsey,C.B.,2001,DevelopmentoftheradiocarbonprogramOxCal: Radiocarbon,v.43,no.2A,p.355–363. Samaranch,R.,andGonzale `z,L.M.,2000,Changesinmorphologywith ageinMediterraneanmonkseals( Monachusmonachus ):Marine MammalScience,v.16,no.1,p.141–157. Savelli,C.,Beccaluva,L.,Deriu,M.,Macciotta,G.,andMaccioni,L., 1979,K/ArgeochronologyandevolutionoftheTertiary‘‘calcalkalic’’volcanismofSardinia(Italy):JournalofVolcanologyand GeothermalResearch,v.5,no.3–4,p.257–269. Sgualdini,L.,2004,Ilcimiterodellefoche:Anthe `o,reviewoftheGruppo Speleo-ArcheologicoGiovanniSpanoCagliari,v.8,p.20–25. Stuiver,M.,Reimer,P.J.,andBrazuinas,T.F.,1998,High-precision radiocarbonagecalibrationforterrestrialandmarinesamples: Radiocarbon,v.40,no.3,p.1127–1151. Vogel,J.S.,Southon,J.R.,Nelson,D.E.,andBrown,T.A.,1984, Performanceofcatalyticallycondensedcarbonforuseinaccelerator massspectrometry:NuclearInstrumentsandMethodsinPhysics Research,v.B5,p.289–293. J.D E W AELE ,G.A.B ROOK AND A.O ERTEL JournalofCaveandKarstStudies, April2009 N 23



PAGE 1

EDITORIAL JournalofCaveandKarstStudies UseofFSC-CertifiedandRecycledPaper M ALCOLM S.F IELD Beginningwiththisissue,the JournalofCaveand KarstStudies nowincludestheForestStewardship Council(FSC)logoontheinsidecover.Accordingto theFSCwebsite(http://www.fsc.org/about-fsc.html) ‘‘FSCisanindependent,non-governmental,notforprofit organizationthatwasestablishedtopromotetheresponsiblemanagementoftheworld’sforests.Itprovides standardsetting,trademarkassuranceandaccreditation servicesforcompaniesandorganizationsinterestedin responsibleforestry.ProductscarryingtheFSClabelare independentlycertifiedtoassureconsumersthatthey comefromforeststhataremanagedtomeetthesocial, economicandecologicalneedsofpresentandfuture generations.’’ FSCcertificationallowsconsumerstoidentifyproducts thatprovideassuranceofsocialandenvironmental responsibilityonthepartoftheproducer.Thisis accomplishedbyrequiringthatallmaterialsbetracked fromthecertifiedsource.Thepathtakenbytheraw materialsharvestedfromanFSC-certifiedsourceare trackedusingachain-of-custodythroughprocessing, manufacturing,distribution,andprintinguntilitisan endproductandreadyfordistribution. TheFSClogoisapplicableandallowedonlyforFSCCertifiedandRecycledPaperuse.Byhavingselectedan FSC-certifiedpaperforprintingthe Journal ,weare providedtherighttousetheFSClogo.Approveduseof thesealensuresthatthepaperandprocessesthatweuse forthe Journal arebeingproducedincompliancewith strictguidelinesprotectingtheenvironment,wildlife, workersandlocalcommunities. Interestingly,the Journal hasactuallybeenconformingtoFSCstandardsformorethanayear,whichreflects theconservationmindsetofthemembersoftheNational SpeleologicalSociety.Asisknowntomanyofyou, conservationisamajorpartoftheNationalSpeleological Society,althoughwearemorefocusedonthemore limitedconceptofcaveandkarstconservation(http:// www.caves.org/committee/conservation/).However,by usingFSC-certifiedpaperwearedirectlysupporting forestconservationandmaybeindirectlysupportingcave andkarstconservation,dependingonwherethepaperis beingharvested.ThisissomethingthattheNational SpeleologicalSocietycanbeproudofanditreflectswell onourmembers’commitmenttobeingconservation minded. E DITORIAL JournalofCaveandKarstStudies, April2009 N 1



PAGE 1

SEASONALDISTRIBUTIONANDCIRCADIANACTIVITYIN THETROGLOPHILELONG-FOOTEDROBBERFROG, ELEUTHERODACTYLUSLONGIPES (ANURA: BRACHYCEPHALIDAE)ATLOSRISCOSCAVE, QUERE TARO,MEXICO:FIELDAND LABORATORYSTUDIES A DRIANA E SPINODEL C ASTILLO 1 ,G ABRIELA C ASTAN O -M ENESES 2,4 ,M AYRA J.D A VILA -M ONTES 1,3 M ANUEL M IRANDA -A NAYA 3 ,J UAN B.M ORALES -M ALACARA 1,4,6 AND R ICARDO P AREDES -L EO N 5 Abstract: LosRiscosCavebelongstotheElAbralimestoneanditsgeographical locationisintheSierraGordaintheStateofQuere taro,Mexico.Thecavehasahigh faunaldiversitythatincludesarthropodsandsomevertebrates,suchasv ampirebatsand anurans,andincludestherobberfrog Eleutherodactyluslongipes (Baird,1859).The abundanceoftherobberfrogchangesnon-randomlybetweendryandrainyse asonsand isrelatedtothesearchforhumidconditionsinsidethecave.Inaddition, therobberfrog waslocatedinareaswheresomescatteredlightmayinfluenceitsdispersi oninsidethe cave;andtherefore,itsactivity.Frogsdisplayedspontaneouscircadia nrhythmsof locomotoractivityfromthefirstdaysoftheexperimentalobservationin constant darkness.Theaverageperiodofcircadianrhythmswas24.85 6 0.93hindicating,in isolatedconditions,adiurnalactivity.Whenexposedtoartificialligh t-darkcycles,the animalslackeddailyactivityrhythms,andultradianactivitywasobserv ed.The preferenceforhighhumidityandlowilluminationinthecaveandapartial endogenous circadianrhythmicityconfirmthetroglophilicaffinityoftherobberfr ogtocave environments. I NTRODUCTION Cavesrepresentwindowstothelithospherewhere differenthabitatsarecharacterizedbypartialortotal darkness,nearlyconstanttemperature,oftenhighlevelsof humidity,andalowflowofnutrients.Nevertheless,a varietyoforganismshavelongcolonizedthesesubterraneanenvironments.Therearestillconstantincursionsto cavesbytrogloxenesandtroglophiles.Inaddition,there areuniquetroglobiticspeciesthathavedevelopeddiverse andspecializedadaptationsduringtheirevolutiontothe lackoflight. InMexico,therearerecordsof27speciesofanurans thatinhabitcaves(Reddell,1981;Hoffmannetal.,1986), whichcorrespondstoapproximately7%ofthetotal amphibiandiversityofthecountry(363).Thishigh percentageofuseofthishabitatforanuransismore frequentthanexpected(Lo pez-OrtegaandCasas-Andreu, 2005). LosRiscosCavehasbeeninvestigatedforitsfaunaover manyyearsandhasalargediversityoffaunathatincludes arthropodsandvertebrates,suchasvampirebats Desmodusrotundus (Geoffroy,1810)and Diphyllaecaudata Spix, 1823,thefrugivorousbat Artibeuslituratus (Olfers,1818), aninsectivorousbat Corynorhinusmexicanus (Allen,1916), andthefish Astyanaxmexicanus (DeFilipi,1853),and anuransas Inciliusvalliceps (Wiegmann,1833)(previously knownas Bufovalliceps ,seeFrost,2008),andtherobber frog Eleutherodactyluslongipes (Baird,1859)(Fig.1), whichisamemberofthefamilyBrachycephalidae, followingFrostetal.,(2006). Eleutherodactyluslongipes isendemictoMexico,being distributedalongtheSierraMadreOrientalfromthestates ofNuevoLeonthroughHidalgo,Tamaulipas,SanLuis Potos andQuere taroinisolatedlocalities(Lynch,1970; IUCN,2006).Itinhabitsmoderateelevationsfrom650to 2000m,anditshowsastrongtendencytooccupycaves (Taylor,1939;Lynch,1970;IUCN,2006).Thebiologyand naturalhistoryofthisfrogareunknown. 6 CorrespondingAuthor(email:jbmm@hp.fciencias.unam.mx). 1 Acarolog a&Bioespeleolog a,DepartamentodeBiolog aComparada,Facultadde Ciencias,UniversidadNacionalAuto nomadeMe xico,Coyoaca n04510,Distrito Federal,Me xico 2 Ecolog aySistema ticadeMicroartro podos,DepartamentodeEcolog ayRecursos Naturales,FacultaddeCiencias,FacultaddeCiencias,UniversidadNaci onal Auto nomadeMe xico,Coyoaca n04510,DistritoFederal,Me xico 3 Biolog aanimalexperimental,DepartamentodeBiolog aCelular,Facultadde Ciencias,UniversidadNacionalAuto nomadeMe xico,Coyoaca n04510,Distrito Federal,Me xico 4 AdditionalAddress:UnidadMultidisciplinar adeDocenciaeInvestigacio n, FacultaddeCiencias,UniversidadNacionalAuto nomadeMe xico,Campus Juriquilla,BoulevardJuriquilla3001,C.P.76230,Quere taro,Quere taro,Me xico 5 DepartamentodeZoolog a,InstitutodeBiolog a,UniversidadNacionalAuto nomadeMe xico,Coyoaca n04510,DistritoFederal,Me xico A.EspinodelCastillo,G.Castan o-Meneses,M.J.Da vila-Montes,M.Miranda-Anaya,J.B.Morales-Malacara,andR.ParedesLeo n– Seasonaldistributionandcircadianactivityinthetroglophilelong-fo otedrobberfrog, Eleutherodactyluslongipes (Anura:Brachycephalidae)at LosRiscosCave,Quere taro,Mexico:Fieldandlaboratorystudies. JournalofCaveandKarstStudies, v.71,no.1,p.24–31. 24 N JournalofCaveandKarstStudies, April2009

PAGE 2

Cavesprovideadiversityofhabitatswhereenvironmentalsignalsmayhavedifferentinfluencesuponthe circadianpropertiesoftheorganismsthatinhabitthem. Nearthemainentryofacave,naturalday-nightcycles,as wellaschangesintemperatureandhumidity,mayplaya significantroleinthedailyorganizationofanimalactivity. Indeepregionsofcaves,constantorpoorlyfluctuating conditionslackthetemporalperiodicitytoentrain endogenouslygeneratedcircadianrhythms,andpoorlight conditionscanalsoaffectotherphysiologicaltraitssuchas visionorskincolor(PoulsonandWhite,1969;Lamprecht andWeber,1992). Diversestudiesofcircadianactivityrhythmshavebeen performedunderlaboratoryconditionsbymonitoringthe locomotoractivityoffish(Trajanoetal.,2005;Pati,2001), crustaceans(DelaO-Mart nezetal.,2004),crickets (Hoenen2005),millipedes(Koilrajetal.,2000),spiders (Gnaspinietal.,2003)andsalamanders(Hervantetal., 2000),amongothers.Resultshavevaried:somespeciesare arrhythmicandothersremaincapableofdisplayinglowamplitudecircadianrhythms,abletosynchronizeto artificiallight-darkcycles. Studyingdifferentmorphological,ontogeneticorethologicaladaptationstoundergroundenvironmentsmayhelp usunderstandtheadaptivemeaningofhavingafunctional circadianclockinorganismsthatliveinsuchunusual habitats. Ouraimistopresentthehabitatandseasonal distributionof E.longipes insideLosRiscosCave,aswell asitslocomotoractivitypatterninlaboratoryconditions, forunderstandinglong-termcircadianlocomotoractivity innaturalsettingsandtheirresponsetoartificiallight-dark cycles. M ATERIALSAND M ETHODS L OCALITY LosRiscosCave(21 u 11 38 N,99 u 30 50 W:Lazcano Sahagu n,1986b)belongstotheElAbralimestoneandits geographicallocationisintheSierraGordaintheStateof Quere taro,Mexico(Alencasteretal.,1999).Itislocated 3kmnortheastfromPuentedeDiosuptheriverJalpanat 1122masl.Waterpersistenceinsidethecavevaries betweendryandrainyperiods.Maximumprecipitation occursfromJunetoNovember,andthedriestperiodis fromDecembertoMay.Averageannualprecipitationis about48.9mmd 1 (CETENAL,1986;CNA,2004),and theaverageannualtemperatureis24 u C(INEGI,1996; CGSNEGI,2004).Theclimateoftheareaiswarm subhumidtype(A)C 1 (wo)(w)sensuKo ¨ppenandmodified byGarcia(1981)andthecategoryofvegetationistropical dryforest(CarabiasLilloetal.,1999). Thiscaverepresentsamixedundergroundsystem (horizontalandvertical)witheightzones,denotedwith thelettersAtoH,includingtunnelsandgalleries(Fig.2). Thecaveismainlyhorizontal,withadepthof35manda lengthof550m(LazcanoSahagu n,1986a,1986b).Criteria fordefiningeachzonearebasedondifferenttopographical eventsandfeaturesinsidethecave,liketunnels,areasof collapseorchangingclimaticconditions.Themainentry, labeledaszoneA,issurroundedbylocalvegetationand sunlightscattersabout35minside.Otherzoneshave diminishedamountoflightoraretotallydark.ZonesB,C, DandErepresentthelargesttransects,thereforeaverage valuesofhumidity,temperature,andlightindicatethe proximal(about40mdeep,designatedas1)anddistal (about80mdeep,designatedas2)locations.ZoneB2 representstheupperchamberattheendofthiszone,which is30mhigh. ZonesGandHincludenarrowcorridorsandvertical pitsandthuspresentedaccesschallenges.Theseareaswere notconsideredforthisstudybecausenofrogswerefound there. F IELD T ECHNIQUES Allsamplingactivitieswereperformedseasonallyevery threemonthsfromNovember2005toMarch2008. Physicalparameterssuchastemperature,humidity,and luminositywererecordedindifferentlocationsinsidethe caveusingathermohydrometer(IAQ-Calc8760),anda luxometer(EXTECHinstruments0–2000luxes).Readings wereobtainedevery20malongsearchtransectsinboth dryandrainyseasons. Thesearchforfrogswascarriedoutalongeachoneof thecavezonesduringday-twilightperiodsbecausewe observedinpreliminaryexplorationsthatapparentlythe frog’sactivityinsidethecaveisthesamealltimesofthe day.Thefrogsarenotveryvisible.Forexample,inzoneB theyweremorecommonlyobservedatrestontopofwalls orcrevicesonwalls,whereasinotherzones(E,F),the Figure1.Long-footedrobberfrog Eleutherodactyluslongipes atLosRiscosCave,Quere taro,Mexico.Scalebaris 2cm. A.E SPINODEL C ASTILLO ,G.C ASTAN O -M ENESES ,M.J.D A VILA -M ONTES ,M.M IRANDA -A NAYA ,J.B.M ORALES -M ALACARA AND R.P AREDES -L EO N JournalofCaveandKarstStudies, April2009 N 25

PAGE 3

frogswerefoundunderrocks.Zoneswherethefrogswere observeddidnotincludepuddlesorpoolsofwater.Some frogsweremarkedandreleased;otherswerecollectedand transportedtothelaboratoryforcircadianstudiesorto serveasvoucherspecimensattheFacultaddeCiencias, UniversidadNacionalAuto nomadeMe xico(FC).Inorder tominimizepossibledampeningofendogenousactivity rhythms,allanimalsusedinthelaboratoryweretransportedinadarkstyrofoamboxslightlycooledwith crushediceandwererecordedinthelaboratoryassoonas possible.Afewindividualswereexaminedforparasites andtoconfirmproperidentification.Thesespecimenswill bedepositedattheUniversidadNacionalAuto nomade Me xico(UNAM),intheMuseodeZoolog adelaFacultad deCiencias(MZFC-UNAM)andColeccio nNacionalde AnfibiosyReptiles(CNAR-IBUNAM). L ABORATORY T ECHNIQUES Ninefrogscollectedduringdifferentseasons(fiveindry seasonandfourinrainyseason)wereusedforthisportion ofthestudy.Frogswereindividuallykeptinglassaquaria (2.5L)andwereeachequippedwithinfraredlight crossingstodetectlocomotoractivityasdescribedelsewhere(Miranda-Anayaetal.,2003).Thesewerekeptin continuouslyventilated,light-proofwoodenboxes,and maintainedat23 6 2 u Cinenvironmentally-controlled roomsattheFCfacilities.Inordertomaintaina sufficientlyhighhumidityinsidetherecordingchamber, eachaquariumhadawetpapertowelspreadonitsbottom whichwasmoistenedwithcleanwateronceaweek. Locomotoractivitydataweresummarizedevery10min forfurtheranalysisbymeansofadataacquisitionboard (NAFRIdisp.Me xicoD.F.). Everyfrogwasinitiallymonitoredforatleastninedays undercontinuousdarkness(DD).Asneeded,adimred lightbulb(1–2lx)wasusedforvisualinspectionand maintenanceofaquaria.Then,light-darkcycles(LD12:12) wereusedduringatleasttendaysinordertoobserveany possibleentrainmentoflocomotoractivityrhythms.Light cycleswereprovidedbymeansofafluorescentlamp (TecnoliteF15T8D)controlledbyatimer.Thelampwas partiallydimmedbyusingblacktapeandlocatedat30cm aboveaquariainordertoprovidehalfofthelightintensity (150lx).Frogswerefedonceaweekwithjuvenilecrickets thatwererenderedleglesstoavoidtheirinterferingwith froglocomotoractivities. D ATA A NALYSIS Seasonalmeansinabundancewerecomparedbya student t -testforindependentsamplesperformedby Statisticasoftware(StatSoft,1999).AFisher’sexacttest (Tocher,1950)wasusedtoevaluatetheassociation betweenseasonsandbetweendifferentcavezones.We performedtheFreemanandHalton(1951)analysisofthe Fisherexactprobabilityforatwo-by-fivecontingency table.AtwowayAnalysisofVariance(ANOVA)was Figure2.Distributionandrelativeabundanceof E.longipes inLosRiscosCave,Quere taro,Mexico.Arrowsrepresentcave entrypoints.Themap,modifiedfromLazcanoSahagu n(1986),showsindetailthezonesofthecave,wherefrogswere sampled.Frog-likesymbolsindicatepercentageoffrequencyofobservat ionineachzone. S EASONALDISTRIBUTIONANDCIRCADIANACTIVITYINTHETROGLOPHILELONG FOOTEDROBBERFROG ELEUTHERODACTYLUSLONGIPES (A NURA : B RACHYCEPHALIDAE ) AT L OS R ISCOS C AVE ,Q UERE TARO ,M EXICO :F IELDANDLABORATORYSTUDIES 26 N JournalofCaveandKarstStudies, April2009

PAGE 4

performedinordertoevaluatetheeffectoftheseasonand cavezoneontheenvironmentalparameterswitha posthoc Tukey’stestusedtoidentifysignificantdifferences.A multiplecorrelationwasperformedbetweentemperature, humidity,luminosity,andfrogabundanceinsiteswhere frogswererecorded. Locomotoractivitydatawereanalyzedindoubleplottedactogramsbasedonaday-by-dayhistogramin whicheachbarindicatestheamountofactivityobservedin atimelapseof10minacrossa24-hourseries.Inthis presentation,thedailytracesstackedverticallyinchronologicalordergivealucidviewoftheactivityenvelope (DeCoursey,2004). Therespectivecircadianperiodsforatleastseven consecutivedays( t )werecalculatedusing x 2 periodograms (SokoloveandBushell,1978)at0–30hintervals.Ultradian activitywasanalyzedbymeansofFastFourierTransform AnalysisusingthesoftwareDISPAC(InstitutodeFisiolog aCe lular(IFC),UNAM,Mexico;Aguilar-Robleroet al.,1997).Periodogramswithspikesabovetheconfidence interval( p 0.05)wereconsideredrhythmic.Resultson periodlengthsoffree-runningactivityboutsunder differentprotocolswereanalyzedusinganon-paired student t -testwiththesoftwareprogramStatistica; differenceswereconsideredsignificantwhen p 0.05.All resultsarepresentedasmeanvalues 6 standarddeviations (SD)unlessotherwisenoted. R ESULTS P HYSICAL P ARAMETERSIN L OS R ISCOS C AVE Thetwo-wayANOVAtestshowednosignificanteffect ofseasonontemperature(F 1,60 0.80; p 0.05), humidity(F 1,60 3.18 p 0.05),orluminosity(F 1,60 0.47 p 0.05).However,thereweresignificanteffectsof cavezoneontheseparameters(temperature:F 9,60 2.35 p 0.05;humidity:F 9,60 5.81 p 0.005;luminosity:F 9,60 37.72 p 0.0005).Temperaturedifferenceswerefound betweenzonesBandD2andbetweenzonesD1andD2. Themaindifferencesforthehumiditywerefoundbetween zoneAandzoneD2.Fortheluminosity,zonesBandC1 togetherweredifferentfromtheotherzonesandzoneA wasdifferentfromallothers(Fig.3). Thezonewiththehighesttemperatureandhumidityin bothseasonswasD2.Thiszonehaspermanentcoloniesof vampirebats( Desmodusrotundus and Diphyllaecaudata ) presentinthedeepestpartofthecave.Twootherbat species, Artibeuslituratus and Corynorhinusmexicanus werefoundinthiszone,butonlyduringasinglesampling event. D ISTRIBUTIONOF E LEUTHERODACTYLUSLONGIPES INTHE C AVE D URING D IFFERENT S EASONS Atotalof43frogswasfoundinthecaveduringthe three-yearstudy.Seasonalaverageabundancefortherainy seasonwas5 6 1.41SD,andforthedryseasonwas11 6 4.24SD;thedifferencebetweenthetwoseasonswas significant( t 4 2.82; p 0.04). Frogswerelocatedmainlybetween1200hand1600h incavezonesA,B,E,andF.DensitywashighestinzoneB andlowestinzoneFforbothseasons(Fig.4).InzoneA, onlyoneindividualwasfoundduringthedryseason,while duringtherainyseason,frogswerefoundoutsideofthe caveontheexternalwallofzoneBatmidnightjustone time.Therewasnoassociationbetweencavedepthand season(Fisher’sexacttest, p 0.29).Forboththedryand rainyseasons,thedistributionoffrogsinthecavewas similar.Thecoefficientofmultiplecorrelationbetweenthe environmentalparametersandfrogabundancewashighly positiveandsignificant( r 0.94; p 0.05).Humidity explainedmostoftheobservedvariationinabundance( r 0.72; p 0.005),followedbyluminosity( r 0.35; p Figure3.Averagetemperature(A),humidity(B),and luminosity(C)alongLosRiscosCave.Lettersdenote significantdifferencesbetweenzonesaccordingto posthoc comparisonsusingTukey’stest( p 0.05).Lightgraybars meandryseasonanddarkgraybarsmeanrainyseason. A.E SPINODEL C ASTILLO ,G.C ASTAN O -M ENESES ,M.J.D A VILA -M ONTES ,M.M IRANDA -A NAYA ,J.B.M ORALES -M ALACARA AND R.P AREDES -L EO N JournalofCaveandKarstStudies, April2009 N 27

PAGE 5

0.05).Temperaturewasnotsignificantlycorrelatedwith abundance( r 0.24; p 0.05).Thisisunsurprising becausethisparameterwasconstantthroughoutthecave (Fig.5).Thesimplecorrelationbetweenluminosityand abundancewasnegativeandsignificant( r 0.11; p 0.05). C IRCADIAN L OCOMOTOR A CTIVITY R HYTHMS Figure6showsadouble-plottedactogram(A)and correspondentactivityprofiles(B)foratypicalrecording oflocomotoractivityunderdifferentlightconditionsfora singlefrog.Asignificantcircadianrhythmoscillatedduring thefirstninedaysofrecordinginDD.Themainboutof activityisprojectedfortheexternal-lightphase.When exposedtoLDcycles(day10),activitywasstillmaintained duringthediurnalphaseofthecycleandcrepuscular activitywasobservedinthisfirstLDcondition,asshown inthecorrespondentactivityprofile(seeBinFig.6). However,whenexposedagaintoconstantDD,therhythm fadedawayandanon-significantcircadianperiodwas observed.ThenextexposuretoLDcyclesdidnotshow circadianactivity. Inasecondgroupofanimalstestedforalongerperiod, theinitialfree-runningcircadianrhythminDDwas observedagain.However,therhythmwasdampenedafter 10to20daysandLDconditionsdidnotreorganizethe activity.Frequencyanalysisofultradianrhythmsbymeans ofFastFouriertransformshowsthatinLD,circadian frequencieswereabsentwhileultradianrhythmsof1to6h wereobserved. D ISCUSSION ThefactorsdeterminingtheenvironmentalcharacteristicsofLosRiscosCaveareassociatedwiththeexternal agentsimposedbylocalclimateandseason,andwefound thatdifferencesoffrogabundancebetweendryandrainy seasonscanberelatedtohumiditylevelsinsidethecave, whichisconsistentwiththenecessityoffrogstoavoid desiccationandlightexposure.Also,amphibiansof temperateareasdisplaydifferentstrategies,suchas hibernation,inordertoavoidstarvationandlow environmentaltemperatures(Pinderetal.,1992).Among thesestrategies,thereisuseofrefugiasuchascracksin rocks,fallentrunks,stumps,cavesandothers(Lo pezOrtegaandCasas-Andreu,2005).Thestudyresults confirmedthatenvironmentalvariablessuchastemperature,solarradiation,andrelativehumiditydeterminethe distributionoftheseorganisms. Itisimportanttopointoutthatalltheindividualswere foundduringthedaytimeinsidethecavebecauseLynch andDuellman(1997)affirmedthatallthe EleutherodactyFigure4.Seasonaldistributionof E.longipes indifferent zonesatLosRiscosCave,Quere taro,Mexico.Lightgray barsmeanrainyseasonanddarkgraybarsmeandryseason. Figure5.Simplecorrelationanalysisbetweentemperature (A),humidity(B),andluminosity(C)withabundanceof E. longipes inLosRiscosCave.DatafromNovember2005to March2008atLosRiscosCave,Quere taro,Mexico.Solid linesmeanthecorrelationthatisadjustedtothedata,and thedashedlinesmeantherangeofdispersionofthedatawith regardtothecorrelationline. S EASONALDISTRIBUTIONANDCIRCADIANACTIVITYINTHETROGLOPHILELONG FOOTEDROBBERFROG ELEUTHERODACTYLUSLONGIPES (A NURA : B RACHYCEPHALIDAE ) AT L OS R ISCOS C AVE ,Q UERE TARO ,M EXICO :F IELDANDLABORATORYSTUDIES 28 N JournalofCaveandKarstStudies, April2009

PAGE 6

lus (withtheexceptionof E.hectus )arenocturnal.The biotic(potentialprey)andabiotic(temperature,humidity, anddarkness)parametersinsidethecaveprobablyincrease therangeofcrepuscularactivityoftherobberfrogas showninFigure6B. Insidethecave,zoneBshowsthemainpopulationof frogsandotherrecordscanbeconsideredasaccidental. Thepopulationlivinghereapparentlydoesnothave contactwiththeexteriorofthecave.However,wecannot discardthepresenceoflittlecracksthatallowsome movementstooutside,whichmeansthefrogscouldnot remaininthecavefulltime.Additionally,thefrogswere observedoutsidethecaveatnightandfarfromthemain populationofzoneB. Preliminaryobservationsindicatethatthereismore diversityofinvertebratesinzonesAandFthaninzones BandE.However,cricketsandspidersinzoneBcanalso serveaspotentialpreyforthefrogs.Nofrogswere observedfeedingoncaveanimalswhileinthecaveand, unfortunately,thecontentsofthefewstomachsanalyzed wereinastateofadvanceddigestion,soitwasnot possibletoidentifythetaxainstomachcontents.ZonesC andDhadappropriatetemperature,humidity,and luminosityconditions,buttheirisolationfromthe externalenvironmentmadetheseconditionsdifficultfor thefrogstoaccess. Endogenouscircadianrhythmicitywasdetectedduring theinitialtendaysofrecordingunderlaboratory conditionswithoutacclimation,whichindicatesafunctionalcircadianclockinthisspecies.Theaveragefreerunningperiodobservedduringthefirst8–10daysof locomotoractivityindicatesawidecircadianvariation, whichaccordingtoAschoff’srule,correspondstoadiurnal species(Aschoff,1960). Thisfrogisvisuallyreactivetothepresenceofliving insectsasprey.However,becausethelightintensityof artificialLDcyclesusedinthelaboratorywashigherthan thatavailableinsidethecave,itispossiblethatbrightlight inducedarhythmicpatternsofultradianexpressioninthis speciesasisknowntohappeninvertebratesexposedto continuousbrightlight(Ohta,etal.,2005). C ONCLUSIONS Robberfrogdistributionamongcaveregions,being largelyconcentratedinzoneB,suggeststhatthepresence ofsomescatteredlightisnecessaryforitsactivityandmay influenceitsdispersioninsidethecave.Thefunctional circadianclocksincaveanimalsmayhavedifferent sensitivitiestolightaccordingtothespecies,becausefish orarachnidscollectedindeeperzoneswereabletoentrain theircircadianrhythmsonsimilarartificiallight-dark cycles(unpublishedobservations).Laboratoryexperiments didnotincludeinteractionwithotherspeciesthatcoexistin thesamecaveareas.Therefore,itispossiblethattiming relationshipsamongspeciesinsidethecavemightbe importanttosustainingcircadianrhythmicityunder naturalconditionsasobservedfordifferentcaveinsects (Odaetal.,2000).Non-circadiandataobservedindifferent cavespeciesintheseconditionsmayrequireadifferent statisticalanalysisthatcanfilterthenoisepatternofsmall peaksasseenfortherhythmofthecavecricket Strinatia brevipennis (Hoenenetal,2001).Thearhythmicpattern observedmayrequiredifferentcriteriaofanalysisthan usedinthepresentstudy,andtheecophysiological significanceofthesefindingsisyettobefullyunderstood. Nonetheless,thisisthefirststudyofsomeaspectsofthe biologyof E.longipes ,andwecannotethatthe environmentalpreferenceforcavesandapartialendogenouscircadianrhythmicityconfirmsthetroglophilicnature oftherobberfrog.Thepaucityofinformationconcerning theecologyandphysiologyoftroglophilicanurans highlightstheimportanceofgeneratingdetailedstudies ofthesetopics. A CKNOWLEDGMENTS Forcommentsonanearlierdraftofthemanuscript,we expressourappreciationtoJoaqu nArroyo-Cabrales (LaboratoriodeArqueozoolog a,InstitutoNacionalde Antropolog aeHistoria,Me xico),FredKraus(Bishop Museum,Hawaii),GerardoLo pez-Ortega(Universidad Figure6.Double-plottedactogram(left)andcorresponding activityprofilesofasinglefrog.Lightconditionsare indicatedbyarrows(DD = constantdarkness,LD = light/ darkcycles),lightphaseisalsoindicatedbyrectangleson leftsideoftheactogram.ActivityprofilesonBshowalsothe LDcycles.Daysconsideredfortheprofilesareindicated. A.E SPINODEL C ASTILLO ,G.C ASTAN O -M ENESES ,M.J.D A VILA -M ONTES ,M.M IRANDA -A NAYA ,J.B.M ORALES -M ALACARA AND R.P AREDES -L EO N JournalofCaveandKarstStudies, April2009 N 29

PAGE 7

Auto nomaMetropolitana–Iztapalapa).Wealsothank MiguelHerna ndez,SaraSoriano,YunuenA vila,Marisol Vega,ItzelSigalaandStephanieOrtega(Biospeleology group,FacultaddeCiencias,UniversidadNacional Auto nomadeMe xico)fortheirhelpfulassitanceinthe fieldexpeditions.WethankAlejandroMart nezMenaand AnaIsabelBielerAntolin,andJose AntonioHerna ndez (LaboratoriodeMicrocine,FacultaddeCiencias,UniversidadNacionalAuto nomadeMe xico)fortheirhelp withphotographs.Financialassistancewasprovidedby theDireccio nGeneraldeAsuntosdelPersonalAcade mico, UniversidadNacionalAuto nomadeMe xico,Grant IN221906toJ.B.M-M. R EFERENCES Aguilar-Roblero,R.,Salazar-Juarez,A.,Rojas-Castan eda,J.,Escobar, C.,andCintra,L.,1997,Organizationofcircadianrhythmicityand suprachiasmaticnucleiinmalnourishedrats:AmericanJournalof Physiology,RegulatoryIntegrativeComparativePhysiology,v.273, p.R1321–R1331. Alencaster,G.,Torres-Herna ndez,R.,Trista n-Gonza lez,M.,BarbosaGudin o,R.,Lo pez-Doncel,R.,Pons,J.M.,andOman a,L.,1999,El AbraformationinthewesternpartoftheValles-SanLuisPotos Platform(Mexico):ErlangerGeologischeAbhandlungen,Sonderband,v.3,p.7–8. Aschoff,J.,1960,Exogenousandendogenouscomponentsincircadian rhythms, in Proceedings,ColdSpringHarborSymposiaonQuantitativeBiology,v.25,p.11–28. Baird,S.F.,1859,ReptilesoftheBoundary, in UnitedStatesandMexican BoundarySurveyundertheOrderofLieut.Col.W.H.Emory,Major FirstCavalry,andUnitedStatesCommisioner.2,Rept.,Pt.2. DepartmentoftheInterior,Washington,35p. CarabiasLillo,J.,Provencio,E.,delaMazaElvira,J.,andRuizCorzo, M.,1999,ProgramadeManejoReservadelaBiosferaSierraGorda, Me xico,InstitutoNacionaldeEcolog a,SEMARNAT(Secretar ade MedioAmbienteyRecursosNaturales),172p. CETENAL(Comisio ndeEstudiosdelTerritorioNacional),1986, Precipitacio nyprobabilidaddelalluviaenlaRepu blicaMexicana ysuevaluacio n,Quere taroeHidalgo,Serieclimas,Institutode Geograf a,UniversidadNacionalAuto nomadeMe xico. CGSNEGI(Coordinacio nGeneraldelosserviciosNacionalesde Estad stica,Geograf aeInforma tica),2004,CartadeClimas, 1:1000000, in AnuarioEstad sticodeQuere tarodeArteaga.http:// www.queretaro.gob.mx/sedesu/deseco/esteco/perfeco/anuest/PDF/ Capitulos/C1.pdf CNA(Comisio nNacionaldelAgua),2004,Registromensualde precipitacio npluvialenmm.Ine dito, in AnuarioEstad sticode Quere tarodeArteaga.http://www.queretaro.gob.mx/sedesu/deseco/ esteco/perfeco/anuest/PDF/Capitulos/C1.pdf DeLaO-Mart nez,A.,Verde,M.A.,Valadez,R.L.,Viccon-Pale,J.A., andFuentes-Pardo,B.,2004,Abouttheexistenceofcircadianactivity incavecrayfish:BiologicalRhythmResearch,v.35,no.3, p.195–204. DeCoursey,J.P.,2004,Overviewofbiologicaltimingfromunicellsto humans, in Dunlap,J.C.,Loros,J.J.,andDeCoursey,J.P.,eds., Chronobiology,1 st ed.:Sunderland,Massachusetts,SinauerAssociatesIncorporatedPublishers,p.3–27. Freeman,G.H.,andHalton,J.H.,1951,Noteonexacttreatmentof contingency,goodnessoffitandotherproblemsofsignificance: Biometrika,v.38,p.141–149. Frost,D.R.,2008,AmphibianSpeciesoftheWorld:AnOnlineReference, Version5.2(19November,2008),NewYork,AmericanMuseumof NaturalHistory,ElectronicDatabaseaccessibleathttp://research. amnh.org/herpetology/amphibia/index.php. Frost,D.R.,Grant,T.,Faivovich,J.,Bain,R.H.,Haas,A.,Haddad, C.F.B.,Channing,A.,Wilkinson,M.,Donnellan,S.C.,Raxworthy, C.J.,Campbell,J.A.,Blotto,B.L.,Moler,P.,Drewes,R.C., Nussbaum,R.A.,Lynch,J.D.,Green,D.M.,andWheeler,W.C., 2006,Theamphibiantreeoflife:NewYork,BulletinoftheAmerican MuseumofNaturalHistory,no.297,370p. Garc a,E.,1981,Modificacionesalsistemadeclasificacio nclima ticade ko ¨ppen:Me xico,D.F.,252p. Gnaspini,P.,Santos,F.H.,andHoenen,S.,2003,Theoccurenceof differentphaseanglesbetweencontrastingseasonsintheactivity patternsofthecaveharvestman Gnosiomaspelaeum (Arachnida, Opiliones):BiologicalRhythmResearch,v.34,no.1,p.31–49. Hervant,F.,Mathieu,J.,andDurand,J.P.,2000,Metabolism andcircadianrhythmsoftheEuropeanblindcavesalamander Proteusanguinus andafacultativecavedweller,thePyreneannewt ( Euproctusasper ):CanadianJournalofZoology,v.78,no.8, p.1427–1432. Hoenen,S.,2005,CircadianpatternsintheactivityoftheBraziliancave cricket Strinatiabrevipennis (EnsiferaPhalangopsidae):European JournalofEntomology,v.102,no.4,p.663–668. Hoenen,S.,Schimmel,M.,andMarques,M.D.,2001,Rescuingrhythms fromnoise:Anewmethodofanalysis:BiologicalRhythmResearch, v.32,no.2,p.271–284. Hoffmann,A.,Palacios-Vargas,J.G.,andMorales-Malacara,J.B.,1986, Manualdebioespeleolog a(connuevasaportacionesdeMorelosy Guerrero,Me xico),UniversidadNacionalAuto nomadeMe xico, 274p. INEGI(InstitutoNacionaldeEstad stica,Geograf aeInforma tica),1986, S ntesisGeogra fica,nomenclatoryanexocartogra ficodelestadode quere taro,143p. IUCN(InternationalUnionforConservationofNature),Conservation International,andNatureServe,2006,Globalamphibianassessment. http://www.globalamphibians.org Koilraj,A.J.,Sharma,V.K.,Marimuthu,G.,andChandrashekaran, M.K.,2000,Presenceofcircadianrhythmsinthelocomotoractivity ofacave-dwellingmillipede Glyphiuluscavernicolussulu (Cambalidae, Spirostreptida):ChronobiologyInternational,v.17,no.6,p.757– 765. Lamprecht,G.,andWeber,F.,1992,Spontaneouslocomotionbehaviour incavernicolousanimals:theregressionoftheendogenouscircadian system, in Camacho,A.I.,ed.,Thenaturalhistoryofbiospeleology: Madrid,MonografiasdelMuseoNacionaldeCienciasNaturales, p.225–262. LazcanoSahagu n,C.,1986a,CavernasdelaSierraGorda.v.I, UniversidadAuto nomadeQuere taro,Mexico,181p. LazcanoSahagu n,C.,1986b,CavernasdelaSierraGorda.v.II, UniversidadAuto nomadeQuere taro,Mexico,206p. Lo pez-Ortega,G.,andCasas-Andreu,G.,2005,Atunnelashibernaculum of Hylaplicata (Anura:Hylidae)atSierraNortedeTlaxco,Tlaxcala, Me xico:RevistadelaSociedadMexicanadeHistoriaNatural,3 a e poca,v.2,no.1,p.160–167. Lynch,J.,1970,ATaxonomicRevisionoftheLeptodactylidFrogGenus Eleutherodactylus Cope:UniversityofKansas,MuseumofNatural History,v.20,no.1,p.1–45. Lynch,J.,andDuellman,W.,1997,FrogsoftheGenus Eleutherodactylus (Leptodactylidae)inWesternEcuador:Systematics,Ecology,and Biogeography:UniversityofKansasNaturalHistoryMuseum, SpecialPublication,no.23,p.50–165. Miranda-Anaya,M.,Barrera-Mera,B.,andRam rez-Lomel ,E.,2003, CircadianLocomotorActivityRhythmintheFreshwaterCrab Pseudothelphusaamericana (DeSaussure,1857):EffectofEyestalk Ablation:BiologicalRhythmResearch,v.34,no.2,p.167–176. Oda,G.A.,Caldas,I.L.,Piqueira,J.R.C.,Waterhouse,J.M.,and Marques,M.D.,2000,Coupledbiologicaloscillatorsincaveinsects: JournalofTheoreticalBiology,v.206,no.4,p.515–524. Ohta,H.,Yamazaki,S.,andMcMahon,D.,2005,Constantlight desynchronizesmammalianclockneurons:NatureNeurosciences, v.8,no.3,p.267–269. Pati,A.K.,2001,Temporalorganizationinlocomotoractivityofthe hypogeanloach, Nemacheilusevezardi ,anditsepigeanancestor: EnvironmentalBiologyofFishes,v.62,no.1–3,p.119–129. Pinder,A.W.,Storey,K.B.,andUltsch,G.R.,1992,Estivationand hibernation, in Feder,M.E.,andBurggren,W.W.,eds.,EnvironmentalPhysiologyoftheAmphibians:ChicagoandLondon,The UniversityofChicagoPress,p.250–276. Poulson,T.L.,andWhite,W.B.,1969,Thecaveenvironment:Science, v.165,no.3897,p.971–981. S EASONALDISTRIBUTIONANDCIRCADIANACTIVITYINTHETROGLOPHILELONG FOOTEDROBBERFROG ELEUTHERODACTYLUSLONGIPES (A NURA : B RACHYCEPHALIDAE ) AT L OS R ISCOS C AVE ,Q UERE TARO ,M EXICO :F IELDANDLABORATORYSTUDIES 30 N JournalofCaveandKarstStudies, April2009

PAGE 8

Reddell,J.R.,1981,AreviewofthecavernicolefaunaofMe xico, GuatemalaandBelize:BulletinoftheTexasMemorialMuseum, v.27,p.1–327. Sokolove,P.G.,andBushell,W.N.,1978,TheChi-Squareperiodogram: Itsutilityforanalysisofcircadianrhythms:JournalofTheoretical Biology,v.72,p.131–60. StatSoftInc.,1999,Statisticaluserguide:CompleteStatisticalSyste m Statsoft.Oklahoma. Taylor,E.,1939,NewSpeciesofMexicanAnura:TheUniversityof Kansas,ScienceBulletin,v.26,no.11,p.385–401. Tocher,K.D.,1950,ExtensionoftheNeyman-Pearsontheoryofteststo discontinuousvariates:Biometrika,v.37,p.130–144. Trajano,E.,Duarte,L.,andMenna-Barreto,L.,2005,Locomotoractivity rhythmsincavefishesfromChapadaDiamantina,northeasternBrazil (Teleostei:Siluriformes):BiologicalRhythmResearch,v.36,no.3, p.229–236. A.E SPINODEL C ASTILLO ,G.C ASTAN O -M ENESES ,M.J.D A VILA -M ONTES ,M.M IRANDA -A NAYA ,J.B.M ORALES -M ALACARA AND R.P AREDES -L EO N JournalofCaveandKarstStudies, April2009 N 31



PAGE 1

OREONETIDESBEATTYI ,ANEWTROGLOBITICSPIDER (ARANEAE:LINYPHIIDAE) FROMEASTERNNORTHAMERICA,AND RE-DESCRIPTIONOF OREONETIDESFLAVUS P.P AQUIN 1,5 ,N.D UPE RRE 2 ,D.J.B UCKLE 3 AND J.J.L EWIS 4 Abstract: AnewtroglobiticLinyphiidae, Oreonetidesbeattyi n.sp.,isdescribedfrom cavesofeasternNorthAmerica.Thespeciesismorphologicallycloseto Oreonetides flavus Emertonandproposedassister-species.Bothspeciesaredescribed,illu stratedand theirdistributionisdocumented.Theintra-specificvariationof O.beattyi isdetailed: femalegenitaliadisplayunusualvariability,butmalesprovidedstable specieslevel diagnosis.AmalefromBullCave(Tennessee)thatshowssignificantgenit alicvariationis problematic,however.Withlimitedsampling,thegeneticbar-codingapp roachdidnot providehelpfulinsightstodetermineifthisspecimenbelongstoadiffer entspecies,is morphologicallyaberrant,orsimplybelongstoapopulationgeographica llydistant enoughtoexplaingeneticvariability.Weproposethecryophilicaffinit ies/relict populationhypothesistoexplaintheecologicalaffinitiesofsomeLinyp hiidaethatare restrictedtocavesinmostoftheirranges,butoccuronthesurfaceatthen orthernedge oftheirdistribution.Wesuggestanevolutionaryscenarioforthedisjun ctdistributionof Oreonetidesbeattyi n.sp.ineasterncavesand O.flavus inmorenorthernlatitudesonthe westcoastofNorthAmerica. I NTRODUCTION Intaxonomy,speciesboundariesaredeterminedbythe examinationofseriesofcloselyrelatedspeciesinorderto identifydistinctivecharactersorgapsingradationof shapes,length,counts,etc.ofvariablemorphological features.Charactersretainedtodelimitspeciesaredetailed andusedasdiagnostic.Intra-specificvariabilityisrarely reportedandremainsaneglectedaspectofmosttaxonomic papers,andconsequently,thereisawidespreadperception thatvariationwithinaspeciesishighlyunusual.Inspider taxonomy,intra-specificvariabilityhasbeendocumented forcolorpatterns[e.g. Araneus (Levi,1971;Courtand Forster,1988), Theridionfrondeum Hentz,1850(Emerton, 1882,plate3,fig.1), Latrodectuskatipo Powell,1870(Vink etal.,2008), Sitticusfasciger (Simon,1880)(Proszynski, 1968)],butcolorationisrathervolatile,easilyalteredin preservedspecimens,andrarelyusedtodelimitspecies. Genitaliccharacters,however,arereputedtobestable withinaspecies,whileprovidingtheneededinformationto distinguishspecies(Eberhard,1986),whichmakesthese featuresidealfortaxonomicpurposes.Intra-specific variabilityofgenitaliaisthereforemuchmoreproblematic. Nonetheless,severalcasesareknown.Forinstance, Roberts(1987,p.180),providedexamplesofintra-specific variabilityofthemalepalpof Araneusdiadematus Clerck (1757)(Araneidae),Levi(1971)illustratedgenitalicvariationofmaleandfemale Araneus ,Gertsch(1984)illustrated thevariationheadmittedforthemalegenitaliaof Eidmanellapallida (Emerton,1875)(Nesticidae)andBlest andVink(2000)documentedthevariabilityoftheretrolateraltibialapophysis(RTA)forafewspeciesof Stiphidiidae.Importantintra-specificvariationsoffemale genitaliahavebeenshownfor Cicurina (Dictynidae) (PaquinandHedin,2004;Paquinetal.,2008),and supportedbygeneticdata(PaquinandHedin,2004, 2007).Theseexamplesaretroublingasmanyspeciesare basedontheexaminationoffewspecimens,wherespecies aredifferentiatedonlybyminorgenitalicdetailsandare foundinsympatry. Asoundevaluationofintra-specificvariabilityislargely dependentonthenumberofspecimensavailable,butinthe caseofrarespecies,oftenknownonlyfromoneortwo specimens,itisimpossibletoassess.Misevaluationofthis variationcanleadtotheerroneousinterpretationofspecies limitsbecausecharactersthatarerandomlyvariablewithin aspeciesmustnotbeusedtoestablishtaxa.Insuchcases,a re-assessmentofthetaxonomybasedonlongerseriesof specimensresultsinsynonymies.Synonymsusuallyhave relativelyminorsignificancebecausespeciesnamesare scientifichypothesestoberefuted,modified,redefined,or improved.However,synonymieshavedeepimpactswhen involvingspeciesthatarelegallyprotected(threatened, 1 CaveandEndangeredInvertebrateResearchLaboratory,SWCAEnvironment al Consultants,4407MontereyOaksBoulevard,Building1,Suite110,Austin ,Texas, 78749,U.S.A.ppaquin@swca.com 2 DivisionofInvertebrateZoology,AmericanMuseumofNaturalHistory,Ce ntral ParkWestat79thStreet,NewYorkNY10024,U.S.A.nduperre@amnh.org 3 620AlbertAvenue,Saskatoon,SK,S7N1G7,Canada,djbuckle@shaw.ca 4 Lewis&AssociatesLLC,Cave,Karst&GroundwaterBiologicalConsulting, 17903StateRoad60,Borden,IN47106-8608,U.S.A.lewisbioconsult@aol. com 5 Thisispublicationno.10oftheKarstBiosciencesandEnvironmentalGeop hysics ResearchLaboratories,SWCAEnvironmentalConsultants P.Paquin,N.Dupe rre ,D.J.Buckle,andJ.J.Lewis– Oreonetidesbeattyi ,anewtroglobiticspider(Araneae:Linyphiidae)fromeastern NorthAmerica,andre-descriptionof Oreonetidesflavus JournalofCaveandKarstStudies, v.71,no.1,p.2–15. 2 N JournalofCaveandKarstStudies, April2009

PAGE 2

Figures1–9. Oreonetidesbeattyi n.sp.1,faceofmale,frontalview(CalfCave,Tennessee);2,faceoffemal e,frontalview (Smith’sFollyCave,Indiana);3,faceoffemale,frontalview(JJ’sSiste rCave,Indiana);4,palpusofmale,retrolateralview; 5,palpusofmale,ventralview;6,clearedepigynum,ventralview;7,clea redepigynum,dorsalview,8,clearedepigynum, lateralview;9clearedepigynum,posteriorview.Abbreviationsused:CD copulatoryducts,COcopulatoryopenings,E embolus,FDfertilizationducts,LClamellacharacteristica,Pparacymb ium,SAsuprategularapophysis,SCscape,S spermatheca,SSsecondaryspermatheca.ScalebarsforFigures1–3,5–9 = 0.1mm;Figure4 = 0.05mm. P.P AQUIN ,N.D UPE RRE ,D.J.B UCKLE AND J.J.L EWIS JournalofCaveandKarstStudies, April2009 N 3

PAGE 3

listed,orspeciesconsideredforlisting)(seeLongacre,2000; Benderetal.,2005).Legalprotectionisultimatelyfunneled intoaspeciesname,asystemthatdoesnotharmonizewell withadisciplinethatprogressesbyproposingrevised hypotheses.Forinstance,Paquinetal.(2008)synonymized twonamesofeyelesstroglobitic Cicurina thatwerethe focalpointofalegaldebatearoundthetaxonomic soundnessofcavespidersthatarespeciesofconcernin Texas(UnitedStates).Thissynonymywasadirect consequenceofaninitialmisevaluationofintra-specific variabilityduetotherarityofidentifiablematerial.Such synonymysuggeststhatcautionshouldbeusedinthe descriptionofcave-restrictedtaxa,becausetherarity, narrowdistributions,andhighdependenceonsensitive habitatsmaketroglobitesidealcandidatesforenhanced conservationmeasures. RecentcavesurveyscarriedoutinIndiana(Lewisand Rafail,2002;Lewisetal.,2004;LewisandLewis,2008a,b) andTennessee(Reeves,2000;Lewis,2005)revealedthe existenceofanewtroglobiticspiderbelongingin Oreonetides Strand1901.Furtherresearchledtothediscoveryof additionalspecimensfromothereasterncavesinmuseum collections.Basedonthisinformation,weconducted additionalsamplingin2004inordertoincreasethe numberofspecimensavailableforstudy,particularlythe males.Thecollectionoffreshmaterialallowedtheuseofa DNAbar-codingapproach(Hebertetal.,2003)foran independentassessmentofspecieslimitsandvariability.In thepresentpaper,wedescribethisnewtroglobiticspider anddocumentitsintra-specificvariability.Were-describe theepigeanspecies Oreonetidesflavus (Emerton,1915) whichishypothesizedassister-species.Thedistributionof thetwospeciesisdiscussedinthelightofapossible evolutionaryscenariobehindthespeciationofthetroglobiticspecies.Thelimitationofthebar-codingapproachis alsobrieflyaddressed. M ETHODS S PECIMEN E XAMINATION Specimenswereexaminedin70%ethanolunderaSMZUNikondissectionmicroscope.ANikonCoolpix950 digitalcameraattachedtothemicroscopewasusedtotake aphotographofthestructurestobeillustrated.Thedigital photowasthenusedtotraceproportions,theillustration wasdetailedandthenshadedbyreferringtothestructure underthemicroscope.Femalegenitaliawereexcisedusing asharpentomologicalneedleandtransferredtolacticacid toclearnon-chitinoustissues.Atemporarylacticacid mountwasusedtoexaminethegenitaliaunderan AmScopeXSGSeriesT-500compoundmicroscope,where genitaliawerephotographedandillustratedasexplained above.Forthestudyoftheembolicdivision,themale palpswereplacedfor 10minutesinwarmKOHand washedin80%alcohol. Allmeasurementsareinmillimetersandweremade usinganopticalmicrometeronthemicroscope.When possible,fivespecimensofeachsexweremeasuredforthe descriptions.CalculationforthelocationofTmIfollows Denis(1949).Palpalandepigynalterminologyfollows Saaristo(1972),vanHeldsingen(1981)andHormiga (1994).Colordescriptionwasdoneunderhalogenlighting, usingtraditionalcolornames.Subsequently,wematched thecolorofthespecimentoareferencePantonechart (PantoneFormulaGuide,solidmatte)andaddedthecolor codetothedescription.Latitudeandlongitudedataare givenindecimalsandshouldbeconsideredanapproximation,andinthecaseofcavelocations,theyarenotgiven inordertopreservetheconfidentialityoftheinformation. M OLECULAR A NALYSIS Specimensrecentlycollectedwerepreservedinthefield in100%ethanolandpreservedonicetoavoidDNA degradation(Vinketal.,2005).DNAextractionwasdone usingaDNEasy H kitfollowingthemanufacturerÂ’s indications.UsingPCR(polymerasechainreaction),we amplifieda 1kbfragmentoftheCytochromeOxidaseI (CO1)mtDNAgeneusingprimersC1-J-1751-SPID,C1-J2309,C1-N-2568,andC1-N-2776-SPID(Hedinand Maddison,2001,Vinketal.,2005)andPCRprotocols similartothosedetailedinPaquinandHedin(2004).PCR productswerepurifiedusingtheWizard H SVGelandPCR Clean-upSystemofPromegafollowingthemanufacturerÂ’s indicationsandsequencedatthecorefacilities(Portland StateUniversityandBerkeleyUniversity).Templateswere sequencedinbothdirectionsforeachfragment,usingPCR primersandexceptfortheshorterfragment,only sequencedfromthe5-foot-endusingC1-N-2776-SPID. Thesequencesreadwereassembledintosequencecontigs andeditedusingSequencer4.5andMacClade4.0 (MaddisonandMaddison,2003).MrModeltestversion 2.2(PosadaandCrandall,1998;Nylander,2004)and PAUP*4.0b10(Swofford,2002)wereusedtoselectabestfitmodelofmolecularevolutionusingtheAkaike InformationCriterion(AIC)(seePosadaandBuckley, 2004).Phylogeneticanalyseswereconductedusing MrBayesversion3.1.2(RonquistandHuelsenbeck,2003) software.WeusedaGTR I Gmodelwithfixedsubstitutionandrateparameters(obtainedinMrModeltest)to conductanun-partitionedBayesiananalysisusingthis best-fitmodelofmolecularevolution.Allanalyseswere runfortenmilliongenerations,samplingevery1000thtree (allotherparameterssettoprogramdefaults(Ronquist andHuelsenbeck,2003)).Majorityruleconsensustrees wereconstructed,discardingthefirst2000treesasburn-in. Theanalysisincludestwoothersurface Oreonetides species, Pithyohyphantes sp.,abasalLinyphiidaeand Pimoa sp. (Pimoidae),asistergrouptoLinyphiidae(Hormiga1994, 2000)usedasoutgroup. ThebulkofspecimenswerecollectedbyJJL,andby PP,NDandJeremyMiller(curatedintheCollection O REONETIDESBEATTYI ANEWTROGLOBITICSPIDER (A RANEAE :L INYPHIIDAE ) FROM E ASTERN N ORTH A MERICA,ANDRE DESCRIPTIONOF O REONETIDESFLAVUS 4 N JournalofCaveandKarstStudies, April2009

PAGE 4

Paquin-Dupe rre ;CPAD,Shefford,Que bec).Voucher specimensarealsodepositedinthecollectionofDonald J.Buckle(DBC,Saskatoon,Saskatchewan,Canada). Specimensfromthefollowingcollectionswerealso examined:AmericanMuseumofNaturalHistory (AMNH,NewYork,USA);CanadianNationalCollection (CNC,Ottawa,OntarioCanada),MuseumofComparativeZoology(MCZ,Harvard,Cambridge,Massachusetts, USA);LymanEntomologicalMuseum(LEM,McGill University,Ste-Anne-de-Bellevue,Canada)andtheBurke MuseumUniversityofWashington(UWBM,Seattle, Washington,UnitedStates). T AXONOMY Family: LinyphiidaeBlackwall,1859 Genus: Oreonetides Strand,1901 TypeSpecies: Oreonetidesvaginatus (Thorell,1872). Diagnosis: SeevanHeldsingen(1981). Composition: Includes15describedspecies,6ofwhich arefoundinNorthAmerica.SeveralNorthAmerican speciesremainundescribed. Distribution: Russia,China,Mongolia,Europe,Japan andNorthAmerica(Platnick,2008). N EW S PECIES Oreonetidesbeattyi (Figs.1–19and27–28) Oreonetidesflavus (Emerton,1915)(Reeves,2000). Misidentification. Oreonetides sp.(Peck,1998;Gertsch,1992). TypeMaterial: HOLOTYPE:UnitedStates:Lawrence Co.,Smith’sFollyCave,TincherHollowSpecialArea, HoosierNationalForest,26.viii.2004,incave,1 = ,P. PaquinandJ.Miller(AMNH). MaterialExamined: UnitedStates: Indiana :Jefferson Co.,GraysCave,onMiddleForkCreek,BigOaks NationalWildlifeRefuge,03.ii.2001,incaveonrotting wood,handcollected,1 R ,J.Lewis,(CPAD);25.viii.2004, incave,handcollected,2 R ,P.Paquin,(CPAD);Lawrence Co.,SullivanCave,2mi.W.Springville,29.xii.20071 R ,J. LewisandS.Lewis(CPAD);JJ’sSisterCave,1mi.SW Bryantsville,29.ix.2000,incave,handcollected,3 R ,J. LewisandR.Burns(CPAD);29.ix.2000,incave,hand collected,1 R ,J.LewisandR.Burns(CPAD);26.viii.2004, incave,handcollected,1 = ,J.Miller(CPAD);Smith’s FollyCave,TincherHollowSpecialArea,Hoosier NationalForest,29.vii.2001,incave,handcollected,1 R J.LewisandS.Rafail(CPAD);29.ix.2000,Berlese extractionofleaflitterfromcave,1 R ,J.LewisandR. Burns(CPAD);25.vii.2002,2 = 1 R ,J.Lewis(DBC); 26.viii.2004,incave,1 = 8 R ,P.PaquinandJ.Miller (CPAD);27.x.2001,incave,1 R ,J.LewisandR.Burns (CPAD);RipleyCo.,LouisNeillCave,BigOaksNational WildlifeRefuge,16.iv.2001,incave,handcollected,1 R ,J. Lewis,S.MillerandT.Vanosdol-Lewis(CPAD); Maryland :WashingtonCo.,Snivley’sCave,nearKeedysville, 12.ix.1968,1 R [nocollector](AMNH);Snivley’sCave No.2,nearEaklesMill,04.viii.1973,betweenrocksand litter,1 R ,A.NordenandB.Ball(AMNH);03.v.1969,1 R (AMNH); Pennsylvania :ArmstrongCo.,HinemanCave,2 mi.W.BuffaloMills,11.vii.1957,1 = ,C.KrekelerandJ.R. Himann(AMNH);DauphinCo.,IndianEchoCave, 16.i.1937,incave,handcollected,1 R ,K.Dearwolf (AMNH);BrownstoneCave,16.i.1937,incave,hand collected,6 R ,K.Dearwolf(AMNH); Tennessee :Blount Co.,BullCave,GreatSmokyMountainsNationalPark, 02.viii.2000,incave,handcollected,1 R ,M.Hedin (CPAD);CalfCave 1,GreatSmokyMountainsNational Park,28.vii.2004,incave,handcollected,1 = ,P.Paquin (CPAD);MarionCo.SpeegleCoveCave,7mi.N.W. Jasper,28.x.2004,insidecave,1 R ,J.LewisandC.Holliday (CPAD); Virginia :TazewellCo.,Rosenbaum’sWater Cave,02.ix.1962,1 R ,J.Holsinger(AMNH);Montgomery Co.,VickersRoadCave,16.x.1971,1 R ,L.M.,T.B.L. FergusonandJ.R.Holsinger(AMNH). Diagnosis: Malesandfemalesof O.beattyi n.sp.differ fromallothermembersofthegenusbythepresenceof noticeablyreducedeyeswhichvaryfromapproximately onethirdthesizeoftheeyesof O.flavus totiny,palewhite spots.Malesarediagnosedbythebidentateridgeofdistal armofparacymbium,theirshortterminalapophysisand lamellacharacteristica.Femalesarecharacterizedbytheir ovalspermathecae,andtheirelongatedsecondaryspermathecaepositionedventrally. Description: Male(n 4):Totallength:1.54 6 0.18; carapacelength:0.72 6 0.04;carapacewidth:0.56 6 0.04; carapaceoff-whitetolightyellow-orange(142M),smooth, shiny,with4–5erectsetaealongmidline.Eyesofirregular form,reducedinparticulartheanteriormedian(AME) andtheposteriormedianeyes(PME)(Figs.1–3),toalmost completelyabsent,withpresenceofwhitepalespots. Cheliceraelightyellow(134M)toyellow-orange(142M), promarginwith4teeth,retromarginwith4–5denticles (Figs.1–3).Cheliceralstridulatoryorganvisible,with 35–40ridges.Abdomenoff-whitetolightgray(Warm gray1M),denselycoveredwithlongsemi-erectsetae, venterofabdomenwithovalstriatedepigastricplates.Legs lightyellow(134M)tolightyellow-orange(142M);leg formula4-1-2-3;tibiaI-IIIwithtwolongdorsalmacrosetae,tibiaIVwithonesuchsetae;metatarsusIwithdorsal trichobothrium,TmIsituatedat0.35-0.41;metatarsusIV lackingdorsaltrichobothrium,coxaIVwithsmallstridulatorypick.TotallengthlegI:2.26 6 0.4;legII:2.16 6 0.12;legIII:1.89 6 0.08;legIV:2.40 6 0.13.Palpalfemur withsmall,basalstridulatorypick.Palpuslength:0.33 6 0.05.Malepalp:cymbiumwithlaterallobe(Fig.4); paracymbium(P)withonebasalprotrusioncup-shaped, tipcoveredwithminusculepapillae,secondprotuberance bearing5setae(Fig.4),trunkofparacymbiumbearingan isolatedsetadistallyandalongitudinalridgebasally,distal armofparacymbiumwithsclerotizedbidentateridge (Fig.4);embolus(E)tri-partate,middlepartbearingthe P.P AQUIN ,N.D UPE RRE ,D.J.B UCKLE AND J.J.L EWIS JournalofCaveandKarstStudies, April2009 N 5

PAGE 5

spermduct(Figs.10and11);radix(R)withasmall pointed,mesalprojection(Figs.10and11);terminal apophysis(TA)short,basallyenlargedandpointed apically(Figs.10and11)lamellacharacteristica(LC), short,large,translucentandcurved(Figs.10and11), distalendslightlyvariable,eitherroundedandrugose (Figs.10–12)orpointedandsmooth(Figs.11and13). Female(n 5):Totallength:1.85 6 0.2;carapace length:0.80 6 0.04;carapacewidth:0.58 6 0.06;overall colorationasinmale.Carapacesmoothandshiny,with4– 5erectsetaealongmidline.Eyesasinmale,with(AME) and(PME)reduced,toalmostcompletelyabsent,with presenceofwhitepalespots.Cheliceralpromarginwith4 teeth,retromarginwith4–5denticles.Cheliceralstridulatoryorganwith 30ridges.Abdomendenselycovered withlongsemi-erectsetae;venterwithovalstriated epigastricplates.Legformula4-1-2-3;tibiaI-IIIwithtwo dorsalmacrosetae,andtibiaIVwithonesuchsetae; metatarsusIwithdorsaltrichobothriumsituatedat0.36– 0.39;metatarsusIVlackingdorsaltrichobothrium;coxaIV withsmallstridulatorypick.TotallengthlegI:2.66 6 0.4; legII:2.57 6 0.24;legIII:2.32 6 0.23;legIV:2.86 6 0.31. Palpalfemurwithsmallbasalstridulatorypick;palpal tarsuswithoutclaws.Epigynumwidth0.24 6 0.03. Epigynumconsistsofatightlyfoldedscape(Fig.8),distal partofscapeprotrudingbasally(Figs.6and7),primary spermathecaeovale(S),secondaryspermathecae(SS) elongated,situatedeitherventro-laterally,ventro-mesally, orventro-internallyinrelationtoprimaryspermathecae (Figs.14–19),copulatoryducts(CD)longandsinuous (Figs.14–19),copulatoryopenings(CO)locatedmidway onventralsurfaceofscape(S)(Fig.7),fertilizationducts (FD)rathershortandstraight(Figs.6and7). Distribution: KnownonlyfromcavesfromtheAppalachianValleyinVirginia,theAppalachianPlateaufrom PennsylvaniasouthtoTennesseeandwesttotheInterior LowPlateausinIndiana(Figs.27–28). Habitat: Allknownspecimenswerecollectedincaves. Etymology: NamedinhonorofJosephA.Beatty, professoremeritus,DepartmentofZoology,Southern IllinoisUniversity-Carbondale. S PECIES Oreonetidesflavus (Emerton,1915),(Figs.20–25and 28) Micronetaflava (Emerton,1915,plateIII,fig.2). Aigolaflava (Crosby,1937,plateI,figs.5–6). Oreonetidesflavus (vanHelsdingen,1981,figs7–12; Crawford,1988;Buckleetal.,2001). TypeMaterial: Micronetaflava Emerton,MCZ,EXAMINED. Label:" Micronetaflava Emerton,Canada:Alberta: Louggan,LakeLouise,inmoss,Aug.10,1905J.H.Emerton Collection"[51.4252 N,116.1805 W]MCZ 21313holotype: male,syntype:female. MaterialExamined: Canada: Alberta :36kmNW Hinton,3.75WofRockLakeRd.[53.9333 N, 118.0833 W]03.-17.vi.2004,1 = ,pitfall,H.Williams (LEM);CameronLake,WatertonLakesNationalPark [49.0166 N,114.0667 W]04.vii.1980,interceptiontrap,1 R H.J.Teskey(CNC);BowPass,64miNWofBanff [53.7167 N,116.5002 W]12.x.1953,Berlesesamplein spruceduff,5 R ,O.Peck(CNC).UnitedStates: Washington ,OkanaganCo.TiffanySpingCamp,6700 [2042m] [48.699 N,119.955 W],31.vii.1985,2 R ,R.Crawford (UWBM). Diagnosis: Malesandfemalesof O.flavus aredistinguishedfrom O.beattyi bythepresenceofwelldeveloped eyes.Malesarefurtherdiagnosedbytheirlongandspineliketerminalapophysis(TA)andtheirsignificantlylonger lamellacharacteristica(LC).Femalesarediagnosedby theirroundedspermathecae(S)andthedorso-mesally positionedsecondaryspermathecae(SS). Description: Male(n 2):Totallength:1.59 6 0.11; carapacelength:0.75 6 0.07;carapacewidth:0.61 6 0.01;carapacesmooth,shiny,lightorange(135M),4–5 erectsetaealongmidline.Eyeswell-developed,rounded andringedwithblackpigment(Fig.20).Sternumlight yellow-orange(134M).Cheliceraelightorange(135M), promarginwith5–6teeth(Fig.20),retromarginwith5–6 denticles.Cheliceralstridulatoryorganvisiblewith 30 striae.Abdomenlightgray(warmgray1M)densely coveredwithlongsemi-erectsetae;venterofabdomen withovalstriatedepigastricplates.Legslightyellow (1205M)tolightorange(134M),legformula4-1-2-3; tibiaI-IVwithtwolongdorsalmacrosetae,metatarsusI withdorsaltrichobothrium,TmIlocatedat0.33, metatarsusIVlackingdorsaltrichobothrium,coxaIV withsmallstridulatorypick.TotallengthlegI:2.15 6 0.04;legII:2.00 6 0.08;legIII:1.79 6 0.13;legIV:2.40 6 0.02.Palpalfemurofwithsmall,basalstridulatory pick.Palpuslength:0.35 6 0.02.Malepalp:cymbium withlaterallobe(Fig.21);paracymbium(P)withone basalcup-shapeprotrusion,coveredwithminuscule papillae,secondbasalprotuberancebearing4setae (Fig.21),trunkofparacymbiumbearinganisolatedseta distallyandalongitudinalridgebasally,distalarmof paracymbiumwithsclerotizedridge(Fig.21);embolus (E)tri-partate(Figs22–23),middlepartbearingthe spermduct;radix(R)withasmallrounded,mesal projection(Fig.23);terminalapophysis(TA)long, smoothlytaperingapically(Fig.23);lamellacharacteristica(LC)transparent,long,curved,withrugosetip (Figs22–23). Female(n 6):Totallength:1.84 6 0.16;carapace length:0.80 6 0.06,carapacewidth0.61 6 0.01;overall colorationasinmale.Carapacesmoothandshiny,with 4-5erectsetaealongmidline.Eyesnormal,rounded, ringedwithblackpigment.Cheliceralpromarginwith5– 6largeteeth,retromarginmarginwith5–6small denticles.Cheliceralstridulatoryorganwith 25striae. O REONETIDESBEATTYI ANEWTROGLOBITICSPIDER (A RANEAE :L INYPHIIDAE ) FROM E ASTERN N ORTH A MERICA,ANDRE DESCRIPTIONOF O REONETIDESFLAVUS 6 N JournalofCaveandKarstStudies, April2009

PAGE 6

Figures10–19. Oreonetidesbeattyi n.sp.10,embolicdivisionofmalepalpus,schematicview(CalfCave,Tenn essee);11, embolicdivisionofmalepalpus,schematicview(Smith’sFollyCave,Indi ana);12,lamellacharacteristicaofmalepalpus(Calf Cave,Tennessee);13,lamellacharacteristicaofmalepalpus(Smith’sFo llyCave,Indiana);14,clearedepigynum,ventralview (BullCave,Tennessee);15,clearedepigynum,dorsalview(BullCave,Ten nessee);16,clearedepigynum,ventralview (Snivley’sCave,Maryland);17,clearedepigynum,dorsalview(Snivley’ sCave,Maryland);18,clearedepigynum,ventralview (Rosenbaum’sCave,Virginia);19,clearedepigynum,dorsalview(Rosenb aum’sCave,Virginia).Abbreviationsused:E embolus,LClamellacharacteristica,Rradix,SAsuprategularapophysis ,SPTsuprategulum,TAterminalapophysis.Scale barsforFigures10–19 = 0.05mm. P.P AQUIN ,N.D UPE RRE ,D.J.B UCKLE AND J.J.L EWIS JournalofCaveandKarstStudies, April2009 N 7

PAGE 7

Abdomencoveredwithlongsemi-erectsetae;venterwith ovalstriatedepigastricplates.Legformula4-1-2-3;tibia I-IVwithtwolongdorsalmacrosetae,metatarsusIwith dorsaltrichobothrium,TmIlocatedat0.33–0.36, metatarsusIVlackingdorsaltrichobothrium,coxaIV withsmallstridulatorypick.TotallengthlegI:2.50 6 0.11;legII:2.32 6 0.12;legIII:2.13 6 0.09;legIV: 2.81 6 0.02.Palpalfemurwithsmallstridulatory picksetbasally;palpaltarsuswithnoclaws.Epigynum width:0.29 6 0.01.Epigynumconsistsofatightlyfolded scape(S),distalpartofscapeprotrudingbasally (Figs.24–25);primaryspermathecaerounded(S) (Figs.24–25);secondaryspermathecae(SS)elongated, situateddorso-mesally(Figs.24–25);copulatoryduct (CD)longandsinuous(Figs.24–25);copulatoryopenings(CO)locatedmidwayonventralsurfaceofscape(S) (Figs.24–25);fertilizationducts(FD)rathershortand curved(Fig.25). Distribution: Canada:Alberta;UnitedStates:Washington(Fig.28). Habitat: Epigeanspecies,apparentlyassociatedwith forestlitter. Note: ThespecimenfromSeaCliff,NewYork,reported byvanHelsdingen(1981)couldnotbelocatedfor examinationdespitenumerouseffortsattheMuseumof ComparativeZoology(Harvard),anditsidentityremains uncertain.Thissurfacerecordisindicatedonthedistributionmapbyastar(Fig.28,seealsodiscussion). R ESULTS M ORPHOLOGICAL V ARIABILITY Theexaminationofavailablematerialclearlyallowsthe distinctionof O.flavus from O.beattyi n.sp.andshows similaritiesthatsuggestacloserelationshipbetweenthe twospecies.Theamountofeyereductionobservedin O. beattyi rangesfromsignificantlyreducedtoalmost completelyabsentwithonlypalespotsremaining,which leavesnodoubtabouttroglobiticadaptations. Oreonetides flavus displayednovariabilityofgenitalicfeatures. Examinationoftheepigynumof O.beattyi ,however, showedsurprisingvariability,particularlyinthedistance betweentheprimaryspermathecaeandthepositionofthe secondaryspermathecaeinrelationtotheprimary spermathecae(Figs.14–19).Suchvariationwasobserved betweenfemalescollectedinthesamecaveandsometimes withinasinglefemaleassomeepigynumswerenot symmetrical.Malesof O.beattyi showednovariationin palpalmorphology,exceptfortheonecollectedfromBull Cave,Tennessee(Figs.10and12),whichdifferedinthe shapeandtextureofthelamellacharacteristicaandcould notbecomfortablyassignedtothespecies. M OLECULAR A NALYSIS Sevenindividualsbelongingto O.beattyi weresequencedinordertotestifmorphologicalvariabilitycould revealmultiplespecies.Thephylogenetictree(Fig.26) showsastronglysupportedtipclade(0.92)offive individualsfromthreecavesinIndianathatdisplaylittle geneticvariability.Asexpected,specimenscollectedinthe samecavesharemoresimilarhaplotypes.However,the specimenfromCalfCave 1inTennessee( Oreonetides -7) isdistinctfromthetipcladewithastronglysupported (value1.00)relativelyshortbranchlength. D ISTRIBUTION Giventheproposedrelationshipassisterspecies,the disjunctdistributionisremarkablewith O.flavus occurring inthewesternpartofthecontinentandcomparatively moretothenorth. O.beattyi n.sp.isconfinedtotheeast andfoundexclusivelyincaves(Fig.28). D ISCUSSION S PECIES V ARIABILITYOF O REONETIDESBEATTYI N.SP. Theseriesofspecimensstudiedprovidesinsightsintothe understandingofgenitalicvariabilityofthespecies, particularlyforfemales.Variationofthespermathecaeis rarelyreportedinLinyphiidae,asetofcharactersotherwise perceivedasstable.However,basedonanumberof specimensthatallowedarobustassessmentofmorphologicalvariation,Schikora(1995)reportedsignificantintraspecificvariabilityofmaleandfemalegenitalicstructures.In atroglobiticlinyphiid,comparableintra-specificvariability hasbeenreportedforfemalesof Porrhommacavernicola (Keyserling,1886)byMiller(2005).Inthepresentcase,the collectionofseveralspecimensof O.beattyi n.sp.fromthe samecavedisplayingvariabilityoftheepigynumis convincingevidencethatthevariationisintra-specific. Thereisnodoubtthatfemalecharacterscannotbeused reliablytorecognizemorethanonespecies.Comparatively, animportantradiationoftroglobiticspidersfoundinTexas ( Cicurina ,subgenus Cicurella )(seeGertsch,1992;Cokendolpher,2004;PaquinandDupe rre ,2009)didnotbenefit fromalongseriesofspecimens,andmanyspecieswere describedonthebasisofasinglefemale,whichledto taxonomicconfusion(Paquinetal.,2008). Theinformationprovidedbythemaleshowever,may suggestadifferentinterpretation.Thegenitaliaofall knownmalesaresimilar,evenbetweenlocalitiesthatare distantfromeachother,suggestingasinglespecies. However,themalecollectedfromBullCave(Tennessee) displayednoticeablevariability,withabroaderlamella characteristicaandadifferentconfigurationoftherugose tip(Figs.10and12).Consideringthestabilityofthis structureinallotherknownmales,thiscouldbe interpretedasaspecies-levelcharacter,althoughslight variabilityofthelamellacharacteristicahasbeenshownin generasuchas Agyneta Hull,1911(SaaristoandKoponen, 1998;Dupe rre andPaquin,2007)and Maro O.PickardCambridge,1906(DondaleandBuckle,2001). O REONETIDESBEATTYI ANEWTROGLOBITICSPIDER (A RANEAE :L INYPHIIDAE ) FROM E ASTERN N ORTH A MERICA,ANDRE DESCRIPTIONOF O REONETIDESFLAVUS 8 N JournalofCaveandKarstStudies, April2009

PAGE 8

Figures20–25. Oreonetidesflavus .20,faceofmale,frontalview;21,palpusofmale,retrolateralview;22, palpusofmale, ventralview;23,embolicdivisionofmalepalpus,schematicview;24,cle aredepigynum,ventralview;25,clearedepigynum, dorsalview.Abbreviationsused:CDcopulatoryducts,COcopulatoryopen ings,Eembolus,FDfertilizationducts,LClamella characteristica,Pparacymbium,Rradix,SAsuprategularapophysis,SCs cape,Sspermatheca,SPTsuprategulum,SS secondaryspermatheca,TAterminalapophysis.ScalebarsforFigures20– 25 = 0.1mm. P.P AQUIN ,N.D UPE RRE ,D.J.B UCKLE AND J.J.L EWIS JournalofCaveandKarstStudies, April2009 N 9

PAGE 9

Themoleculardataprovidesupportfortheidentificationofthevariabilityoffemalegenitaliaasintra-specific (greybox,Fig.26).ForthemalefromBullCave,the phylogenetictreedoesnotsupportorrejectthatthespecies foundinthatcavediffersfrom O.beattyi .Afemale ( Oreonetides -7,arrowsinFig.26)fromCalfCave 1, whichislocatedonlyafewmetersfromBullCavewithin thesamesinkholeandlikelyharborsthesame Oreonetides species,wasincludedforDNAanalysis.Thegenetic distinctivenessof Oreonetides -7doesnotallowusto determinewhetherthisisduetoitsbeingadifferent species,partofanisolatedpopulation,restrictedgeneflow asexpectedbetweencavepopulations,orgeographic distancebetweenBullCaveandCalfCave 1andthe clusterfromIndiana(seeFig.27).Sucharesultclearly exemplifiesoneofthenumerouslimitationsinthe applicationofthegeneticbar-codeapproachforspecies levelidentificationasadvocatedbyHebertetal.(2003).A treewith Oreonetides -7nestedwithinthetipclade,or Oreonetides -7asdistinctas O.filicatus or O.flavescens wouldhaveprovidedinterpretableinsights,butthetree obtainedhereisinconclusiveinthatregard.Thephysical barrierstodispersalandgeneflowinherenttocavelife resultinadditionaldifficultiesforaccurateinterpretations, especiallywithalimitednumberofspecimens,asisoften thecasefortroglobites.PaquinandHedin(2007)showed geneticdivergenceswithinasingletroglobiticspiderinan areaof 30 3 15km,thataregreaterthanforallNorth AmericaformobilespeciesofLepidoptera(Niceetal., 2005).Suchdiscrepanciesarestillpoorlyunderstoodand suggestthecautioususeofthebar-codeapproach. Aswiththemorphologicalandmoleculardata,geographyanddistributiondonotprovideinsightsthatfavor eitheradistinctspeciesorintra-specificvariability.The clusterfromTennessee(thatincludesBullCaveandCalf Cave 1)couldeitherharboradifferentspeciesoradistinct populationofthesamespecies,andbothscenarioswouldbe geographicallycohesive.Basedonavailabledata,itseems besttoconsiderthemalefromBullCavetobelongto O. beattyi n.sp.andrepresentvariationwithinthesamespecies. Asinglespecimendoesnotprovideenoughcertitudeto discardthepossibilityofanaberrantspecimenandproposea robustspecieshypothesis.However,thisinterpretationmay beeasilyrefutedbyfuturecollectionsofadditionalmales displayingmorphologysimilartothatspecimen. T ROGLOBITIC L INYPHIIDAEANDTHE C RYOPHILIC A FFINITIES /R ELICT P OPULATION H YPOTHESIS Affinitiesforcaveshavebeenpreviouslyreportedinthe genus: Oreonetidesshimizui (Yaginuma,1972)isfoundin Japanesecaves,althoughthespeciesdoesnotdisplay obviousmorphologicaladaptationstocavelifesuchas noticeableeyereduction(Yaniguma,1972;Eskov,1992). IneasternNorthAmerica,severallinyphiidspeciesdisplay differentdegreesoftroglobiticadaptations. Anthrobia mammouthia Tellkampf,1844showspronouncedadaptationswithlegelongationandatotallackofeyes, Porrhommacavernicola (Keyserling,1886)and Islandiana spp.stillhaveeyeremnants,while Phanettasubterranea (Emerton,1875)knownfromatleastelevenU.S.statesand morethanathousandlocalities,doesnotshowstriking morphologicaladaptationstocavelifesuchaspronounced eyereduction.Noneofthesespeciesareknownfrom surfacerecords,even P.subterranea, withitsabsenceof pronouncedtroglomorphiccharacters.Inothercases, however,thedependencetocavehabitatsisnotasclear. Bathyphantesweyeri (Emerton,1875), Centromeruslatidens (Emerton,1882), Taranucnusornithes (Barrows,1940)are widespreadincavesoftheeasternNorthAmerica;almost allrecordsareknownfromcaves,butsomespecimenshave beenfoundonthesurfaceatthenorthernedgeoftheir distributions.Asimilarpatternisobservedfor Oaphantes n.sp.onthewestcoast.Weproposethecryophilic affinities/relictpopulationhypothesistoexplainthis troublesomerestrictiontocavesinsouthernlocationsand surfacerecordsatthenorthernedge,withinthesame species.Theimplicationsofthishypothesishavebeen referredto,atleastindirectly,bysomeauthors(seeBarr, 1967;Peck,1973;PeckandLewis,1978;Barrand Holsinger,1985),butnotformallyproposedassuch. Affinityforcolderconditions,oravoidanceofwarm climate,isadrivingforcebehindthedynamicof distributionranges(ParmesanandYohe,2003)and invasionofcaves(Barr,1967).Facingconditionstending Figure26.BayesianconsensusphylogrambasedonCOI sequencedata.Valuesabovebranchesareposterior probabilities. O REONETIDESBEATTYI ANEWTROGLOBITICSPIDER (A RANEAE :L INYPHIIDAE ) FROM E ASTERN N ORTH A MERICA,ANDRE DESCRIPTIONOF O REONETIDESFLAVUS 10 N JournalofCaveandKarstStudies, April2009

PAGE 10

Figure28.Distributionmap Oreonetidesflavus inNorthAmerica.Thestarsymbolrepresentsanolderrecordof O.flavus thatcouldnotbeverified.Thedistributionof O.beattyi n.sp.isprovided(shadedarea)forcomparativepurposes. Figure27.Distributionmapof Oreonetidesbeattyi n.sp.ineasternNorthAmerica. P.P AQUIN ,N.D UPE RRE ,D.J.B UCKLE AND J.J.L EWIS JournalofCaveandKarstStudies, April2009 N 11

PAGE 11

towardswarmeranddryerclimate,cryophilicspecies shiftedtheirrangesnorthward,leavingbehindpockets ofpopulationsincoolhabitatsuchasbogs,mountain tops,screeslopes,orcaves(Fig.29).Theseareasare subsequentlyunderwarmerconditions,whichleadtothe extinctionofepigeanpopulations,andfavorthesurvivalin thesubterraneanhabitat,orothersuitablehabitat.Such pressurediffersorisabsentatthenorthernedge,allowing speciestosurviveonthesurface. LikemanycavespidersfromeasternNorthAmerica, Oreonetidesbeattyi n.sp.hasabroaddistributionacross multiplephysiographicregions,includingseveralkarst areasoftheAppalachiansandInteriorLowPlateaus (Hunt,1974).Thetheoreticalbackgroundforthesebroad distributionsisunclear(butseeBarrandHolsinger,1985), particularlywithsuchdifferencesinthedegreeofcave adaptationforthedifferentspecies.Ononehand,this suggeststhatcaveinvasionsbylinyphiidspecieswerenot synchronizedandaremuchmorerecentthancave formation,whichwouldhavelimitedandshapedthe speciesdistributionsaccordingtothephysio-geographic evolutionoftheseareas.Forinstance,aspeciesrestricted toaparticulargeologicalformationwouldsuggestthatthe mechanismsdrivingsuchdistributionarerelatedtothe evolutionofthegeologicalunit,butadistributionthat encompassesseveralkarstunitssuggestsotherwise.Onthe otherhand,themechanismsbehindtheevolutionand dispersaloftroglobitesarepoorlyunderstoodandcould Figure29.Evolutivehypothesisof O.beattyi n.sp.and O.flavus proposedassisterspecies.A)Borealdistributionofthe hypothesisedancestorassociatedwithcoolandmoistforesthabitats.B) Glaciationsaresweepingdowntheancestorintotwo areasdefinedbyforestcomponents.C)Isolationandspeciation.D)Follo wingtheretreatoftheWisconsinglacialicesheet, speciesshiftupnorth.Inthecaseof O.beattyi n.sp.,thepopulationsfoundinthesouthernlimitsarerestrictedtocave sand northernmostprobablerecordsaresurfaceones; O.flavus isnowfoundinpreviouslyglaciatedareas. O REONETIDESBEATTYI ANEWTROGLOBITICSPIDER (A RANEAE :L INYPHIIDAE ) FROM E ASTERN N ORTH A MERICA,ANDRE DESCRIPTIONOF O REONETIDESFLAVUS 12 N JournalofCaveandKarstStudies, April2009

PAGE 12

involvealternativescenarios.Forinstance,speciesmay dispersebywayofalternatehypogeanhabitatssuchas mammalburrows(SkelleyandKovarik,2001),scree slopes(Ruszykaand Klimes,2005),mesocavernor espace sous-terrainsuperficiel (subterraneanundergroundcompartments)(Juberthieetal.,1980),whichprovidea connectionbetweenapparentlyisolatedcavesystems.In addition,theperceptionofbeingisolatedisoften qualifiedbytheiractualstate;apparentlydisconnected fragmentsmayhavebeenpartofacomplexnetworkthat evolvedintoafragmentedhabitat,givingabiased impressionofindependenceofitsinherentunits.Such fragmentationmaynothavebeenthephysicalconditions presentwhencavelifeevolvedinthesesystems.Another possibilityisthatthedistributionofthesetroglobitesis theresultofmultiplecavecolonizationsbyepigean ancestors.Theintermediatestagesofcavedependence suggestedbythecryophilicaffinities/relictpopulation hypothesisprovidesatheoreticalpathwayforthe evolutionanddistributionoftroglobitesthatexplains distributionsbypostulatingsurfaceconnectionsbetween cavesystemsincoldertimes,followedbytheisolationof disjunctcavepopulationsastheclimatewarmed.Given therangeofspeciesmentionedabove,itseemsunlikely thatasinglescenariocouldexplaintheirwidespread distributions.Theinformationpresentlyavailableisnot sufficienttoconclusivelyfavoranyofthesealternatives, butthewidespreaddistributionandthedifferentdegrees oftroglomorphismobservedineasterncaveLinyphiidae likelyrepresentdifferent,andrelativelyearly,stagesinthe evolutiontowardsstrictdependenceoncavehabitatwhen comparedtothe 80NorthAmericancavespidersthat aretotallyeyeless,displayhighlyfragmenteddistribution, orextremelynarrowendemism(Gertsch,1974;1984; 1992). T HE P ROBLEMATIC R ECORDFROM S EA C LIFF (N EW Y ORK ) Inthelightofactualdata,theidentityofthespecimen reportedbyvanHelsdingen(1981)fromSeaCliff(New York)remainsproblematic(seeFig.28).Giventhe similarityofbothspecies,itispossiblethatthisrecordis amisidentificationof O.beattyi n.sp.,assuggestedbyits locationontheeasternsideofthecontinent.Suchrecord wouldbethefirstsurfacementionofthespecies,otherwise restrictedtocaves.Asurfacerecordwouldindicatesimilar affinitiestothosedisplayedbyto B.weyeri and C.latidens, whicharefoundatthesurfaceatthenorthernedgeoftheir distribution.However,itisalsopossiblethatthisrecordis indeedof O.flavus ,andtheapparentdistributiongapdue toincompletesamplingortoadisjunctdistributionlike thatof Poecilonetaaggressa (ChamberlinandIvie1943) (Paquinetal.,2001).LocationoftheSeaCliffspecimen,or thecollectionofadditionalspecimens,isnecessaryto clarifythesituation. S PECIATION H YPOTHESIS Thestudyof O.beattyi n.sp.and O.flavus leavesno doubtabouttheircloserelatedness.Giventhehighly disjunctdistributionofthetwospeciesandtheclear affinitiesof O.beattyi n.sp.forcavehabitat,wepropose thefollowingevolutionaryscenariotoexplaintheir particulardistribution.LikemanyLinyphiidae,andmost NorthAmerican Oreonetides species(vanHelsdingen, 1981),theproto beattyi/flavus ancestordisplayedaffinities forcoolandmoistenvironmentsandhadwidespread borealdistribution(Fig.29a).DuringtheWisconsinanage, mostlifeformsoccurringinnorthernlatitudeswere displacedsouthwardbyaglaciericecomplexthatcovered nearlyallofCanada(Prestetal.,1967;Matthews,1979; Danks,1993).IncentralNorthAmerica,theice-front habitatconsistedofathinbandofmixedopenconifer forestandtundra,withdryplainsbeyond.Someboreal specieswereabletoliveinthisenvironmentbutmanywere not,andtheirpopulationsweresplitintowidelyseparated Cordilleranandeasternforestcomponents(Scudder,1979; Pielou,1991)(Fig.29b).Prolongedisolationledto speciationintowesternandeasternforms( O.flavus and O.beattyi n.sp.)(Fig.29c).Suchdisjunctdistributionof relatedspeciesissimilartootherpairsoftaxathatdisplay asimilareastern-westerndivision(Scudder,1979).For instance,thetwoNorthAmericanspeciesof Cryphoeca (Araneae,Hahniidae)aresimilarlydistributed: C.montana Emerton1909isfoundintheEast,while C.exlineae Roth 1988isrestrictedtotheWest.Similarpatternsalsohave beenreportedforseveralspeciesofbeetlessuchas Carabidae,Staphylinidae,Helodidae(Scudder,1979).Such eastern-westerndistributionsarewell-knownandusedas broadcategoriesinDanks(1994).Thewarmingclimate andretreatoftheWisconsinglacialicesheetfrom14to11 ka(Peck,1988;Schwert,1992)providedsuitablehabitat forrapidnorthwardexpansionforspecieswithaffinities forcoldandmoistconditions.Theprogressivenorthward movementofspecieswithsuchrequirementsiswellknown (Schwert,1992).Followingthecryophilicaffinities/relict populationhypothesisproposedhere, O.beattyi n.sp shiftednorthwardintheeast,leavingbehindisolated populationsincaves,while O.flavus reoccupiedmore northerlyareaspreviouslynotavailable(Fig.29d).This scenarioisclosetothespeciationhypothesisproposedfor thecavefaunaofIllinoisbyPeckandLewis(1978).It providesbettertheoreticalsupportfortheactualdistributionofthespeciesthantreating O.beattyi n.sp.asa simplepostglacialoffshootfrom O.flavus .Despitetheir clearphylogeneticaffinities,themorphologicaldifferences observedbetweenthetwospeciesarelargeenoughto suggestalongerhistorythanoneoriginatingonlyfrom theendofthelastglaciation.Therarityofthespecies couldresultinincompletesamplingthatmayobscurean accurateassessmentoftheirdistributions,butthe proposedhypothesisseemsthemostplausibleonegiven availabledata.Hopefully,additionalcollectionswilleither P.P AQUIN ,N.D UPE RRE ,D.J.B UCKLE AND J.J.L EWIS JournalofCaveandKarstStudies, April2009 N 13

PAGE 13

confirmtheproposedscenariobyproducingsurface specimensof O.beattyi fromnortheasternNorthAmerica orsuggestabetteronetoexplaintheparticular distributionsanddynamicsoftheserelatedspecies. A CKNOWLEDGMENTS WewouldliketothankJeremyMiller,NathanielMann, andDavidCaudlefortheircompanionshipduringthefield workrelatedtothatproject.Themolecularportionhas beenaccomplishedthankstothegenerosityofSusan Masta(PortlandStateUniversity).Wearealsogratefulto NormanPlatnickandLouSorkin(AMNH,NewYork, NY),CharlesDondale(CNC,Ottawa,ON),Laura LeibenspergerandGonzaloGiribet(MCZHarvard,Cambridge,MA),TerryWheeler(LEM,McGillUniversity, Ste-AnnedeBellevue,QC)andRodCrawford(UWBM, WashingtonUniversity,Seattle,WA)whopermittedusto workwiththematerialundertheirresponsibility.Weare alsogratefultoStewartB.Peck(CarletonUniversity, Ottawa,ON)foraninformativediscussiononthetopic. Finally,weappreciatethecommentsoftwoanonymous reviewers. R EFERENCES Barr,T.C.,1967,Observationsontheecologyofcaves:TheAmerican Naturalist,v.101,p.475–491. Barr,T.C.,andHolsinger,J.R.,1985,Speciationincavefaunas:Annual ReviewofEcologyandSystematics,v.16,p.313–337. Bender,S.,Shelton,S.,Bender,K.C.,andKalmbach,A.,2005,Texas comprehensivewildlifeconservationstrategy(TWAP),TexasParks andWildlife,p.i-xv 1–1131. Blest,A.D.,andVink,C.J.,2000,NewZealandspiders:Stiphidiidae: RecordsoftheCanterburyMuseum,v.13,p.1–27. Buckle,D.J.,Carroll,D.,Crawford,R.L.,andRoth,V.D.,2001, LinyphiidaeandPimoidaeofAmericanorthofMexico:Checklist, synonymy,andliterature, in Paquin,P.,andBuckle,D.J.,eds., Contributionsa `laConnaissancedesAraigne es(Araneae)d’Ame rique duNord,Fabreries,Supple ment10,p.89–191. Cokendolpher,J.C.,2004, Cicurina spidersfromcavesinBexarCounty, Texas:TexasMemorialMuseumSpeleologicalMonographs,v.6, StudiesonthecaveandendogeanfaunaofNorthAmerica,v.4, p.13–58. Court,D.J.,andForster,R.R.,1988,Araneidae-Araneinae:TheSpiders ofNewZealand,v.6,p.68–124. Crawford,R.L.,1988,Anannotatedchecklistofthespidersof Washington:BurkeMuseumContributionsinAnthropologyand NaturalHistory,v.5,p.1–48. Crosby,C.R.,1937,StudiesinAmericanspiders.Thegenus Aigola Chamberlin, in ProceedingsoftheBiologicalSocietyofWashington, v.50,p.35–42. Danks,H.V.,1993,PatternsofdiversityintheCanadianinsectfauna, in Ball,G.E.,andDanks,H.V.,eds.,Systematicsandentomology: diversity,distribution,adaptationandapplication,Memoirsofthe EntomologicalSocietyofCanada,v.165,p.51–74. Danks,H.V.,1994,RegionaldiversityofinsectsinNorthAmerica: AmericanEntomologist,v.41,p.50–55. Denis,J.,1949,NotessurlesE rigonides.XVI.Essaisurlade termination desfemellesd’e rigonides:BulletindelaSocie te d’HistoireNaturellede Toulouse,v.83,p.129–158. Dondale,C.D.,andBuckle,D.J.,2001,Thegenus Maro inNorthAmerica (Araneae:Linyphiidae):Fabreries,v.26,p.9–15. Dupe rre ,N.,andPaquin,P.,2007,Descriptionoffivenewspidersfrom Canada(Araneae,Linyphiidae):Zootaxa,v.1632,p.1–20. Eberhard,W.G.,1986,Whyaregenitaliagoodspeciescharacters?, in Lubin,Y.D.,Eberhard,W.G.,andRobinson,B.C.,eds.,Proceedings oftheNinthInternationalCongressofArachnology,Panama,1983, Washington,D.C.,SmithsonianInstitutionPress,333p. Emerton,J.H.,1882,NewEnglandspidersofthefamilyTheridiidae: TransactionsoftheConnecticutAcademyofArtsandSciences,v.6, p.1–86 pl.I–XXIV. Emerton,J.H.,1915,Canadianspiders,II:TransactionsoftheConnecticutAcademyofArtsandSciences,v.20,p.147–156 pl.II–III. Eskov,K.Y.,1992,Arestudyofthegenericcompositionofthelinyphiid spiderfaunaoftheFarEast(Araneida:Linyphiidae):Entomologica Scandinavia,v.23,p.153–168. Gertsch,W.J.,1974,ThespiderfamilyLeptonetidaeinNorthAmerica: JournalofArachnology,v.1,p.145–203. Gertsch,W.J.,1984,ThespiderfamilyNesticidae(Araneae)inNorth America,CentralAmerica,andtheWestIndies:BulletinoftheTexas MemorialMuseum,v.31,p.1–91. Gertsch,W.J.,1992,DistributionpatternsandspeciationinNorth Americancavespiderswithalistofthetroglobitesandrevisionofthe cicurinasofthesubgenus Cicurella :TexasMemorialMuseum SpeleologicalMonographs,v.3,Studiesontheendogeanfaunaof NorthAmerica,Volume2:Austin,TexasMemorialMuseum, p.75–122. Hebert,P.D.N.,Cywinska,A.,Ball,S.L.,anddeWaard,J.R.,2003, BiologicalidentificationsthroughDNAbarcodes, in Proceedings,RoyalSocietyofLondonSeriesB,supplement,v.270, p.313–321. Hedin,M.C.,andMaddison,W.P.,2001,Acombinedmolecular approachtophylogenyoftheJumpingSpidersubfamilyDendryphantinae(Araneae:Salticidae):MolecularBiologyandEvolution, v.18,p.386–403. Hormiga,G.,1994,Cladisticsandthecomparativemorphologyof linyphiidspidersandtheirrelatives(Araneae,Araneoidea,Linyphiidae):ZoologicalJournaloftheLinneanSociety,v.111,p.1–71. Hormiga,G.,2000,Higherlevelphylogeneticsoferigoninespiders (Aranea,Linyphiidae,Erigoninae):SmithsonianContributionsto Zoology,v.609,p.1–160. Hunt,C.B.,1974,NaturalregionsoftheUnitedStatesandCanada, Secondedition,SanFrancisco,W.H.Freeman&Company,725p. Juberthie,C.,Delay,B.,andBouillon,M.,1980,Surl’existenced’un milieusouterrainsuperficielenzonenoncalcaire:Comptesrendus hebdomadairesdesse ancesdel’Acade miedeSciencesdeParis(Se rie D),v.290,p.49–52. Lewis,J.J.,2005,ThesubterraneanfaunaoftheTennesseeCumberland Plateau,FinalReport,Nashville,TheNatureConservancyof Tennessee,156p. Lewis,J.J.,Burns,R.,andLewis,S.,2004,Thesubterraneanfaunaofthe HoosierNationalForest,Bedford,Indiana,U.S.Departmentof Agriculture,ForestService,190p. Lewis,J.J.,andLewis,S.,2008a,ThesubterraneanfaunaoftheHoosier NationalForest,FinalReport,Bedford,Indiana,U.S.Departmentof Agriculture,ForestService,186p. Lewis,J.J.,andLewis,S.,2008b,ThefaunaofSullivanCave,Lawrence County,IndianaFinalReport,Indianapolis,IndianaDepartmentof NaturalResources,DivisionofNaturePreserves,22p. Lewis,J.J.,andRafail,S.,2002,ThesubterraneanfaunaoftheBigOaks NationalWildlifeRefuge,U.S.DepartmentoftheInterior,Fishand WildlifeService,IndianaDepartmentofNaturalResources,Natural HeritageProgram,77p. Levi,H.W.,1971,The diadematus groupoftheorb-weavergenus Araneus northofMexico(Araneae:Araneidae):BulletinoftheMuseumof ComparativeZoology,v.141,p.131–179. Longacre,C.,2000,DepartmentoftheInterior,FishandWildlifeService 50CFRpart17,RIN1018-AF33,Endangeredandthreatenedwildlife andplants,FinalruletolistnineBexarCounty,TexasInvertebrate speciesasendangered,Washington,D.C.,FederalRegister, v.65(248),p.81419–81433. Maddison,D.R.,andMaddison,W.P.,2003,MacClade4,Release Version4.07,Sunderland,MA,SinauerAssociates,Inc. Matthews,J.V.jr.,1979,Tertiaryandquaternaryenvironments:histori cal backgroundforananalysisoftheCanadianinsectfauna, in Danks, H.V.,ed.,Canadaanditsinsectfauna,MemoirsoftheEntomological SocietyofCanada,v.108,p.31–86. O REONETIDESBEATTYI ANEWTROGLOBITICSPIDER (A RANEAE :L INYPHIIDAE ) FROM E ASTERN N ORTH A MERICA,ANDRE DESCRIPTIONOF O REONETIDESFLAVUS 14 N JournalofCaveandKarstStudies, April2009

PAGE 14

Miller,J.A.,2005,Aredescriptionof Porrhommacavernicola Keyserling (Araneae,Linyphiidae)withnotesonAppalachiantroglobites: JournalofArachnology,v.33,p.426–438. Nice,C.,Anthony,N.,Gelembiuk,G.,Ratermans,D.,andFrenchConstant,R.,2005,Thehistoryandgeographyofdiversification withinthebutterflygenus Lycaeides inNorthAmerica:Molecular Ecology,v.14,p.1741–1754. Nylander,J.A.A.,2004,MrModeltestv2.,EvolutionaryBiologyCentre, UppsalaUniversity. Paquin,P.,andDupe rre ,N.,2009,Afirststeptowardstherevisionof Cicurina :redescriptionoftypespecimensof60troglobiticspeciesof thesubgenus Cicurella (Araneae:Dictynidae),andafirstvisual assessmentoftheirdistribution:Zootaxa,v.2002,p.1–67. Paquin,P.,Dupe rre ,N.,Cokendolpher,J.C.,White,K.,andHedin,M., 2008,Fundamentalimportanceoftaxonomyinconservationbiology: Thecaseoftheeyelessspider Cicurinabandida (Araneae:Dictynidae) ofCentralTexas,includingnewsynonymsandthedescriptionofthe maleofthespecies:InvertebrateSystematics,v.22,p.139–149. Paquin,P.,andHedin,M.,2004,Thepowerandperilsof’molecular taxonomy’:acasestudyofeyelessandendangered Cicurina (Araneae:Dictynidae)fromTexascaves:MolecularEcology,v.13, p.3239–3255. Paquin,P.,andHedin,M.,2007,Geneticandmorphologicalanalysisof specieslimitsin Cicurinaspiders (Araneae,Dictynidae)fromsouthern TravisandnorthernHaysCounties(Texas),withemphasison Cicurinacueva Gertschandrelatives:Austin,Texas,ReporttoUnited StatesFishandWildlifeServices,p.1–40. Paquin,P.,LeSage,L.,andDupe rre ,N.,2001,FirstCanadianrecordsof Tenuiphantescracens and Walckenaeriaclavipalpis (Araneae:Linyphiidae),plusthirteennewprovincialrecordsandaconfirmationfor Que bec:EntomologicalNews,v.112,p.271–277. Parmesan,C.,andYohe,G.,2003,Agloballycoherentfingerprintof climatechangeimpactsacrossnaturalsystems:Nature,v.421, p.37–42. Peck,S.B.,1973,Asystematicrevisionandtheevolutionarybiologyofth e Ptomaphagus( Adelops )beetlesofNorthAmerica(Coleoptera; Leiodidae,Catopinae),withemphasisoncave-inhabitingspecies: BulletinoftheMuseumofComparativeZoology,v.145,p.29–162. Peck,S.B.,1988,AreviewofthecavefaunaofCanada,andthe compositionandecologyoftheinvertebratefaunaofcavesandmines inOntario:CanadianJournalofZoology,v.66,p.1197–1213. Peck,S.B.,1998,Asummaryofdiversityanddistributionoftheobligate cave-inhabitingfaunasoftheUnitedStatesandCanada:Bulletinof theNationalSpeleologicalSociety,v.60,p.18–26. Peck,S.B.,andLewis,J.J.,1978,Zoogeographyandevolutionofthe subterraneaninvertebratefaunasofIllinoisandsoutheasternMissouri:TheNationalSpeleologicalSocietyBulletin,v.40,p.39–63. Pielou,E.C.,1991,AftertheIceAge:ThereturnoflifetoglaciatedNorth America,Chicago,UniversityofChicagoPress,366p. Platnick,N.I.,2008,TheWorldSpiderCatalog,Version8.5,American MuseumofNaturalHistory[accessedMarch2008]http://research. amnh.org/entomology/saders/catalog Posada,D.,andBuckley,T.R.,2004,Modelselectionandmodel averaginginphylogenetics:AdvantagesofAkaikeinformation criterionandBayesianapproachesoverlikelihoodratiotests: SystematicBiology,v.53,p.793–808. Posada,D.,andCrandall,K.A.,1998,MODELTEST:Testingthemodel ofDNAsubstitution:Bioinformatics,v.14,p.817–818. Prest,V.K.,Grant,D.R.,andRampton,V.N.,1967,GlacialmapofCanada: GeologicalSurveyofCanada.Map1253A:(Scale1:5000000). Proszynski,J.,1968,Revisionofthespidergenus Sitticus Simon1901 (Araneida,Salticidae),I.The terebratus group:AnnalesZoologici (Warszawa),v.26,p.391–407. Reeves,W.K.,2000,InvertebratecavernicolesoftheGreatSmoky MountainsNationalPark,USA:JournaloftheElishaMitchell ScientificSociety,v.116,p.334–343. Roberts,M.J.,1987,TheSpidersofGreatBritainandIreland,VolumeII, DescriptionofSpecies-Linyphiidae;ChecklistoftheBritishSpiders, Colchester,England,HarleyBooksPublishers,204p. Ronquist,F.,andHuelsenbeck,J.P.,2003,MRBAYES3:Bayesian phylogeneticinferenceundermixedmodels:Bioinformatics,v.19, p.1572–1574. Ruzicka,V.,andKlimes,L.,2005,Spider(Araneae)communitiesofscree slopesintheCzechRepublic:JournalofArachnology,v.33,p.280–189. Saaristo,M.I.,1972,Redelimitationofthegenus Oreonetides Strand,1901 (Araneae,Linyphiidae)basedonananalysisofthegenitalorgans: AnnalesZoologiciFennici,v.9,p.69–74. Saaristo,M.I.,andKoponen,S.,1998,AreviewofnorthernCanadian spidersofthegenus Agyneta (Araneae:Linyphiidae),withdescriptions oftwonewspecies:CanadianJournalofZoology,v.76,p.566–583. Schikora,H.-B.,1995,Intraspecificvariationintaxonomiccharacters ,and notesondistributionandhabitatsof Meionetamossica Schikoraand M.saxatilis (Blackwall),twocloselyrelatedspidersfromnorthland andcentralEurope(Araneae:Linyphiidae):BulletinoftheBritish ArachnologicalSociety,v.10,p.65–74. Schwert,D.P.,1992,Faunaltransitionsinresponsetoaniceage:thelate WisconsinanrecordofColeopterainthenorth-centralUnitedStates: TheColeopteristsBulletin,v.46,p.68–94. Scudder,G.G.E.,1979,PresentpatternsinthefaunaandfloraofCanada, in Danks,H.V.,ed.,Canadaanditsinsectfauna,Memoirsofthe EntomologicalSocietyofCanada,v.108,p.87–179. Skelley,P.E.,andKovarik,P.W.,2001,Insectsurveysinthesoutheast: investigatingarelictualentomofauna:FloridaEntomologist,v.84, p.552–555. Swofford,D.L.,2002,PAUP*v.4.0b10PPC:Phylogeneticanalysisusing Parsimony,Sunderland,MA,SinauerAssociates. vanHelsdingen,P.J.,1981,TheNearcticspeciesof Oreonetides (Araneae, Linyphiidae):BulletinoftheAmericanMuseumofNaturalHistory, v.170,p.229–241. Vink,C.J.,Thomas,S.M.,Paquin,P.,Hayashi,C.Y.,andHedin,M., 2005,TheeffectofpreservativesandtemperaturesonarachnidDNA: InvertebrateSystematics,v.19,p.99–104. Vink,C.J.,Sirvid,P.J.,Malumbres-Olarte,J.,Griffiths,J.M.,Paquin ,P., andPaterson,A.M.,2008,Speciesstatusandconservationissuesof NewZealand’sendemic Latrodectus spiderspecies(Theridiidae: Araneae):InvertebrateSystematics,v.22,p.589–604. Yagimuma,T.,1972,SpidersfromtuffcavesinYamagataPrefecture, Japan:FacultyofLettersReview,OtemonGakuinUniversity,v.6, p.81–94. P.P AQUIN ,N.D UPE RRE ,D.J.B UCKLE AND J.J.L EWIS JournalofCaveandKarstStudies, April2009 N 15



PAGE 1

CAVESASSEALEVELANDUPLIFTINDICATORS, KANGAROOISLAND,SOUTHAUSTRALIA J OHN E.M YLROIEAND J OAN R.M YLROIE DepartmentofGeosciences,MississippiStateUniversity,MississippiS tate,MS39762USA,mylroie@geosci.msstate.edu Abstract: FlankmargincaveshavebeenobservedinQuaternaryBridgewater FormationeolianitesonKangarooIsland,SouthAustralia.Horizonsoffl ankmargin cavedevelopmentat25m,30m,and35melevationdemonstratetectonicupli ftoftens ofmetersduringtheQuaternary,asthecaveelevationsarehigherthanany reported Quaternaryglacioeustaticsea-levelhighstand.Distinctcavehorizons indicatethat episodicupliftwaspossible.Wave-cutnotchesatHansonBay,at30to35me levation, alsosupporttheinterpretationfromcavesthatrelativesealevelwasonc eatthe 30-melevationrange.AdmiralsArch,previouslypresentedasformingsolelyb ywaveerosion, isaflankmargincavebreachedandmodifiedbywaveerosion.PointEllenco ntainsa LatePliocenesubtidalcarbonateunitthatformedwithinthereachofwave base,was upliftedandcliffedbywaveprocesses,andthenwaskarstifiedbeforebei ngburiedby QuaternaryBridgewaterFormationeolianites.Apossibleflankmarginca vedeveloped atPointEllenat3mabovemodernsealevelisconsistentwithearlierinter pretationsof notchingofthenearbycoastatasimilarelevationduringthelastintergl acialsea-level highstand(MIS5e);andtherefore,notectonicupliftinthelast120ka.In contrast,the tafoniofRemarkableRockspresentacautionarynoteonevidenceofcavewa ll morphologicalcharacteristicsasproofofdissolutionalorigin. I NTRODUCTION KangarooIslandislocated16kmsouthwestacrossthe BackstairsPassagefromCapeJervisontheFleurieu Peninsula,SouthAustralia(Fig.1).Theislandisa rectangleroughly145kmeasttowest,and60kmnorth tosouthwithanareaof4,350km 2 andacoastlinelength of457km(ShortandFotheringham,1986).Kangaroo Islandisageologicallydiverseenvironmentwithrocks presentfromtheProterozoicthroughthePaleozoic, Mesozoic,andCainozoic(BelperioandFlint,1999).The islandisageologicextensionoftheFleurieuPeninsulaon theAustralianmainland(BelperioandFlint,1999;James andClark,2002).AlthoughisolatedPaleozoicoutcropsof carbonaterockexistonKangarooIsland,thedominant carbonateunitsareCainozoic(JamesandClark,2002). Especiallyprevalent,primarilyalongthesouthernand westerncoasts,areeoliancalcarenitesofLatePliocene throughHoloceneage(Ludbrook,1983;Shortand Fotheringham,1986).TheCambrianKanmantooGroup metasedimentsformthebasementoftheisland(Jamesand Clark,2002),andcommonlytheeolianitesrestdirectly uponthemalongthesouthernandwesternshoreof KangarooIsland.Cambro-OrdoviciangranitesalsooutcropatthesouthwestendofKangarooIsland(Jamesand Clark,2002)andeolianitesoccasionallyoverliethose exposures. Theeoliancalcarenitesareassignedbymostauthorsto theBridgewaterFormation,thenamegiventoeolian calcarenitesacrossSouthAustraliaandVictoria(Drexel andPreiss,1995).Theseeoliancalcarenitesrepresent depositionalepisodesassociatedwithglacioeustaticQuaternarysea-levelfluctuationsandconsistofatleast16 separateeventsontheAustralianmainland(Drexeland Preiss,1995).OnKangarooIsland,theeoliandepositional eventsarethoughtbysomeauthorstorepresentonlyfour glacialeventsbecausethoseauthorsrecognizeonlythe traditional,continental-basedfourPleistoceneglaciations (TwidaleandBourne,2002),whileotherauthorsindicate thatmorethan16sea-levelhighstandsoccurredinthe QuaternaryonKangarooIsland(ShortandFotheringham, 1986).Mostauthors(e.g.,DrexelandPreiss,1995) considertheBridgewaterFormationtobePleistocenein age.However,Ludbrook(1983)reportsthateolian calcarenitesoftheBridgewaterFormationinterfingerwith LatePliocenePointEllenFormationsubtidalfaciesat PointReynolds,eastofPointEllen.Thisobservationraises thepossibilitythattheinitialdepositionoftheBridgewater FormationeoliancalcarenitespredatesthePliocene-PleistoceneboundaryonKangarooIsland.TheBridgewater Formationeoliancalcarenitesandthecavesandkarst featuresdevelopedinthemarethefocusofthisreconnaissancestudy. KangarooIslandwasvisitedinSeptember2006forfive daysasareconnaissanceexpeditionwiththepurposeof comparingcavedevelopmentinthecoastaleoliancalcarenitesonthatislandwiththosefoundinsimilar environmentsintheBahamianArchipelago,halftheworld away.Theobservationsreportedherearenecessarilybrief andlackdetailsuchascavesurveys.However,thesebrief observationsestablishimportantinformationregardingthe natureofcavedevelopmentonKangarooIslandand J.E.MylroieandJ.R.Mylroie–Cavesassealevelandupliftindicators,Ka ngarooIsland,SouthAustralia. JournalofCaveandKarst Studies, v.71,no.1,p.32–47. 32 N JournalofCaveandKarstStudies, April2009

PAGE 2

importantsealevelandtectonicinterpretationsthatcanbe generalizedfromthoseobservations. DecadesofworkintheBahamas(Mylroie,1980;Vogel etal.,1990;Mylroieetal.,1995;Rothetal.,2006;Mylroie andMylroie,2007)developedanextensivedatabaseon cavedevelopmentineoliancalcarenites.Thoseinvestigationsledtothedevelopmentoftheflankmargincave modeltoexplainrapidcavedevelopmentinBahamian eolianitesasaresultofmixing-zonedissolution.These cavesdevelopinthedistalmarginofthefresh-waterlens, undertheflankoftheenclosinglandmass.Atthislocation, thefavorabledissolutionalhorizonsatthetopofthelens (vadoseandphreaticfresh-watermixing)andatthebaseof thelens(sea-waterandfresh-watermixing)aresuperimposed(MylroieandCarew,1990,1995).Theselens boundariesarealsodensityinterfaces,whichcollect organicmaterial.OxidationoftheseorganicscreatesCO 2 todrivefurtherdissolution,andwithextremeorganic loading,anoxicconditionsdevelopandH 2 S-mediated dissolutioncanoccur(Bottrelletal.,1993).Thedecrease inlenscross-sectionalareaatthelensmarginincreases water-flowvelocity,suchthatreactantsandproductsmove swiftlythroughthesystem(RaeisiandMylroie,1995).The modelwasexpandedtonon-eolianitecarbonateislandsof increasingcomplexity,suchasIsladeMona,PuertoRico (Franketal.,1998)andGuamintheMarianas(Mylroieet al.,2001).Oneoftheprominentaspectsofflankmargin cavedevelopmentisthatthecavescutacrosscarbonate facieswithoutregardtograinsize,porosity,orprimary structuresintherock(MylroieandCarew,1995);freshwaterlensposition,andhencesealevel,istheprimary factorincaveposition.Theflankmargincavemodelhas evolvedintotheCarbonateIslandKarstModel,orCIKM, whichtakesintoaccountmixingdissolution,glacioeustasy, tectonics,carbonate/noncarbonateislandrelationships, anddiageneticmaturity(Jensonetal.,2006;Mylroieand Mylroie,2007).Flankmargincavesformasisolated chambersorgroupsofchambers(i.e.,theydevelopas mixingchamberswithoutentrances).Asimilarrapid reconnaissanceinAugustof2006atRottnestIsland, Australia,resultedinanumberofnewandimportant observationsintheQuaternaryeolianitesandtheflank margincavesthere(MylroieandMylroie,2009). Flankmargincavedevelopmentiscontrolledbythe positionofthefresh-waterlens,whichinturn,istiedto sea-levelposition.Otherclassicsea-levelindicatorson carbonatecoasts,suchasintertidalandsubtidaldeposits, andcoastalnotching,remainonthelandsurfaceafterlocal upliftand/oreustaticsea-levelfall.Flankmargincavesalso occurinacoastalsetting,butinsidethelandmass.This differenceinplacementmeansthatassurficialerosion stripsawaysurfacedepositsandlandforms,flankmargin cavesarestillpresent.Assurfaceerosioncontinues,the flankmargincavesarebreachedandbecomevisibletothe surfaceobserver.Asaresult,flankmargincavesarenot onlysea-levelindicators,butalsoindicatorsofratesof surfacedenudation(Figs.2and3). IntheBahamas,breachedflankmargincaveswereonce identifiedasfossilbioerosionnotchesbeforebeing recognizedaserosionally-exposedsubsurfacefeatures (MylroieandCarew,1991).TheBahamasaretectonically stable,subsidingat1to2mper100ka(Carewand Mylroie,1995;McNeill,2005),therefore,anysea-level indicatorsfoundinthesubaerialenvironmenttodayarethe resultofeustasy.AsthesurficialrocksoftheBahamasare lessthan800,000yearsold(CarewandMylroie,1997),this eustaticsea-levelchangeisaresultofglacial-deglacial cycles.TheBahamiansituationissomewhatsimplified,as Figure1.MapofKangarooIslandshowingitspositionofftheAustralianco ast,andlocationsdiscussedinthetext. J.E.M YLROIEAND J.R.M YLROIE JournalofCaveandKarstStudies, April2009 N 33

PAGE 3

sealevelorsubtidalindicatorsabovemodernsealevel today,suchasfossilreefs,abandonedcoastalnotches, breachedflankmargincaves,orherringbonecrossbedding, mustbetheresultofaglacioeustaticsea-levelhighstand abovemodernlevels.Thathighstandmustalsohavebeen recentenoughthatitsdepositshaveremainedabove modernsealeveldespitetheisostaticsubsidencerateof1 to2mper100ka.Basedontheseconstraints,theonlysealevelhighstandbelievedtobepervasivelyrecordedinthe Bahamasisthatofthelastinterglacial,marineisotope substage5e,orMIS5e(CarewandMylroie,1997).Inthe AtlanticBasin,thathighstandreached 6mandlasted from131kato119ka(Chenetal.,1991),a12,000year timewindowinwhichtocreateallthepre-modernsea-level indicatorscurrentlyobservedabovesealevelinthe Bahamas,includingtheflankmargincaves. KangarooIslandisinanenvironmentthathasbeen upliftedduringtheQuaternary(ShortandFotheringham, 1986;JamesandClark,2002;TwidaleandBourne,2002); thedebatehasbeenovertheamountofuplift.Inhigh-relief coastalsettings,suchasonKangarooIsland,cliffretreat undercurrentsea-levelconditionshasremovedmanyof thetypicalsurfaceindicatorsofpast,highersealevels (Fig.2),anduplifthasdisplacedthemfromtheelevation oftheirinitialformation(Fig.3).Numerouscavesare foundintheeolianitecliffsofthesouthernandwestern coastsofKangarooIsland,andtheirlocationispotentially arepresentationofapastfresh-waterlensposition,and hence,apastsea-levelposition.Thispaperreportsonhow useofcaves,especiallyflankmargincaves,canhelprefine ourunderstandingofQuaternaryprocessesonKangaroo Island. M ETHODOLOGY Conceptually,themethodologyissimple;walkthe eolianiteoutcropsandrecordthelocationandelevationof flankmargincaves.Ideally,allcavesidentifiedare representativeofapaleosea-levelposition,butcareis neededbecausethecavescanbeoftwomaintypes: dissolutionalcaves;andpseudokarstcavesproducedby processesotherthandissolution.Adistinctionmustbe madebetweenthetwomaincavetypes.Ifthecaveis determinedtobeadissolutionalcave,thenthenatureof thatdissolutionmustbedeterminedtoestablishwhether thecaveformedinthedistalmarginofafresh-waterlensas aflankmargincave. Cliffedcoastalsettingsineoliancalcarenitescommonly havetwomaintypesofpseudokarstcaves:seacaves,and tafonicaves.Seacavesaretheresultofwaveenergyand wavetransporteddebrismechanicallyerodingahollowor caveintoabedrockoutcrop(Waterstrat,2007and referencestherein).Suchcavesformfromtheoutside inwardasthewavescontinuetheirwork.Seacavesarea sea-levelindicator,butbecausetheyareproducedfromthe surfaceinward,theyremainneartheactiveerosivefaceof thelandsurfaceandarevulnerabletolaterremovalby scarpretreat.Tafonicavesarevoidsandhollowsproduced bysubaerialweatheringprocessesthatselectivelyremove rockgrainsfromaspecificareainanoutcrop.The processesinvolvewind,crystalwedgingofevaporite minerals,wettinganddrying,andavarietyofother activities(Owen,2007andreferencestherein).Similarto seacaves,tafonicavesformfromtheoutsideinwardand arevulnerabletoremovalbyscarpretreat.Unlikesea caves,however,theirpositiononaclifffaceisrandomand notrelatedtoaspecificpastsea-levelposition.Recent workinBahamianeoliancalcareniteshasresultedinthe developmentofcriteriatoallowthedifferentiationof breachedflankmargincavesandseacaves(Waterstrat, 2007;Mylroieetal.,2008a),aswellasthedifferentiationof seacavesandtafonicavesfromeachotherandfrom Figure2.WestcoastofKangarooIsland,northofSnake LagoonandRockyRiver.A–Overallscene,withlightcolouredBridgewaterFormationeolianitesoverlyingdarker KanmantooGrouprocks.Thewhiteboxdenotesthearea presentedin(B).B–IsolatedBridgewaterFormationeolian calcareniteoutcrops(light-colouredpatches)onKanmantoo Grouprocks,demonstratingthataonce-continuouseolian calcareniteunithasbeenstrippedawaybyHolocene coastalerosion. C AVESASSEALEVELANDUPLIFTINDICATORS ,K ANGAROO I SLAND ,S OUTH A USTRALIA 34 N JournalofCaveandKarstStudies, April2009

PAGE 4

breachedflankmargincaves(Owen,2007;Mylroieetal., 2008a). Oneoftheprimarycriteriatodifferentiatethecave typesisinspectionoftheinteriorrocksurfacesforthe uniquecurvilinearshapesassociatedwithcarbonate dissolutionalsurfaces.Suchinspectioncanbeproblematic ifaflankmargincavehasbeenbreachedwhilestillinthe surfzonebecausewave-dominatedmechanicalprocesses mayhavescoured,obscured,andoverprintedtheoriginal dissolutionalsurfaces.Theuseofmetrics,suchastheratio ofcaveareaovercavewallperimeter,andtheratioofcave entrancewidthovermaximumcavewidthreliablydifferentiateflankmargincaves,seacaves,andtafonicavesfrom eachotherintheBahamas(Waterstrat,2007,Owen,2007, Mylroieetal.,2008a).Thesemetricshavebeendemonstratedtoworkforseacavesandflankmargincavesin coastalPuertoRico(Lace,2008).Thedissolutionalorigin oftheflankmargincavecreatesaverycomplexperimeter, whereastheseacaveandtafonicavetendtohavesmooth perimeters.Thisdifferencecreatesdistinctareatoperimeterratiosforthecavetypes.Becausetheseacavesand tafonicavesarecreatedbyerosiveforcesworkingfromthe outsideinward,theycommonlyhaveacaveentrancewidth overmaximumcavewidthratioofnearone(theentranceis thewidestpartofthecave).Flankmargincavesinitiateas dissolutionalchambersthatarelaterbreachedbysurface erosion.Theentrancesarecommonlysmallormedium sized,andthecaveentrancewidthtomaximumcavewidth ratioismuchlessthanone.Whenaflankmargincave displaysarationearone,thatresultisanindicationthat thecavehasbeenbreachedanddenudedtotheextentthat itisapproximatelyhalformoredestroyed(Staffordetal., 2005,theirFig.11).Finally,sealedcavechamberswith humiditynear100%precipitatedense,crystallinecalcite speleothemssuchasstalactites,stalagmites,flowstone,and otherformsasaresultofCO 2 diffusionfromdripwater intothecaveatmosphere.Inopenairenvironments,such calciteprecipitationisnotpossible,asevaporation competeswithCO 2 diffusiontocontroltheCaCO 3 chemistry,andcrumblytuffaceousdepositsform.Taboros i etal.(2006)provideexamplesandcriteriatodetermine whichcalciteprecipitatesfoundinopencavechambers Figure3.ViewsofflankmargincaveentrancesinthelowerRockyRiverandS nakeLagoonarea,developedinBridgewater Formationeolianites.A–Valleynorthwall,showingawell-developedban dofcavesat 30-m-elevation,andothercavesat 25-m-elevation.B–Valleysouthwall,showingthreehorizonsofflankmar gincavedevelopment,at 25-, 30-and 35-melevation.Aswiththevalleynorthwall,thebestdevelopmentisat 30m. J.E.M YLROIEAND J.R.M YLROIE JournalofCaveandKarstStudies, April2009 N 35

PAGE 5

formedinsideasealedcaveandwhicharetheresultof open-airprecipitation.Seacavesandtafonicavesoriginate asopenvoidsanddonotcommonlycontaindensecalcite speleothems.Cavespeleothemsarecommonlymodifiedby surficialexposure,degradingintoamoretuffaceous character,whichcancomplicateinterpretations(Taboros i etal.,2006) Onceithasbeenestablishedthatagivencaveisthe resultofdissolution,thatitisakarstcave,thenthetypeof karstcavemustbedetermined.Karstcavesfallintotwo maingroups(Palmer,1991):epigeniccavesthatusually developasconduitsystemswithrapidturbulentflow,that aredirectlycoupledtothesurfacehydrologyand hypogeniccavesthatdevelopinalaminarorslowflow system,decoupledfromthesurfacehydrology.Flank margincaves,bythisdescription,aresimilartohypogenic cavesbecausetheyaremixingchambersanddonothave rapidturbulentconduitflow.Epigeniccavesformedby conduitflowcommonlycontainwallmarkingsthatreveal theflowtobefastandturbulent.Dissolutionalfeatures calledscallopsareespeciallyeffectiveindicatorsof turbulentflowdirectionandvelocity(Curl,1966,1974). Turbulentconduitflowcommonlyleavesbehinddiagnostic sedimentarydeposits,anddistinctpassageshapes,suchas vadosecanyons.Othercavefeatures,suchassolutionpipes andvadoseshafts,canalsobepresent.Theseotherfeatures representpartofthecouplingsystemthatconnectsthe epikarsttothewatertable,andforepigeniccaves,they couplethecavedirectlytothesurfacehydrology,butfor hypogenicandflankmargincaves,theyrepresentrandom intersections.TheKellyHillCavesonKangarooIsland (Hill,1984)areanexampleofconduitflowofwater throughaneolianiteridge,perchedonunderlyinginsoluble rocks.Subsequentcollapsehascreatedamazeofchambers andpassagewayswithinandabovethecollapsematerial, andthetruenatureoftheoriginaldissolutionalpassagesis obscuredinmostofthisepigeniccave. Flankmargincavesformentirelyasphreaticfeatures underlaminarorslowflowconditions.Theirwallsculpture consistsofcuspatepockets,bedrockcolumnsandspans commonlyofdelicateconfiguration,andcurvilinearforms thatcutacrossprimaryandsecondarybedrockfeatures (MylroieandCarew,1990,1995).Flankmargincaves containnoevidenceofrapidturbulentflow,eitheraswall sculptureorassedimentarydeposits.Theirdevelopmentin thethin,distalmarginofthefresh-waterlensresultsinlow, widechambersthatintersectrandomlybecauseeach chamberwasaninitiationpointformixingdissolution (Roth,2004;Labourdette,etal.,2007).Thispassage patternisespeciallytrueofflankmargincavesdeveloped indiageneticallyimmature,oreogenetic,carbonaterocks suchasQuaternaryeoliancalcarenites,whereprimary porositycanbeasmuchas30%andthewatersareableto mixacrossabroadvolumeoftherockmass(Vacherand Mylroie,2002).Flankmargincavesizeiscontrolled primarilybythedurationoftimethatthefresh-waterlens, andhencesealevel,wasinastableposition(Mylroieand Mylroie,2007;Mylroieetal.,2008b). ThesouthernandwesterncoastsofKangarooIsland areareasofhighwindandoceanenergy(Shortand Fotheringham,1986).Thecoastaloutcropsofeolianiteare cliffedandhaveretreatedlandward.Anideaofhowmuch retreathasoccurredcanbeseenonthewestcoastofthe islandwheretheeoliancalcarenitesrestunconformablyon theKanmantooGroupbasementrocks.Remnantsof eoliancalcarenitecanbeseenthataretensofmetersaway fromthecurrenteoliancalcarenitescarp,indicating significantHoloceneerosionalremovalofeoliancalcarenitematerial(Fig.2).Suchlarge-scaleerosionnotonly stripsoffsurfacefeaturessuchasintertidaldeposits,but alsoremovesseacaves,tafoni,andevenflankmargin caves,leavingnopastsea-levelrecord.Toobtaina preservedeoliancalacarenitesectionrequiresinvestigating embaymentsandstreamvalleysthatareprotectedfrom directmarineassault,butwhichwouldhaveheldafreshwaterlensincontactwithseawateratapast,highersealevelposition.Forthesereasons,RockyRiver,reachingthe coastfromSnakeLagoontoMaupertiusBay(Figs.1,3, and4),wasselectedastheprimefieldinvestigationlocality tosearchforflankmargincavesandevidenceofpastsealevelhighstands.CapeduCouedicwasalsoinvestigated becauseoffshoreislandsprovidedsomewaveprotection (Fig.5). Eachlocationwasexaminedbypreliminaryreconnaissanceandrevealeditsownsetofobservations.Follow-up quantitativemappingandsubsequentmapinterpretation, asdescribedintheMethodssection,wasnotdonedueto lackoftimeatthefieldlocalities.Asaresult,the interpretationsmadeastotheoriginofthesecaveswere basedonphysicalconfigurationandappearanceonly. Eachofthelocalitiesbelowispresentedasanobservation setandeachlocalityisthenusedtogenerateasetof preliminaryinterpretations.Theoveralloutcomeofthe studyisthenreviewedintheSummarysection. R ESULTSAND D ISCUSSION S NAKE L AGOON TheFlindersChaseNationalParktrailalongRocky RiverfromSnakeLagoonwasusedastheaccessrouteto thewestcoast(Fig.1).Atthelocationwherethewooden footbridgecrossesoverRockyRiver,KanmantooGroup rocksarevisibleinthestreambed,andcaveopeningscan beseenbacktotheeast,onthenorthbank,highuponthe eolianitevalleywall.Thesecaveswerenotdirectlyvisited, buthavetheappearanceofsimilarfeaturesintheBahamas thatareflankmargincaves.Continuingwestdownstream, morecaveopeningsarevisiblehighonthenorthbank. Whiletheyhavetheappearancefromadistanceofflank margincaves,theywerenotvisitedandtheirorigincannot beconfirmed.AsRockyRiverandthetrailapproachthe coast,numerouscaveopeningsappearinthecliffsonboth C AVESASSEALEVELANDUPLIFTINDICATORS ,K ANGAROO I SLAND ,S OUTH A USTRALIA 36 N JournalofCaveandKarstStudies, April2009

PAGE 6

sidesofthestream(Figs.3and4).Thosecavestothenorth werenotvisitedbecausetheywereinaverticalwall (Figs.3Aand4A),buttheopeningsonthesouthbank wereaccessibleandwerethoroughlyinvestigated(Figs.3B, 4B,4C,and4D). Thesinglemostobviousaspectofthecavesistheir developmentasaseriesofchambersthatcommonly connectinternally.Thisobservationisknownasbeads onastring(Mylroieetal.,2001)andreflectsthedegreeto whichindividualflankmargincavechambersdidordid notintersectastheygrewbymixingdissolutioninthe distalmarginofthefresh-waterlens.Theflankmargin dissolutionaloriginofthecavesisdemonstratedbypassage shapeandconfiguration,abundantcavespeleothems,and dissolutionalwallmorphologies(Fig.4).Thecavesonthe northsideofthestreamchannelarefoundattwoprimary horizons,atapproximately25mand30melevation (Fig.3A).Thehillonthesouthsideishigher,andcontains evidenceofthreecavehorizonsatapproximately25m, 30m,and35m(Fig.3B).Thetwohorizonsonthenorth sideappeartocorrelatewiththetwolowerhorizons(at 25mand30m)onthesouthside.Ateachcavehorizon, thecavesextendlaterallyoveradistanceofupto100m. Thecavesarenotverydeep,penetratingintothehillside generallylessthan10m.Becausethecaveshavebeen breachedbyscarpretreat,theoriginalvoidshaddimensions,perpendiculartothehillside,ofover10m.The largestandmostcontinuousbandofcaves,oneachsideof Figure4.FlankmargincavesatSnakeLagoon.A–Caveentranceat 25-m-elevationinthevalleynorthwall,showing abundantflowstone,stalactiteandstalagmitedevelopment.B–Caveat 25-m-elevationinthevalleysouthwall,showing interconnectingchambersandstalactites.C–Lookingnorthwestfromafl ankmargincaveinthevalleysouthwall,showingthe oceanandKanmantooGroupbasementrocksthatunderlietheBridgewaterFo rmationeolianites.D–Seriesoferodedflank margincavechambersonthevalleysouthwall,showingsmoothphreaticdis solutionalsurfacesandsecondary vadosestalactites. J.E.M YLROIEAND J.R.M YLROIE JournalofCaveandKarstStudies, April2009 N 37

PAGE 7

thevalley,istheoneat30m,indicatingperhapsalonger sea-levelstillstandatthathorizonthanoccurredatthe 25mor35mhorizons. TheSnakeLagooncavesfitallthedirectobservational criteriathatidentifythemasflankmargincaves.Assuch, thecavesrepresentpastsea-levelpositions.Allthecaves describedareaboveanypastQuaternaryglacioeustatic sea-levelhighstand.Therefore,upliftofKangarooIslandis requiredtohaveoccurredtoplacethecavesattheircurrent positionwithrespecttomodernsealevel.RecordsofsealevelhighstandsonKangarooIslandabove10mare regardedasequivocal(TwidaleandBourne,2002). However,Bauer(1961)indicatedthatamarineerosion terraceat100to110feet(30.5to33.5m)wasthemost significantofthefiveterracesherecognizedat20to25feet (6to7.6m)andhigheronKangarooIsland.Thecave observationspresentedheredemonstratearecordofsea levelwellabove6m,andatleastthreeclosely-spaced highstandsarerecorded.Thedurationofthehighstands canbe,inpart,determinedbyhowlargethecavesare.Cliff retreatsincetheirformationhasobviouslydecreasedtheir size,butnonetheless,theyaresmallerthanmanyflank margincavesintheBahamas,whichhad12,000yearsto form.Thedevelopmentofthelargestandmostcontinuous cavesat 30-m-elevationagreeswellwithBauerÂ’s(1961) best-developedterraceatthatelevation.Duringglacioeustasy,theBahamianrecorddemonstratesthatonlywhensea levelisturningaroundfromalowstandorahighstandisit stablelongenoughtocreatelargeflankmargincaves (MylroieandMylroie,2007).Ifupliftisalsoinvolved,then thetimeofsea-levelstabilitywillbeevenless.TheNew Zealandflankmargincaverecord,inatectonicallyactive environment,providestimelimitsonlensstability(Mylroie etal.,2008b).TheflankmargincavesatSnakeLagoon indicatethatuplifthasdefinitelyoccurred,butithasbeen slowenough,andepisodicenough,suchthatglacioeustatic stillstandscanstillleaveaflankmargincavesignature. Giventhatthelastinterglacial(oxygenisotopesubstage5e) was6mhigherthanatpresent,suchaeustaticsea-level elevationvalueshouldbeconsideredwheninvestigatingthe elevationsofthecavesfoundatSnakeLagoontoday.In otherwords,thecavehorizonat 30-m-elevationcould indicateanupliftofonly24m,butthatupliftwouldhave hadtooccurinthelast120ka,andotherevidenceatPoint Ellen(below),suggestsstabilityforthelast120ka.The clusterofcavesbetween 25mand 35mmayrepresent twouplifteventsonasingleglacioeustaticsea-level highstand,ornoupliftepisodeswhilethreedifferent glacioeustaticsea-levelhighstandsoccurred,oracombinationofthetwo. Figure5.OverviewofCapedeCouedic,theCasuarinaIslets,andAdmiralsA rch,KangarooIsland.Thetwoislandsinthe distancearetheCasuarinaIslets,alsoknownasTheBrothers.Theeastope ningofAdmiralsArchislabeledintheforeground. TheblackverticalarrowinthebackgroundpointstothecaveshowninFigur e6A.LightcoloredBridgewaterFormation eolianitesoverliedarkKanmantooGroupbasementrocks. C AVESASSEALEVELANDUPLIFTINDICATORS ,K ANGAROO I SLAND ,S OUTH A USTRALIA 38 N JournalofCaveandKarstStudies, April2009

PAGE 8

C APEDU C OUEDIC Thislocation(Fig.1)isfamousforAdmiralsArch,a largevoidpenetratingarockypointcomposedofBridgewaterFormationeoliancalcarenitesoverlyingKanmantoo Groupbasementrocks(Fig.5).Thesteepeolianitecliffsto theeastcontainanumberofbreachedflankmargincaves (Fig.6).Unfortunately,FlindersChaseNationalPark accessregulationspreventeddirectinvestigation.Visual examinationdemonstratesthatthesefeatureshavedissolutionalwallmorphologies,calcitespeleoethems,andalign horizontallywithothercavesonthesamecliffface. AdmiralsArchwasalsonotavailableforindependent directinvestigation,butpublicaccessplacestheobserver whereclearviewscanbehadofthecaveinterior(Fig.7). ThepublicdisplayonthetourpathpresentsAdmirals Archasaproductsolelyofwaveerosion.However,visual examinationofthenorthwallofthearchrevealsthatithas aseriesofphreaticdissolutionpockets(Fig.7C)andthat theoriginalfloorofthearchwashorizontalanddeveloped inlimestoneabovethedippingcontactwiththeunderlying KanmantooGroupbasement(Figs.7Aand7D).Thecave hasabundantcalcitespeleothems. AdmiralsArchappearstobeabreachedflankmargin cave,whichhasbeenmodifiedbywaveactiononthe current(andperhapslastinterglacial)sea-levelhighstand(s).Thephreaticdissolutionsurfacesandabundant Figure6.ErodedflankmargincavesintheCapedeCouedicarea.A–Caveentr anceattheeolianite-basementcontact, showingstalactitesandstalagmites,onthelandwardofthetwoCasuarina Islets(Figure5).Theprominentstalagmiteinthe caveentranceis 1-m-high.B-Backwallofaflankmargincaveintersectedbycliffretreat, inasmallpointjusteastof AdmiralsArch.Peopleattopforscale.Notethecurvilinearshapeoftheca vewall,andthestalactitesandflowstone.C-Large breachedflankmargincaveatthemajorheadlandbetweenAdmiralsArchand RemarkableRocks.NotetheBridgewater FormationcontactwiththeKanmantooGroupbasementrocksnearsealevel, andthemanystalactitesandstalagmitespresent. D–Smallflankmargincave,andassociatedphreaticpocketsincliffwall, betweenAdmiralsArchandthebreachedcaveseen inFigure6B.Thecaveentranceis 3-m-high. J.E.M YLROIEAND J.R.M YLROIE JournalofCaveandKarstStudies, April2009 N 39

PAGE 9

calcitespeleothemsindicateacavethatformedbyphreatic, mixed-waterdissolution,thenwasdrainedsuchthat vadosespeleothemscoulddevelopinasealedcave chamber.Theremnantflatlimestoneflooronthenorth sideofthecaveisanotherindicationofflankmargincave development.Thatfloorhasbeenmostlystrippedawayby modernwaveactionrampinguptheslopingKanmantoo Groupbasementrocks. Observationfromadistanceofthenearerofthetwo CasuarinaIslets(TheBrothers)revealedtwocavesinthe eoliancalcarenitesontheclifffacingAdmiralsArch (Figs.5and6A).Remnantspeleothemscouldbeseen, butlittleofinteriorconfigurationwasobservable.The cavesarequiteclosetotheKanmantooGroupbasement contact.Theyappeartohavephreaticmorphologies,and astheislandsaretoosmalltosupportconduitflow,the mostlikelyinterpretationisthattheyareflankmargin caves. P OINT E LLEN AtPointEllen,onthewesternsideofVivonneBay (Fig.1),asequenceofcarbonateandnon-carbonaterocks isexposed.AsdescribedbyLudbrook(1983),eolian calcarenitesoftheBridgewaterFormationoverliesubtidal carbonatesoftheLatePliocenePointEllenFormation, whichinturnrestunconformablyontheKanmantoo Figure7.ImagesfromAdmiralsArch.A–LookingnorthwestthroughtheArch .NotethesteepdipoftheKanmantooGroup basementrocks,andstalactitesinupperforeground.Recliningseal,1.5 -m-long,inthecenteroftheimageforscale(white arrow;samesealasinCandD).B–LookingwestthroughtheArch,showingnum erousstalactitessilhouettedbythewestern entrance.Thetwistedandgnarlyappearanceofthestalactitesisanoutco meofmodificationbybothevaporationandalgal growth.C–PhreaticpocketsalongthenorthwalloftheArch,formedintheB ridgewaterFormationeolianitesalonga horizontaldatum,justabovetheslopingKanmantooGroupbasementrocks. Seal, 1.5-m-long,inlowerleftforegroundfor scale(whitearrow).D–Survivingsectionoftheoriginalhorizontalfloo roftheArch.ThephreaticpocketsofFigure7Carein shadowaheadandtotherightintheimage.Seallyinginbackgroundforscal e(whitearrow). C AVESASSEALEVELANDUPLIFTINDICATORS ,K ANGAROO I SLAND ,S OUTH A USTRALIA 40 N JournalofCaveandKarstStudies, April2009

PAGE 10

Groupbasementrocks(Fig.8).Ourobservationsindicate thattheeoliancalcarenitesdrapeoverthePointEllen Formation,andafossilepikarstwithaterrarossapaleosol separatesthetwounits.Theeoliancalcarenitesextend seawardofthePointEllenFormationandsitdirectlyin contactwithKanmantooGroupbasementrocks(Fig.9A). Theoutcropisverycomplexandcontainsawealthof information.ThecontactofthePointEllenFormation withtheunderlyingKanmantooGroupbasementrocks commonlycontainsroundedclastsofthebasementrocks inthefirst1to2metersofthePointEllenFormation (Figs.9Band9C).Suchevidenceisanindicationofwave baseactivelyerodingthebasementrocksatthetimeof PointEllenFormationdeposition.Thiswave-baseevidence placeslimitsonthedepthofdepositionofthePointEllen Formation.ThetransportoftheseerodedKanmantoo Grouprocksdownasubmarineslopecannotbediscounted,however.ThereportbyLudbrook(1983)thateolian calcarenitesoftheBridgewaterFormationinterfingerwith LatePliocenePointEllenFormationsubtidalfaciesat PointReynolds,eastofPointEllen,isanotherindication thatthePointEllenFormationwasdepositedinrelatively shallowwater. ThecontactofthePointEllenFormationwiththe overlyingBridgewaterFormationisapaleokarst,a fossilizedepikarstwithaterrarossapaleosol.Apaleokarst requiresthatthePointEllenFormationwassubaerially exposedforasubstantialtime.Subsequently,theBridgewaterFormationwasdeposited.Theareawherethe BridgewaterFormationeoliancalcarenitesextendover thePointEllenFormationhasareliefofseveralmeters (Fig.10),andatthispoint,thePointEllenFormationis cliffedandapaleo-talusoccupiesthespacebetweenthe BridgewaterFormationandthePointEllenFormation. ThesettingissuggestivethatthePointEllenFormation wasdepositedinwaterswithinreachofwavebase,and thenasupliftoccurred,thePointEllenmaterialwascliffed bywaveactionasittransitedintothesupratidal environment.Itremainedexposedforaperiodoftime longenoughtodevelopamatureepikarstandterrarossa Figure8.PointEllenoutcrop.A–Panoramaphotoofthemajoroutcrop,look ingnorth.TheforegroundisKanmantooGroup basementrocks,thecaveandlaterallyadjacentrocksarePointEllenForm ationmarinecarbonates,andtheoverlyingrocks areBridgewaterFormationeolianites.Personinwhiteovalforscale.B–O utcropofthePointEllenFormation,showing numerousmolluskshells.Pencilis15-cm-longforscale(arrow).C–Close rviewofthesectionin(A).KanmantooGroup basementrocksintheforeground,gradeupwardintoaKanmantooboulderan drubblefaciesinterfingeredwithPointEllen Formation,whichformsthebackwallofthecave.TheBridgewaterFormatio neolianitesformthecaveroofandtopofthe section.Personinblackovalforscale. J.E.M YLROIEAND J.R.M YLROIE JournalofCaveandKarstStudies, April2009 N 41

PAGE 11

soil.Atalusformedalongtheformerseacliff.Basedon Bahamianexamples,theminimumtimeframeforterrarossapalesoldevelopmentwouldbeinthe50to100ka range(CarewandMylroie,1997).Subsequently,thePoint EllenFormationwasentombedbyBridgewaterFormation eoliancalcarenitesthatoverrodetheunit,overrodethe talusdeposit,andextendedontotheKanmantooGroup basementrocksatasealevellowerthanpresent. Theoutcropalsohasacaveinit(Figs.8Aand8C). Thiscaveiswithinreachofstormwaves.Itisdifficultto determineifthecaveisabreachedflankmargincaveora seacave.Thecavecontainsfloortoceilingcolumnsthat arenotspeleothems,butratherremnantsolutionpipes (Milnesetal.,1983).Becausethecolumnwallsbecame micritizedwhentheywerepartofanactiveepikarst,they arenowstrongerthanthehostrockandweatheroutin relief(Fig.11).SuchfeaturesarecommononcoastalSouth AustraliainBridgewaterFormationeoliancalcarenites. Thesurvivalofthesecolumnswithinacaveallowstheir connectiontotheoriginalepikarstsurfacetobeobserved. Thesolutionpipecolumnstotheeast(rightfacingintothe cave)aremadeupofpaleosolmaterial(Fig.11C),butto thewest(leftfacingintothecave)thematerialisstronglyinduratedPointEllenmaterialthatthesolutionpipehad drilledinto(Fig.11B).Thecorrectinterpretationofthe caveisimportant.Ifitisaseacave,producedasaresultof stormactivityonthiscoast,thenitisaHolocenefeature.If itisaflankmargincave,thenitisatleast125kaoldand formedonthelastinterglacialsea-levelhighstand(MIS 5e).Asbenchesat3melevationfromthelastinterglacial arereportedonthesouthernshoreofKangarooIsland (ShortandFotheringham,1986;TwidealeandBourne, Figure9.A–BridgewaterFormationeolianiteslyingdirectlyondeformed KanmantooGroupbasementrocks.ThePoint EllenFormationismissing.ThelocationiswherethephotographofFigure 8Awastaken,seawardofthemainoutcropby about30m.B.–West(left)aroundthepointshowninthefarleftofFigure8A ,thePointEllenFormationrestsonaplanated benchofKanmantooGroupbasementrocks.Weatheredbouldersandcobbleso fKanmantooGroupbasementrocksarevisible 1to2mabovethecontact.C–PointEllenFormationrocks,withabundantsub tidalfossils,intermixedwith,andoverlying, KanmantooGroupbasementrockspresentasbouldersandcobbles.Pencilis 15-cm-longforscale. C AVESASSEALEVELANDUPLIFTINDICATORS ,K ANGAROO I SLAND ,S OUTH A USTRALIA 42 N JournalofCaveandKarstStudies, April2009

PAGE 12

2002),afresh-waterlenscouldeasilyhaveexistedinthis outcropduringthelastinterglacialandaflankmargincave formed.Ifthesebenches,andthecave,aretrulylast interglacial,thenupliftoverthelast120kahasbeen minimalandtheupliftatSnakeCreekwasatleast30m andoccurredpriortoMIS5e. R EMARKABLE R OCKS ThetafonidevelopedinCambro-Ordoviciangraniteat RemarkableRocks(Figs.1and12)provideacautionary tale.Theinteriorofseveralofthelargertafonihavewall sculpturethatisespeciallycuspateanddimpledandbarea strikingresemblancetodissolutionalwallmorphologyas foundinflankmargincaves(Fig.12CversusFig.12D). Theyalsohavebeenmisidentified.TwidaleandBourne (2002,theirFigure14c)presentedaphotographofthe interiorofatafoniatRemarkableRocks,callingthe wallsurfacemammillated.Thiserrorisaresultofprinting thepictureupsidedown,suchthatlightingandshadows inverttheapparentreliefinthepicture.Thefeatures areclearlycuspateandnotmammillary,asseenin Figures12BandC.Becausesuchcuspatefeaturesare partofthevisualinventoryusedtodefineacavein limestoneasphreaticinorigin,theRemarkableRocks exampleindicatesthatmultiplelinesofevidenceshouldbe usedtoidentifyacaveÂ’sorigin.Areviewoftafoni,their mechanismsofformation,andthetechniquesutilizedto differentiatethemfromdissolutionalcavescanbefoundin Owen(2007). H ANSON B AY TheeastsideofHansonBay(Fig.1)beginsasastretch ofbeach,andgraduallytrendingsoutheastward,becomesa higheolianridgewithseacliffsdowntotheoceanbelow. Highuponthesecliffsareaseriesofplanatednotches (Fig.13A)thatcouldeasilyrepresentsmallwave-eroded platforms.Theplatformshavearubbledepositofrounded clastsinagreymatrix(Fig.13B).Theseareclearlynota paleosollayer,inwhichtheclastswouldbemoreangular andthematrixwouldcarrytheredcolorofaterrarossa paleosol.Suchdeposits,whenfoundintheBahamas, indicateabackbeachorrockplatformrubblefacies (Floreaetal.,2001).Itisthereforelikelythatthesenotches indicateasea-levelhighstandapproximately30mto35m abovemodernsealevel,whichwouldrequiretectonic uplift.Suchasea-levelinterpretationsupportstheobservationsofflankmargincavesatsimilarelevationsatSnake Lagoontothewest. Figure10.OutcropsectionatPointEllen,theleftportionofFigure8A.Ka nmantooGroupbasementrocksatthebottom, passingthrougharubblefaciesintoPointEllenFormationmarinecarbona tes.Apaleo-cliffseparatesthePointEllen Formationlaterallyfromapaleo-talustotheleft(south).BridgewaterF ormationeolianitesoverliethePointEllenFormation andthepaleo-talusunit.Seaward(leftorsouth)ofthislocation,theBri dgewaterFormationeolianitesoverlietheKanmantoo Groupbasementrocksdirectly,asshowninFigure9A. J.E.M YLROIEAND J.R.M YLROIE JournalofCaveandKarstStudies, April2009 N 43

PAGE 13

S UMMARY TheobservationsmadeonKangarooIslandinSeptember2006areadmittedlycursoryandsuperficial.They consistofsimplevisualdescriptionsatthemacroscopic scale,withoutdetailedsitesurveyorrock-sampleanalysis. Ontheotherhand,thesimpleobservationsallownew interpretationstobeofferedthatmayhelpilluminate geologicprocessesontheisland.Noneoftheprevious workerswhointerpretedtheCainozoicgeologyofKangarooIslandutilizedthepotentialdatastoredincavesonthe island. Itisclearthatflankmargincavesarepresenton KangarooIsland,andthispaperisthefirstreportoftheir existencethere.Thepositionsoftheflankmargincavesat SnakeLagoonrevealsea-levelhighstandsofatleastthree elevations: 25m, 30m,and 35m,substantiating earlyclaimsbyBauer(1961)thatTwidaleandBourne (2002)latercalledintoquestion.Theabsenceofflank margincavesfrommanyhigh-energycoastsunderlainby KanmantooGrouprocksverifiesthevulnerabilityofflank margincavestodestructionbypowerfulwave-generated slope-retreatprocesses.Insuchlocales,flankmargincaves arepreservedinembaymentsandsurfacewatercourse incisions,asatSnakeLagoon,orbyoffshorebarriers,asat CapeduCouedic.ObservationsfromHansonBayshow marineerosionfeaturesconsistentwithdevelopment duringoneormoresea-levelhighstandsatapproximately 30mto35m,concurringwithobservationsatSnake Lagoon,whichwouldrequiretectonicuplift. AtCapeduCouedic,anarchinBridgewaterFormation eolianitesrestingonKanmantooGrouprocksispresented tothepublicasbeingtheresultofwaveerosionwithwave energybeingfocusedonthepointbythepresenceofthe Figure11.ThecaveatPointEllen.A–Lookingintothecave,withtheBridge waterFormationeolianitesformingthecave roof,andinfilledsolutionpitsdescendingintothePointEllenFormatio n.B–CloseupofBridgewaterFormationeolianiteson topofPointEllenmarinecarbonates.Thecontactistheupwardconvexline archingthroughthetopportionofthephotograph (longverticalarrow).Pencilis15-cm-longforscale(shorthorizontala rrow).C–VerticalcontactofBridgewaterFormation eolianitesandpaleosolinfillingasolutionpittotheleft,withPointEl lenFormationmarinecarbonatestotheright.Pencilis 15-cm-longforscale. C AVESASSEALEVELANDUPLIFTINDICATORS ,K ANGAROO I SLAND ,S OUTH A USTRALIA 44 N JournalofCaveandKarstStudies, April2009

PAGE 14

islandsoffshore.Theevidencefromtheeolianiteportionof thearchsuggeststhattheoriginalvoidformedasaflank margincaveandwassubsequentlybreachedbywave erosion.Theoffshoreislandsnotonlyactedasafocusing mechanismforwaveenergy,butalsoprovidedabarrier functionthathaspreventedtheentireeolianitesectionat AdmiralsArchfrombeingremovedbywaveerosion. AtPointEllen,acavehasbeenusefulininterpretingthe stratigraphicsectionofeolian,marine,andbasementrock relationships.Thepreservedfossilepikarstandpaleosolat thislocationplaceboundaryconditionsonthetimingof thecarbonate-depositionalevents.Apaleo-talusisdescribedhereforthefirsttime.Thecaveitselfis indeterminateinorigin,butifitisaflankmargincave,it wouldprovideasecondlineofevidencetosuggestthat upliftonKangarooIslandhasbeenminimalforthelast 120ka.RemarkableRocksdemonstratehownon-dissolutionalerosiveforcescanproducesurfacesintafonithat mimiconeoftheclassicindicatorsofflankmargincave development,andassuch,areawarningaboutusingsingle linesofevidencetomakeimportantcaveorigininterpretations. A CKNOWLEDGMENTS TheauthorsthanktheDepartmentofGeosciences, MississippiStateUniversity,forgrantingJohnMylroiea sabbatical,andJoanMylroiealeaveofabsence,andfor providingsupportforthefieldexpedition.MichaelKidd, KellyHillCave,providedhelpfulinformationand guidanceonKangarooIsland.PeterBell,GrantGartrell, KenGrimes,JohnWebb,NickWhiteandSueWhite Figure12.RemarkableRocks.A–TheRemarkableRocks,wheretafonihavede velopedingraniticrocks.B–Classic cavernousweatheringtoproduceacomplicatedtafoni.C–Insideoneofthe largertafoni,withpocketsorcuspserodedintothe ceiling.D–ChamberinSaltPondCave,LongIsland,Bahamas,ineoliancalc arenites,showingthepocketsandcusps consideredasoneofthediagnosticindicatorsofcaveformationbyphreat icdissolution.ComparewithFigure12C.Twopeople inbackground,leftandright,forscale. J.E.M YLROIEAND J.R.M YLROIE JournalofCaveandKarstStudies, April2009 N 45

PAGE 15

providedsignificantlogisticalsupportandimportant scientificinsight. R EFERENCES Bauer,F.H.,1961,Chronicproblemsofterracestudy,SouthAustralia: Zeitschriftfu ¨rGeomorphologie,v.3,p.57–72. Belperio,A.P.,andFlint,R.B.,1999,KangarooIslandbiologicalsurvey : Geomorphologyandgeology, in Robinson,A.C.,andArmstrong, D.M.,eds.,AbiologicalsurveyofKangarooIsland,SouthAustralia: DepartmentofEnvironment,Heritage,andAboriginalAffairs,South Australia,p.19–31. Bottrell,S.H.,Carew,J.L.,andMylroie,J.E.,1993,Bacterialsulphate reductioninflankmarginenvironments:Evidencefromsulphur isotopes, in White,B.,ed.,Proceedingsofthe6thSymposiumonthe GeologyoftheBahamas:PortCharlotte,Florida,BahamianField Station,p.17–21. Carew,J.L.,andMylroie,J.E.,1995,Quaternarytectonicstabilityofth e BahamianArchipelago:Evidencefromfossilcoralreefsandflank margincaves:QuaternaryScienceReviews,v.14,p.144–153. Carew,J.L.,andMylroie,J.E.,1997,GeologyoftheBahamas, in Vacher, H.L.,andQuinn,T.M.,eds.,Geologyandhydrogeologyofcarbonate islands:Amsterdam,Elsevier,DevelopmentsinSedimentology,v.54, p.91–139. Chen,J.H.,Curran,H.A.,White,B.,andWasserburg,G.J.,1991,Precise chronologyofthelastinterglacialperiod: 234 U230 Thdatafromfossil coralreefsintheBahamas:GeologicalSocietyofAmericaBulletin, v.103,p.82–97. Curl,R.L.,1966,Scallopsandflutes:TransactionsoftheCaveResearch GroupofGreatBritain,v.7,p.121–160. Curl,R.L.,1974,Deducingflowvelocityincaveconduitsfromscallops: NationalSpeleologicalSocietyBulletin,v.36,p.1–5. Drexel,J.F.,andPreiss,W.V.,1995,ThegeologyofSouthAustralia, Volume2,ThePhanerozoic:GeologicalSurveyofSouthAustralia, Bulletin54,347p. Florea,L.,Mylroie,J.,andCarew,J.,2001,Karstgeneticmodelforthe FrenchBayBrecciadeposits,SanSalvador,Bahamas:Theoreticaland AppliedKarstology,v.13–14,p.57–65. Frank,E.F.,Mylroie,J.,Troester,J.,Alexander,E.C.,andCarew,J.L., 1998,Karstdevelopmentandspeleologensis,IsladeMona,Puerto Rico:JournalofCaveandKarstStudies,v.60,no.2,p.73–83. Hill,A.L.,1984,OriginoftheKellyHillCaves:Helictite,v.22,p.6–10. James,P.R.,andClark,I.F.,2002,Geology, in Davis,M.,Twidale,C.R., andTyler,M.J.,eds.,NaturalHistoryofKangarooIsland: Richmond,Australia,RoyalSocietyofSouthAustralia,p.1–22. Jenson,J.W.,Keel,T.M.,Mylroie,J.R.,Mylroie,J.E.,Stafford,K.W., Taborosi,D.,andWexel,C.,2006,KarstoftheMarianaIslands:The interactionoftectonics,glacioeustasyandfresh-water/sea-water mixinginislandcarbonates:GeologicalSocietyofAmericaSpecial Paper404,p.129–138. Labourdette,R.,Lascu,I.,Mylroie,J.,andRoth,M.,2007,Process-like modelingofflankmargincaves:Fromgenesistoburialevolution: JournalofSedimentaryResearch,v.77,p.965–979. Lace,M.J.,2008,CoastalcavedevelopmentinPuertoRico:Journalof CoastalProcesses,v.24,no.2,p.508–518. Ludbrook,N.H.,1983,MolluscanfaunasoftheEarlyPleistocenePoint EllenFormationandBurnhamLimestone,SouthAustralia: TransactionsoftheRoyalSocietyofSouthAustralia,v.107, p.37–49. McNeill,D.F.,2005,Accumulationratesfromwell-datedlateNeogene carbonateplatformsandmargins:SedimentaryGeology,v.175, p.73–87. Milnes,A.R.,Ludbrook,N.H.,Lindsay,J.M.,andCooper,B.J.,1983, ThesuccessionofCainozoicmarinesedimentsonKangarooIsland, SouthAustralia:TransactionsoftheRoyalSocietyofSouth Australia,v.107,p.1–36. Mylroie,J.E.,1980,CavesandKarstofSanSalvador:FieldGuidetoSan SalvadorIsland,Bahamas:FtLauderdale,Florida,CollegeCenterof theFingerLakes,p.67–91. Mylroie,J.E.,andCarew,J.L.,1990,Theflankmarginmodelfor dissolutioncavedevelopmentincarbonateplatforms:EarthSurface ProcessesandLandforms,v.15,p.413–424. Mylroie,J.E.,andCarew,J.L.,1991,ErosionalnotchesinBahamian carbonates:Bioerosionorgroundwaterdissolution?, in Bain,R.J.,ed., Proceedingsofthe5thSymposiumontheGeologyoftheBahamas: PortCharlotte,Florida,BahamianFieldStation,p.185–191. Mylroie,J.E.,andCarew,J.L.,1995,Chapter3,Karstdevelopmenton carbonateislands, in Budd,D.A.,Harris,P.M.,andSaller,A.,eds., UnconformitiesandPorosityinCarbonateStrata:AmericanAssociationofPetroleumGeologistsMemoir63,p.55–76. Mylroie,J.E.,Carew,J.L.,andVacher,H.L.,1995,Karstdevelopmentin theBahamasandBermuda, in Curran,H.A.,andWhite,B.,eds., TerrestrialandShallowMarineGeologyoftheBahamasand Bermuda:GeologicalSocietyofAmericaSpecialPaper300, p.251–267. Mylroie,J.E.,Jenson,J.W.,Taborosi,D.,Jocson,J.M.U.,Vann,D.T., andWexel,C.,2001,KarstfeaturesofGuamintermsofageneral modelofcarbonateislandkarst:JournalofCaveandKarstStudies, v.63,no.1,p.9–22. Mylroie,J.E.,andMylroie,J.R.,2007,DevelopmentoftheCarbonate IslandKarstModel:JournalofCaveandKarstStudies,v.69, p.59–75. Mylroie,J.R.,Mylroie,J.E.,Owen,A.M.,andWaterstrat,W.J.,2008a, CoastalcavesinBahamianeolianites:Originasflankmargincaves, Figure13.HansonBay.A–Long,linearandlevelnotches cutintotheeolianitesoftheBridgewaterFormation.Vertical whitebar1-m-longforscale.B–Rubblefaciesfoundonthe floorofthenotchesshowninFigure13A.Thematrixis sandyandwhiteorgray,notred,andtheclastsaremore roundedthannormallyseeninapaleosol,andareinterpreted asaback-beachrubblefacies.Pencil15-cm-longforscale (blackarrow,samelengthasthepencil). C AVESASSEALEVELANDUPLIFTINDICATORS ,K ANGAROO I SLAND ,S OUTH A USTRALIA 46 N JournalofCaveandKarstStudies, April2009

PAGE 16

seacaves,andtafonicaves[abs]:JournalofCaveandKarstStudies, v.70,no.3,p.179–180. Mylroie,J.E.,Mylroie,J.R.,andNelson,C.N.,2008b,Flankmargincave developmentintelogeneticlimestonesofNewZealand:Acta Carsologica,v.37,no.1,p.15–40. Mylroie,J.E.,andMylroie,J.R.,2009,Arapidreconnaissanceofa QuaternaryeolianiteislandofAustralia:RottnestIsland,with comparisonstotheBahamas, in Martin,J.B.,andSiewers,F.D., eds.,Proceedingsofthe14 th Symposiumonthegeologyofthe Bahamasandothercarbonateregions,June2008,GeraceResearch Centre,SanSalvadorIsland,Bahamas:(inpress). Owen,A.M.,2007,TafonicavesinQuaternarycarbonateeolianites: ExamplesfromTheBahamas[MSc.thesis]:MississippiState, MississippiStateUniversity,187p.http://library.msstate.edu/etd/ show.asp?etd etd-05142007-143443 Palmer,A.N.,1991,Originandmorphologyoflimestonecaves: GeologicalSocietyofAmericaBulletin,v.103,p.1–25. Raeisi,E.,andMylroie,J.E.,1995,Hydrodynamicbehaviorofcaves formedinthefresh-waterlensofcarbonateislands:Carbonatesand Evaporites,v.10,no.2,p.207–214. Roth,M.J.,2004,Inventoryandgeometricanalysisofflankmargincaves oftheBahamas[MSc.thesis]:MississippiStateUniversity,117p. http://library.msstate.edu/etd/show.asp?etd etd-07062004-164930 Roth,M.J.,Mylroie,J.E.,Mylroie,J.R.,Ersek,V.,Ersek,C.C.,and Carew,J.L.,2006,FlankMarginCaveInventoryoftheBahamas, in Davis,R.L.,andGamble,D.W.,eds.,Proceedingsofthe12 th SymposiumontheGeologyoftheBahamasandOtherCarbonate Regions:SanSalvador,Bahamas,GeraceResearchCenter, p.153–161. Short,A.D.,andFotheringham,D.G.,1986,Coastalmorphodynamics andHoloceneevolutionoftheKangarooIslandcoast,South Australia,TechReportNo.86/1:Sidney,Australia,Universityof Sydney,CoastalStudiesUnit,92p. Stafford,K.W.,Mylroie,J.E.,Taboros i,D.,Jenson,J.W.,andMylroie, J.R.,2005,KarstdevelopmentonTinian,Commonwealthofthe NorthernMarianaIslands:Controlsondissolutioninrelationtothe carbonateislandkarstmodel:JournalofCaveandKarstStudies, v.67,no.1,p.14–27. Taboros i,D.,Mylroie,J.E.,andKirakawa,K.,2006,Stalactiteson tropicalcliffs:Remnantsofbreachedcavesorsubaerialtufadeposits? ZeitschriftfurGeomorphologie,v.50,p.117–139. Twidale,C.R.,andBourne,J.A.,2002,TheLandSurface, in Davis,M., Twidale,C.R.,andTyler,M.J.,eds.,NaturalHistoryofKangarooIsland: Richmond,Australia,RoyalSocietyofSouthAustralia,p.23–35. Vacher,H.L.,andMylroie,J.E.,2002,Eogenetickarstfromthe perspectiveofanequivalentporousmedium:Carbonatesand Evaporites,v.17,no.2,p.182–196. Vogel,P.N.,Mylroie,J.E.,andCarew,J.L.,1990,Limestonepetrology andcavemorphologyonSanSalvadorIsland,Bahamas:Cave Science,v.17,p.19–30. Waterstrat,W.J.,2007,Morphometricdifferentiationofflankmargin cavesandlittoral,orseacaves[MSc.thesis]:MississippiState University,201p.http://library.msstate.edu/etd/show.asp?etd etd-04052007-150907 J.E.M YLROIEAND J.R.M YLROIE JournalofCaveandKarstStudies, April2009 N 47



PAGE 1

FORMATIONOFSEASONALICEBODIESAND ASSOCIATEDCRYOGENICCARBONATESINCAVERNEDE L’OURS,QUE BEC,CANADA:KINETICISOTOPEEFFECTS ANDPSEUDO-BIOGENICCRYSTALSTRUCTURES D ENIS L ACELLE 1 ,B ERNARD L AURIOL 2 AND I AN D.C LARK 3 Abstract: Thisstudyexaminesthekineticsofformationofseasonalcaveiceformati ons (stalagmites,stalactites,hoar,curtain,andfloorice)andtheassocia tedcryogeniccalcite powdersinCavernedel’Ours(QC,Canada),ashallow,thermally-responsi vecave.The seasonaliceformations,whicheitherformedbythe:(1)freezingofdripp ingwater(ice stalagmiteandstalactite);(2)freezingofstagnantorslowmovingwater (flooriceand curtainice)and;(3)condensationofwatervapor(hoarice),all(exceptf loorice)showed kineticisotopeeffectsassociatedwiththerapidfreezingofcalcium–bi carbonatewater. Thiswasmadeevidentinthe d D, d 18 Oand d (deuteriumexcess)compositionsofthe formedicewheretheyplotalongakineticfreezingline.Thecryogeniccal citepowders, whicharefoundonthesurfaceoftheseasonaliceformations,alsoshowkin eticisotope effects.Their d 13 Cand d 18 Ovaluesareamongthehighestmeasuredincold-climate carbonatesandarecausedbytherapidrateoffreezing,whichresultsinst rongC-O disequilibriumbetweenthewater,dissolvedCspeciesinthewater,andpr ecipitating calcite.Althoughthecryogeniccalciteprecipitatedaspowders,divers ecrystalhabits wereobservedunderscanningelectronmicroscope,whichincludedrhombs ,aggregated rhombs,spheres,needles,andaggregatedstructures.Therhombcrystalh abitswere observedinsamplesstoredandobservedatroomtemperature,whereasthes phereand needlestructureswereobservedinthesampleskeptandobservedundercry ogenic conditions.Consideringthattheformationofcryogeniccalciteispurel yabiotic(freezing ofcalcium–bicarbonatewater),thepresenceofsphericalstructures,co mmonly associatedwithbioticprocesses,mightrepresentvaterite,apolymorph ofcalcitestable onlyatlowtemperatures.Itisthereforesuggestedthatcareshouldbetak enbefore suggestingbiologicalorigintocalciteprecipitatesbasedsolelyoncry stalhabitsbecause theymightrepresentpseudo-biogenicstructuresformedthroughabiotic processes. I NTRODUCTION MostterrestrialfreshwatershaveachemistrydominatedbyCa 2 andHCO 3 solutesthatoriginatesfromthe preferentialdissolutionofcalcareouscomponentsofthe bedrock.Evenincrystallinebedrockenvironments,where thebedrockcancompriselessthan1%carbonate,calcite dissolutionwilldominateoversilicateweatheringdueto thehighersolubilityofcalciteoversilicate(Whiteetal., 1999).Therefore,whenasolutioncontainingbothCa 2 andHCO 3 solutesfreezes,precipitationofcryogenic calcite(CaCO 3 ),orotherformsoflow-temperature carbonates,likevaterite( m -CaCO 3 )andikaite(CaCO 3 6H 2 O),isexpected,irrespectiveofthelocalgeology.Inthe naturalenvironment,cryogeniccarbonatesarequite commoninareaswheretheairtemperaturesfallbelow thefreezingpointforatleastafewmonthsoftheyear. Freezingcaves,locatedinareasoflimestonebedrock,are amongthemostsusceptibleenvironmentsinwhichtofind cryogeniccarbonateprecipitates(ClarkandLauriol,1992; Zaketal.,2004;Lacelle,2007;Zaketal.,2008;Richterand Riechelmann,2008).However,cryogeniccarbonatesare alsocommonlyencounteredonthesurfaceofaufeisor rivericings(Hall,1980;Pollard,1983;ClarkandLauriol, 1997;Lacelleetal.,2006)andonthesurfaceofclasts (Hallet,1976;Fairchildetal.,1993;Marlinetal.,1993; Courtyetal.,1994).Themostcommonlyprecipitated carbonatemineraliscalcite,butvateriteandikaitehave alsobeenreportedfromsomehighArcticandAntarctic environments(Pauly,1963;Suessetal,1982;Marion,2001; Omelonetal.,2001;Grasby,2003).However,these metastablemineralshaveneverbeenidentifiedinfreezing caves. ExperimentalworkdonebyHallet(1976),Fairchildet al.(1996),andKillaweeetal.(1998)haveshownthatthe formationofcryogeniccarbonatemineralsinvolveda 1 PlanetaryExplorationandSpaceAstronomy,CanadianSpaceAgency,6767r oute del’aeroport,St-Hubert,QC,J3Y8Y9,Canada.denis.lacelle@asc-csa.g c.ca 2 DepartmentofGeography,UniversityofOttawa,60UniversitySt.,Ottawa ,ON, K1N6N5,Canada.blauriol@uottawa.ca 3 DepartmentofEarthSciences,UniversityofOttawa,140LouisPasteur,Ot tawa, ON,K1N6N5,Canada.idclark@uottawa.ca D.Lacelle,B.Lauriol,andI.D.Clark–Formationofseasonalicebodiesan dassociatedcryogeniccarbonatesinCavernedel’Ours, Que bec,Canada:Kineticisotopeeffectsandpseudo-biogeniccrystalstruc tures. JournalofCaveandKarstStudies, v.71,no.1,p.48–62. 48 N JournalofCaveandKarstStudies, April2009

PAGE 2

seriesofchemicalprocessesanduniquekineticsof dissolutionleadingtotheprecipitationofcarbonates. Duringinitialfreezingofacalciumbicarbonatesolution, theCa 2 andHCO 3 solutesincreasetoapointwhere eventuallytheirionactivityproductmightreachand exceedthecalcitesaturationpoint,causingcalciteto precipitatefromthesolution.Duringtheirexperimental work,Fairchildetal.(1996)identifiedtheprecipitationof bothcalciteandvateriteminerals,representedbyrhombs (oraggregatedrhombs)andspheres,respectively.For cryogeniccalciteprecipitatedunderequilibriumconditions, itwasrecentlydemonstratedbyLacelleetal.(2006)thatin additiontotherateoffreezing,thedegreeofC-Oisotope fractionationisalsocontrolledbytheattainmentofcalcite saturation.However,whentherateoffreezingisincreased, theamountofC-Oisotopefractionationbetweencalcite andwaterresultsinstrongC-Oisotopedisequilibrium, producingcalcitewithhighstableC-Oisotopecomposition (ClarkandLauriol,1992).Therefore,thechemical processesandtherateatwhichtheyoccurinthecalcitewater-gassystemplayanimportantroleindeterminingthe stableC-Oisotopecompositionandcrystalhabitsof cryogeniccarbonates. Infreezingcaves,cryogeniccarbonateprecipitatesare commonlyobservedonthesurfaceofperennial/seasonal iceformations(i.e.,iceplugs,stalagmites,andstalactites) ascryptocrystallinecarbonatepowders,oronthefloorof freezingcavesasloosecalcitepearlsorcarbonatepowders (e.g.,ClarkandLauriol,1992;Zaketal.,2004).Inthis study,we(1)examinethegeochemicalandstableO-H isotopecompositionofthevarioustypesofseasonalice formationsinCavernedel’Ours(Que bec,Canada),a thermally-responsivecave,and(2)analyzethemineralogy (XRD),micro-morphologies(SEM)andstableC-O isotopecompositionofthecryogeniccarbonatepowders associatedwiththeformationofthevariousiceformations.Wealsodocumentanewtypeofcarbonate discoveredinCavernedel’Ours,spidersilkcalcite.The micro-morphologiesofthecryogeniccalcitepowderswere examinedundercryogenicconditionsandatroom temperaturetodeterminethepotentialpresenceofcalcite polymorphs(vateriteorikaite),whichhaveshown characteristiccrystalhabits(e.g.,Omelonetal.,2001; Grasby,2003).Tofurtherunderstandtheconditionsand processesunderwhichthecryogeniccavecalciteprecipitated,theirmicro-morphologiesandstableC-Oisotope compositionarecomparedtothoserelatedtoaufeis. Aufeis,whicharesheet-likemassesofhorizontallylayered icethataccumulateonriverchannelsbysuccessive overflowofperennialgroundwaterfedspringsupon exposuretocoldtemperature,containvariouscryptocrystallinepowders,includingcalcite,vateriteandikaite, withintheindividualicelayers.Consequentlycryogenic aufeisandcavecalcitepowdersaretheonlyknowntypesof cryogeniccalciteprecipitatingascryptocrystallinepowders. C AVERNEDE L’O URS S ITE D ESCRIPTION Cavernedel’Ours(45 u 40 N;75 u 39 W)islocatedinthe OttawavalleyregiononPrecambrianGrenvillemarble outcropandneartheeasternlimitoftheCanadianShield (Fig.1).Thecaveislocatedinaregioncharacterizedby largeseasonaltemperaturevariationsandrelativelyhigh precipitation.Themeanannualairtemperature T (1970– 2000)recordedattheOttawameteorologicalstationis6.0 6 0.8 u C(Januarymean T : 10.8 6 2.9 u C;Julymean T : 20.9 6 1.1 u C),andtheareareceivesatotalof945mmof precipitationannually,ofwhichone-thirdfallsassnow (EnvironmentCanada,2004).Thevegetationsurrounding thecaveconsistsofamixed-deciduousforestcomposedof spruce( Picea ),hemlock( Tsuga ),cedar( Thuja ),birch ( Betula ),andmaple( Acer ),whichischaracteristicofthe middleOttawazoneoftheGreatLakes–St.Lawrence forestregion(Rowe,1972).Thesoiloverlyingthecave consistsofaslightlyacidicorganicmatterandplantlitter (pHof4.5),typicalofsoilscoveredbyadeciduousforest (Hagen-Thornetal.,2004).The p CO 2 inthesoilis approximately10 3.3 to10 3.1 ppmVinJulyanddecreases to10 3.5 to10 3.4 ppmVinJanuary,reflectingthe biologicalactivityoftheoverlyingvegetation. TheageofformationofCavernedel’Oursisunknown, butgiventheabsenceofglacialsedimentsinsidethecave,it wasprobablyinitiallyscouredintothePrecambrian Figure1.LocationandtopographyofCavernedel’Ours (QC,Canada).Theseasonaliceformationsandassociated cryogeniccarbonateswerecollectedintheopencavity (sectionI).Moonmilkdepositsarefoundinthemain passages(sectionIII). D.L ACELLE ,B.L AURIOL AND I.D.C LARK JournalofCaveandKarstStudies, April2009 N 49

PAGE 3

Grenvillemarbleoutcropbysubglacialmeltwaterflowing fromtheLaurentideicesheet.TheGrenvillemarble,which consistsofmetamorphosedlimestone(CaCO 3 95%; d 18 O 7.9 % ; d 13 C 2.3 % )(Kretz,1980,2001),ishighly solubleandcontainsnumerousinclusionsofquartz, gabbro,garnetandfeldspar(DresserandDenis,1946; Pre vostandLauriol,1994). Cavernedel’Ours,whichmeasuresupto280min length,isdividedintothreesections:(1)alargeopencavity measuring30-m-longand5-m-wide;(2)themainundergroundsectionconsistingof250mofnarrowsubhorizontalpassages;and(3)a30-m-longandverynarrow passagethatconnectstheopencavitytothemain undergroundpassages(Fig.1).Theopencavityandmain passagesbothreachafewmetersinheight.Thereisasmall closed-basinlakeadjacenttotheentranceoftheopen cavity,andthepresenceofabeaverdampreventsthelake waterfromfillingmostoftheopencavity.However,the beaverdamallowsforasmallstreamtoflow(ca.4– 5Ls 1 )alongtheflooroftheopencavity,andthestream reachesthemainpassagesofthecavethroughanarrow networkoffissures.Neartheentranceofthecavity,the streampartiallyfreezesduringwinter,butitremains unfrozen(streamtemperaturenear4 u C)attheendof thecavity.Thecryogeniccavecalcitedepositsdiscussedin thisstudyareonlyfoundintheopencavity;insidethe remainderofthecave,moonmilkisthedominanttypeof speleothem,althoughafewHolocene-ageflowstonesare alsopresent(Lacelleetal.,2004). M ICROCLIMATE ,S EASONAL I CE D ISTRIBUTIONAND C RYOGENIC C ALCITE P OWDERS Themicroclimateinsidethemainundergroundpassages ofthecavewasdescribedinLacelleetal.(2004),anditwas foundthattheseasonalairtemperatureandrelative humidityfluctuatedbetween5–15 u Cand85–100%, respectively.However,themicroclimateintheopencavity tendstoreflectthatoftheoutsideairtemperature.January airtemperaturesrangefrom 10 u Cneartheentrance,to near0 u Cattheendofthecavity.Therefore,thecolddense airthatentersthecavityinwintercirculatesalongthe floor.Asitprogressesintheopencavity,theairwarmsup, rises,andflowsbacktowardstheentrancealongtheroof (Fig.2).Thedistributionoficeformationsinthecavityis inpartcontrolledbythewinter0 u Cisotherm,which extendstoapproximately20minsidethecavity. InCavernedel’Ours,asinmostfreezingcavesin Canada(FordandWilliams,2007),theseasonalice formationsformedeitherbythe:(1)freezingofdripping water(icestalagmiteandstalactite),(2)freezingofstagnant orslowmovingwater(flooriceandcurtainice),or(3) condensationofwatervapor(hoarice)(Fig.3).Figure2 presentsthevariousformsofseasonalcaveiceformations inrelationtothemicroclimaticzonesinthecavity.Within thefirst5mfromtheentrance,afewsmallicestalagmites (5-to15-cm-high)arefoundonthelimestoneblocks,and theceilingiscoveredbyhoarice.Between5and10mfrom theentrance,theicestalagmites,shapedasinversebowling pins,aremoreabundantandmeasureupto1-m-highand 20-cm-wide.Numerousicestalactitesarealsopresentin thissection.Thestalactiteshaveaconicalshapethattapers attheirtipsandmeasureupto1-m-long.Thesetwotypes oficeformations,whichareformedfromthefreezingof drippingwater,growmorerapidlywhentheairtemperaturesinthecavityareverycold,whereaswhentheair temperaturesareslightlybelow0 u C,thedrippingwater slowlycirculatesontheirsurface,resultinginathickening oftheiceformations.Inthismiddlesection(5to10m)and neartheentranceofthecavity,theshapesoftheice stalagmitesandstalactitesarenotonlycontrolledbythe Figure2.Schematicdiagramillustratingthemicro-climaticzonesinthe opencavityinCavernedel’Ours,QC,Canada,in relationtothevarioustypesofseasonaliceformations. F ORMATIONOFSEASONALICEBODIESANDASSOCIATEDCRYOGENICCARBONATESIN C AVERNEDEL ’O URS ,Q UE BEC ,C ANADA :K INETICISOTOPEEFFECTS ANDPSEUDO BIOGENICCRYSTALSTRUCTURES 50 N JournalofCaveandKarstStudies, April2009

PAGE 4

freezingofdrippingwater,butalsobysublimationofthe icesurfaces.Thisprocessliesintheequilibriumbetween temperatureandhumiditydifferencesbetweentheairand theiceformations.Thecoldanddryairthatentersthe opencavityflowsaroundtheicestalagmitesand stalactites,andtheinteractionofthewatermoleculesat theboundarylayerbetweentheiceandairmassescausesa transfer(removal)oficetotheair.Alsofoundinthe middlesectionofthecavityishoaricealongtheroofand curtainicegrowingperpendicularontheupperwalls. Furtherawayfromtheentrance(10to20m),theice stalagmiteshaveatubularformandmeasureupto1.2-mhigh,indicatingthatsublimationisnolongeractiveinthe farthestsection.Icestalactitesandhoariceareabsentin thissectionbecausetheairtemperaturealongtheroofis greaterthan0 u C,exceptonverycoldwinterdays. Throughouttheopencavity(exceptneartheend),the flooriscoveredbyice.Thisiceformsbythefreezingof lakewaterpassingbeneaththebeaverdaminwinter.Near theentrance,theflooricemeasures25-cm-thick,andis composedofcandleicecrystalsupto5-cm-long.These characteristicsapproachthoseofaufeis,andassuch,the formationoftheflooriceresemblesthatofaufeisinthe Arctic,albeitatamuchsmallerscale(discussedfurtherin text). Cryogeniccalcitepowders,whichformduringthe freezingofasolutioncontainingdissolvedcalciumand bicarbonatesolutes,arevisibleonthesurfaceonthefloor iceandalsoonthesublimatedsectionsoftheice stalagmitesandstalactites.Deeperintheopencavity, wheretheiceformationsarenotmodifiedbysublimation, nocalcitepowdersarefoundontheirsurface;however,it shouldbenotedthatthemeltingoficestalagmitesgrowing deeperinthecavealsoreleasedcalcitepowders.The cryogeniccalcitepowdershaveawhitishtoyellowishcolor andbecomeprogressivelythickerontheicesurfacesasthe wintermonthsadvance,withmaximumaccumulations reaching1–2mminMarch. Calcitepowdersarenotonlyfoundonthesurfaceofice formations,butalsowereobservedonspiderwebs attachedtothetipsofthesmallicestalagmitesgrowing attheentranceofthecavity.Intheliterature,thistypeof calciteisrarelydocumented,butMuraseetal.(2001) nameditspidersilkcalcite,followingtheirdiscoveryina laboratoryexperiment.Theonlyspiderspeciesobservedin thesummerintheentranceofCavernedel’Ourswas Meta ovalis ,andthisspeciesisthemostcommonspiderlivingin theentranceofcavesinNorthAmerica(Dondaleetal., 2003).Thespidersilkhasastrongabilityforwater condensation;andtherefore,thecondensationandsubseFigure3.Schematicdiagramofvarioustypesofseasonaliceformationsen counteredintheopencavityinCavernedel’Ours, QC,Canada. D.L ACELLE ,B.L AURIOL AND I.D.C LARK JournalofCaveandKarstStudies, April2009 N 51

PAGE 5

quentfreezingofwatervaporcontainingcaveaerosolson thespidersilkcouldproducecryogenicspidersilkcalcite, assumingthecaveaerosolscontainsomedissolvedcalcium andbicarbonatespecies. F IELD S AMPLINGAND A NALYTICAL P ROCEDURES C AVE I CE Inthewintersof2006–2007,varioustypesofseasonal caveiceformations(hoar,curtain,floor,stalagmite,and stalactiteice)foundintheopencavityofCavernedel’Ours werecollectedforgeochemicalandstableO-Hisotopes. Sincethehoarandcurtainicecouldreadilybebrokenoff, theywerecollectedandtransferreddirectlyintosealed plasticbags,whereastheflooricewassampledusinganice axeandthentransferredintosealedplasticbags.Nearthe entranceofthecavity,anentireicestalagmite(50cm)and stalactite(80cm)werebrokenofffromtheirlimestone blockandceiling,respectively,andbroughtbacktothe laboratoryinathermallyinsulatedbox.Waterdripping fromtheceilingintheopencavitywasalsocollectedduring themonthsofOctoberandDecember2006andApril2007 inglassamberbottlesforgeochemicalandstableO-H-C isotopemeasurements. PriortogeochemicalandstableO-Hisotopeanalyses, allicesamplesweremeltedinthelaboratory,filtered through0.45 mporediameterfilters,andtransferredin 20mlpre-rinsedpolyethylenebottles.However,theice stalagmiteandstalactiteweresectionedwithapre-cleaned sawinto2-cm-thickslicesalongtheirgrowthaxistoverify thechemicalandstableisotope(O-H)variationsduring theiraccretion.ThepHofthemeltedicesamples,which representsanequilibriumvaluebetweenthewaterand potentiallyanydissolvedcryogeniccalciteunderthe laboratorypartialpressureofCO 2 ,wasdeterminedusing aFisherAccumet610ApHmetercalibratedwithpH4 and7buffersolutions.Majorcations(Ca 2 ,Mg 2 ,Na and K )wereanalyzedandacidifiedtopH2usingultra-pure nitricacidbyInductivelyCoupledPlasmaAtomicEmissionSpectroscopy(ICP-AES).Allsampleswererunin duplicateattheUniversityofOttawa(Departmentof EarthSciences)andtheanalyticalreproducibilitywas 6 5%. The 18 O/ 16 Oratioofthemeltedicesampleswas determinedonCO 2 equilibratedwiththewaterat25 u C. TheD/HratiowasmeasuredonH 2 isotopicallyequilibratedwiththewaterat25 u CusingaPtbasedcatalyst. Bothstableisotopemeasurementsweremadeonthesame sampleusingaGasBenchIIinterfacedwithaFinnigan MatDelta XPisotopemassspectrometerattheG.G. HatchLaboratory(UniversityofOttawa).Resultsare presentedusingthe d -notation,where d representsthe partsperthousanddifferenceof 18 O/ 16 OorD/Hina samplewithrespecttoViennaStandardMeanOcean Water(VSMOW).Analyticalreproducibilitywasof 6 0.1 % for d 18 Oand 6 1.5 % for d D. C RYOGENIC C AVE C ALCITE Themineralcompositionofthecryogeniccarbonate deposits,identifiedascalciteinallcases,wasdetermineda fewmonthsaftercollection.Thesampleswerepowdered, mixedwithacetoneandspreadoveraglassslideand analyzedusingaPhillipsPW-1800x-raydiffractometer withastepsizeof0.02andscanningspeedof0.4seconds persteptorecordthex-raydiffractionspectra. Thecryogeniccalcitepowderswerealsocollectedfrom theiceformationsusingdifferentmethodsandexamined underscanningelectronmicroscope(SEM)eitherunder cryogenicorroomtemperatureconditionstoverifythe effectofpost-fieldstoringontheirmicro-morphologiesand todeterminethepotentialpresenceofcalcitepolymorphs. Fourmethodswereused: (1)Cryogeniccalcitepowderswerecollecteddirectly fromthesurfaceoftheiceformations,placedin sterileroll-topplasticbagsandkeptatroom temperatureuntilanalyzedunderSEM. (2)Cryogeniccalcitepowderswerecollectedalongwith theiceonwhichtheyrested,stored,andanalyzed usingaSEMundercryogenicconditionstoverifythe undisturbedmicro-morphologies. (3)Thethirdmethodconsistedofcollectingasectionof icestalagmitethatwasmeltedinaglassbeakerback inthelaboratory.Thecalcitepowderswereretrieved fromthebeakerafterthewaterhadcompletely evaporatedandthenexaminedunderSEMatroom temperature. (4)Forthefinalmethod,anicestalagmitewassectioned intosmallblocks,placedinabeakercoveredbyan aluminumsheetanddesiccated(1atmandtemperatureof 5 u C)inacommercialdessicator(LabconcoFreezone)attheGeologicalSurveyofCanada. Theresidualcalcitepowderswerecollectedfromthe beakerandkeptatsub-freezingtemperatureuntil SEMexaminationundercryogenicconditions. Calcitepowderswerealsocollectedfromthespider websattachedonthesmallicestalagmitesnearthe entranceofthecavitytoexaminethemorphologyofthe crystals.Thecryogenicspidersilkcalcitewascollected directlyonacarbontapemountedontoanaluminumstub andkeptatsub-freezingtemperatureuntilexamination underSEM.PriortoexaminationunderSEM,allcalcite powders(exceptforthepowdercollectedwiththeice substrate,whichwasputdirectlyuncoveredontoan aluminumstub)weremountedontoanaluminumstub usingdoubled-sidedcarbontapeandthensputter-coated withgoldfor60seconds.Themicro-morphologiesof calciteprecipitateswereexaminedusingaJEOL6400 SEMattheUniversite duQue beca `Montre al. The 18 O/ 16 Oand 13 C/ 12 Cratiosofthecryogeniccave calcitepowdersweredeterminedonCO 2 gasproducedby reactingthepowderedcalcitewith100%phosphoricacid (H 3 PO 4 )inglassseptumvialsfor24hoursat25 u C.The F ORMATIONOFSEASONALICEBODIESANDASSOCIATEDCRYOGENICCARBONATESIN C AVERNEDEL ’O URS ,Q UE BEC ,C ANADA :K INETICISOTOPEEFFECTS ANDPSEUDO BIOGENICCRYSTALSTRUCTURES 52 N JournalofCaveandKarstStudies, April2009

PAGE 6

evolvedCO 2 gaswasanalyzedincontinuousflowusinga GasBenchIIinterfacedwithaFinniganMatDelta XP isotopemassspectrometerattheG.G.HatchIsotope Laboratory,UniversityofOttawa.Stableisotopedatafor CandOareexpressedin d -notation,where d representsthe partsperthousanddifferenceof 13 C/ 12 Cand 18 O/ 16 Oina samplewithrespecttotheViennaPee-DeeBelemnite standard(VPDB).Analyticalreproducibilityis 6 0.15 % forbothisotopes. R ESULTS C AVE I CE F ORMATIONS ThegeochemicalandstableO-Hisotopecompositions ofthevariousseasonalcaveiceformationsinCavernede lÂ’OursarepresentedinFigures4and5.Theiceformations havesimilargeochemicalandisotopiccompositions:apH rangingbetween7and8,aCa 2 concentrationaveraging 15.9 6 6.7mgL 1 ,typicalofkarstwaterinthearea (Pre vostandLauriol,1994),and d 18 Oand d Dvalues averaging 8.2 6 0.2 % and 57.6 6 2.8 % ,respectively (Fig.4A).Althoughthe d 18 Ocompositionsoftheice stalagmite,icestalactite,curtainandhoaricearesimilar, separateregressionlinesareobtainedina d D d 18 O diagram.Individualregressionslopesof5.1( d D 5.1 d 18 O 16.2; R 2 0.58),5.6( d D 5.6 d 18 O 11.1; R 2 0.57), and2.7( d D 2.7 d 18 O 37.6; R 2 0.40)arecalculated fortheicestalactite,icestalagmite,andicecurtain, respectively(Fig.4B),suggestingpotentiallyuniquemechanismofformationforeachtypeofice. Samplingalongthegrowthaxisoftheicestalactiteand stalagmiterevealedslightgeochemicalandisotopicvariationsrelatedtotheirseparateaccretionprocesses(Fig.5). Inthe80-cm-longicestalactite,thepHincreasesfrom6.8 atthebaseto7.3atitstip.Thegeochemicalcompositionof theicestalactiteisdominatedbyCa 2 (14 6 2.3mgL 1 ), followedbyNa (1.7 6 0.9mgL 1 )andK (1.2 6 0.8mgL 1 ),buttheirconcentrationsdonotshowany trendsfromthebasetothesummit.Itisinterestingto observethattheconcentrationsofK andNa fluctuate together,whereasthevariationintheconcentrationof Ca 2 isindependent.ThesourceofK andNa isprobably relatedtosomeclasticcomponent,whichcanbeeitherclay transportedtogetherbythedripwaterorcaveaerosols depositedonthewetsurfaceoftheicestalactite.The d 18 O compositionoftheicestalactiteshowsaprogressive depletiontrendfrom 7.1 % atitsbaseto 8.7 % atitstip. Inthe50-cm-longicestalagmite,thepHalsoprogressivelyincreasesasitgrew,rangingfrom6.9atthebaseto 7.9atitstip.TheK andNa concentrationintheice stalagmiteissimilartothatofthestalactite.However,the Ca 2 solutesconcentrationreachesamuchgreaterconcentration(upto47.2mgL 1 ).Liketheicestalactite,the soluteconcentrationsinthestalagmitedonotshowany trendsduringiceaccretion(Fig.5).The d 18 Ooftheice stalagmitevariesbetween 7.9and 8.9 % .Itwas interestingtoobservethatoncethawed,thecolorofthe wateroftheicestalagmiteandstalactitewasyellowish, suggestingthepresenceofaminoacids.Infact,the cavedripwaterhasameasureddissolvedorganic content(DOC)rangingfrom4.1to6.9mgL 1 (unpublisheddata). C RYOGENIC C AVE C ALCITE P OWDERS The d 18 Oand d 13 Ccompositionofthecryogeniccalcite powdersispresentedinFigure6.Thestableisotopic compositionofthecryogeniccalcitepowdersisinvariantof thecaveiceformationfromwhichtheywerecollectedand Figure4.A)RangeofstableO-Hisotopecompositionof seasonaliceformationsinCavernedelÂ’Ours,QC,Canada, comparedtothatoftheseasonalprecipitationrecordedin Ottawa,whichisdefinedbytheLocalMeteoricWaterLine (LMWL: d D = 7.5 d 18 O 5.9;IAEA/WMO,2004).B) StableO-Hisotopecompositionofseasonaliceformations (icestalagmite,icestalactite,hoariceandcurtainice)in CavernedelÂ’Oursandtheirrespectiveregressionline( S ). D.L ACELLE ,B.L AURIOL AND I.D.C LARK JournalofCaveandKarstStudies, April2009 N 53

PAGE 7

ofthepost-samplinghandling.Thecryogeniccalcite powdersallhave d 18 Ovaluesrangingfrom 11.3to 3.7 % and d 13 Cvaluesbetween5.8and11.9 % ,which areamongthemostpositivevaluesintheliterature.The d 18 Oand d 13 Cofcalcitepowderscollectedfromthe sublimatedicestalagmitefallalsowithinthisisotopic range.Cryogeniccalcitepowderscollectedfromthesurface offlooriceinthenearbyLuskcave(45 u 39 N;75 u 38 W) yieldedverysimilar d 18 Oand d 13 Cvalues(Clarkand Lauriol,1992). Unlikethe d 18 Oand d 13 Cvalues,thecryogeniccalcite powdersproducedvariedcrystalarrangementsdepending onthepost-samplingstorageandanalyticalmethods (roomorsub-freezingtemperature)duringexamination underSEM(Figs.7and8).Thecrystalarrangementofthe firstgroupofcryogeniccalcitepowders,whichwere collectedfromthesurfaceoftheicestalagmitesand stalactitesandstoredatroomtemperaturepriorto analysis,iscomposedof3–8 mrhombohedralcrystals (Fig.7A–C).Thesurfaceofthecalcitecrystalsisoften pittedandtheiredgesareetched,suggestingcalcite disintegrationafterprecipitationbylocaldissolution.In contrast,thecryogeniccalcitepowdersanalyzeddirectly fromthesurfaceoftheiceandthosecollectedfrom sublimatedicestalagmitesundercryogenicconditions showedcrystalhabitsdifferentfromthefirstgroup (Fig.8).Thesecalcitecrystalsarecomposedofspheres rangingfrom 1to2 mindiameterandthickcalcite needlesupto20 mlong.Insomeinstances,several spheresaggregatetogether,producingachain-likeappearance(Fig.8B,D).Finally,thecryogeniccalcitepowders analyzedfromevaporatedmeltedicestalagmitesand stalactitesarecomposedof3–10 mrhombohedralcrystals thatareoftenstackedtogether(Fig.7D–F).Thesurfaceof thecrystalsissmooth,buttheiredgesareetched,similarto thefirstgroupofcalcitepowders(collectedfromthe surfaceoftheiceformationsandkeptatroomtemperature priortoanalysis).Overall,noneofthesecrystalhabits resemblethatofikaite,whichistypicallycomposedof anhedralcalcitecrystals(Omelonetal.,2001).However, thespherical-shapedcrystalaggregatesobservedunder cryogenicconditionscloselyresemblethoseofvaterite precipitatedundernaturalandlaboratorysetting(e.g., Turnbull,1973;Fairchildetal.,1996;VechtandIreland, 2000;Grasby,2003). Thespidersilkcalciteproducedauniquecrystal habit(Fig.9).Thecalciteattachedtothespider’ssilk showedelongatedcrystalsupto3 mlongthatformed clusterslessthan10 mwidealongthespider’ssilk (Fig.9C–F).Thesedimensionsofsingleclustersare probablyconstrainedbytheweightofthecalcitecrystals onthesilk. Figure5.Geochemical(Ca 2 ,Na ,K andpH)and d 18 O compositionalonggrowthaxisofanicestalactiteandice stalagmitefoundintheopencavityinCavernedel’Ours, QC,Canada. Figure6.StableC-Oisotopecompositionofcryogenic calcitepowdersfoundonthesurfaceofthevariousseasonal iceformationsinCavernedel’Ours.Alsoshownfor comparisonpurposesisthe d 13 Cand d 18 Ocompositionof moonmilkfoundinthemainpassagesofCavernedel’Ours (Fig.1),andcryogenicaufeiscalcitepowders(sampledfrom limestoneterraininthenorthernYukon,Canada),theonly otherknowntypeofcryogeniccalciteprecipitatingas cryptocrystallinepowders. d 13 Cand d 18 Odataformoonmilk andcryogenicaufeiscalcitepowdersderivedfromLacelleet al.(2004)andClarkandLauriol(1997),respectively. F ORMATIONOFSEASONALICEBODIESANDASSOCIATEDCRYOGENICCARBONATESIN C AVERNEDEL ’O URS ,Q UE BEC ,C ANADA :K INETICISOTOPEEFFECTS ANDPSEUDO BIOGENICCRYSTALSTRUCTURES 54 N JournalofCaveandKarstStudies, April2009

PAGE 8

Figure7.Secondaryelectronimagesofcryogeniccalcitepowderscollect edonthesurfaceofvariousiceformationsinCaverne del’Ours.TheimageswereacquiredatroomtemperaturewithaSEM.A–C)cry ogeniccalcitepowderscollectedfromthe surfaceoficestalagmitesandstalactites;andD–F)calcitepowdersreco veredfromabeakerinwhichasectionoficestalagmite wasmeltedandlefttoevaporateinthelaboratory. D.L ACELLE ,B.L AURIOL AND I.D.C LARK JournalofCaveandKarstStudies, April2009 N 55

PAGE 9

Figure8.Secondaryelectronimagesofcryogeniccalcitepowderscollect edonthesurfaceofvariousiceformationsinCaverne del’Ours.Theimageswereacquiredatsub-freezingtemperaturewithaSEM .A–C)cryogeniccalcitepowderscollectedfrom thesurfaceoficestalagmiteusingcryogenicsampletransportandanalys is;andD–F)calcitepowdersrecoveredfromabeaker inwhichasectionoficestalagmitewassublimatedinthelaboratoryunder cryogeniccondition,transport,andanalysis. F ORMATIONOFSEASONALICEBODIESANDASSOCIATEDCRYOGENICCARBONATESIN C AVERNEDEL ’O URS ,Q UE BEC ,C ANADA :K INETICISOTOPEEFFECTS ANDPSEUDO BIOGENICCRYSTALSTRUCTURES 56 N JournalofCaveandKarstStudies, April2009

PAGE 10

Figure9.A–B)Fieldphotographsofcalcitepowdersonspidersilkattache dtoasmallicestalagmiteneartheentranceofthe opencavityofCavernedel’Ours.C–F)Secondaryelectronimagesofspider silkcalciteacquiredatroomtemperaturewith aSEM. D.L ACELLE ,B.L AURIOL AND I.D.C LARK JournalofCaveandKarstStudies, April2009 N 57

PAGE 11

D ISCUSSION O RIGINOFTHE S EASONAL C AVE I CE F ORMATIONS : E VIDENCEFOR K INETIC I SOTOPE E FFECTS Theformationoftheseasonalcaveiceformations examinedinthisstudyiswellknownintheliterature(i.e., FordandWilliams,2007)andithasbeendescribedherein aprevioussection.However,littleisknownregardingthe degreeofisotopeequilibrium/disequilibriumduringthe formationofthehoar,curtain,stalagmiteandstalactiteice bodies.Suchinformationcanbeprovidedbyexamining the d D d 18 Orelationintheiceformations(Fig.4). Duringequilibriumfreezing,theicesamplesarealigned alongaregressionlinethatwillbelowerthanthelocal meteoricwaterline(LMWL; d D 7.5 d 18 O 5.9; R 2 0.97;IAEA/WMO,2004)becausetheamountofincorporationoftheheavierisotopesintheice(Dand 18 O),which followsaRayleigh-typefractionation,isslightlydifferent forbothisotopes(JouzelandSouchez,1982;Souchezand Jouzel,1984).AccordingtoJouzelandSouchez(1982),the slopeofthefreezinglinedependsmainlyontheinitial isotopiccompositionofthewater,withthemoredepleted watershavingalowerslopevalue.Theconditions prevailingduringfreezing(openversusclosedsystem), andtherateofsupplyofwatertothefreezingfronts,have littleeffectonthefreezingslope(SouchezandJouzel, 1984).BasedontheJouzelandSouchez(1982)modeland usinganinitialwater d 18 Oand d Dvaluesof 11.4 % and 78.8 % ,respectively,andtheaverageannualprecipitation valuesinOttawa(IAEA/WMO,2004),theiceformations inthecaveresultingfromequilibriumfreezingshouldplot alongaslopeof5.9. Anotherparameterthatprovidescluesintotherateof freezingduringtheformationoficebodiesisthe d d D relation,where d representsthedeuteriumexcess( d d D 8 d 18 O)(Dansgaard,1961).Duringequilibriumfreezing, thefirsticethatformshasagreater d D(and d 18 O)value duetothepreferentialpartitioningoftheDisotopeinthe ice(i.e.,withanassociateddepletionofheavyisotopesin theresidualwater),butasfreezingprogresses,the d D valuesoftheicebecomeprogressivelylower.Thisis accompaniedbyaconcurrentincreasein d valuesbecause thefreezingequilibriumslopehasavaluelowerthanthe LMWL.Asaresult,anegativerelationisexpectedbetween d and d Dduringequilibriumfreezing(Souchezetal., 2000). Ifweexaminethe d D d 18 Oand d d Drelationsin theCavernedelÂ’Oursseasonaliceformations,afewkey featuresemerge.First,allcaveiceformationshave d 18 O and d Dvalueswithintherangeofsummerprecipitation andofthestreamflowinginsidetheopencavity(Fig.4A). Secondly,thestalagmiteandstalactiteiceformationslieon positiveslopesthatarelessthantheLMWL(7.5),but similartothetheoreticalfreezingslope(5.9)(Fig.4B). Individualregressionslopesof5.1( d D 5.1 d 18 O 16.2; R 2 0.58)and5.6( d D 5.6 d 18 O 11.1; R 2 0.57)are calculatedfortheicestalactiteandicestalagmite, respectively(Fig.4B).Bycontrast,thecurtainicesamples plotalongaregressiveslope d D 2.7 d 18 O 37.6; R 2 0.40)thatismuchlessthantheLMWLandpredicted freezingslope.Thirdly,thereisnonegativerelation between d d Dinallcaveiceformations(Fig.10), suggestingthenon-existenceofafreezingslope;orinother words,freezingoccurredundernon-equilibriumconditions.Infact,eventhoughregressionslopeswere calculatedforthevariousiceformations,theircorrelation coefficientisratherweak( R 2 0.58).Finally,thehoarice ontheceilingofthecaveshowsstrongenrichmentinD,as itplotswellabovetheLMWL(Fig.4B).Overall,these unusualisotopicfeaturescannotbeexplainedbythe progressivefreezingofwaterunderequilibriumconditions, butcanbeattributedtokineticisotopeeffectsduringthe successivefreezingofthinlayersofwater.Thearguments infavorofnon-equilibriumfreezingfortheformationof eachseasonalicetypeformationsarepresentedbelow. Bydefinition,hoariceformsbydirectcondensationof watervaporonthecoldceiling.Theeffectofcondensation isevidentinthe d D d 18 Odiagram,wherethehoarice samplesplotwellabovetheLMWL(Fig.4B).Thisis attributedtotheeffectofaircirculationdynamicsinside theopencavity.Thenon-equilibriumevaporationofthe streaminsidetheopencavitywouldproducewatervapor plottingabovetheLMWL,followedbyequilibrium condensationofthewatervaporalongtheLMWL (Fig.11).Thisprocessproduces d valuesmuchgreater than10 % thatareincreasingunderdecreasingrelative humidityconditions.Asimplecalculationusingthe Figure10.Deuteriumexcess( d )andstableDisotope compositionofthevariousseasonaliceformationsin CavernedelÂ’Ours,QC,Canada.Horizontaldashedline represents d inlocalprecipitation. F ORMATIONOFSEASONALICEBODIESANDASSOCIATEDCRYOGENICCARBONATESIN C AVERNEDEL Â’O URS ,Q UE BEC ,C ANADA :K INETICISOTOPEEFFECTS ANDPSEUDO BIOGENICCRYSTALSTRUCTURES 58 N JournalofCaveandKarstStudies, April2009

PAGE 12

average d 18 Oand d Dofthesmallstreamintheopencavity inwinter( 7.5 6 1.0 % and 56.6 6 2.1 % ,respectively) (Lacelleetal.,2004),thefractionationfactorbetweenice andvapor( d 18 Oice-vapor 14.7 % ; d Dice-vapor 127 % ,bothat0 u C;Majoube,1971),andaveragerelative humidityinwinter(74.7%;thisvalueisprobablyslightly lessintheopencavitysincetheairbecomeswarmer), showsthatthecondensingicewouldhavea d 18 Oand d D compositionof 7.9 % and 39.2 % ,respectively,anda d excessof24.7 % .Thesepredictedvaluesaresimilartothe measured d 18 O, d Dand d valuesofthehoariceinCaverne delÂ’Ours(Figs.4Band10),indicatingthatthestableO-D isotopecompositionofhoarpreservedthecondensation signature. Icestalagmitesandstalactitesandicecurtainsoriginate fromthefreezingofdrippingwater.Asthe d 18 Ovaluesof theseiceformationsarewithintherangeofsummer precipitation(Fig.4A),thepercolatingwaterintheopen cavityduringthewinteroriginatesfromrainfallground waterstoredintheoverlyingsoil(epikarstzone),andnot fromsnowmeltwater.Inthe d Dand d 18 Odiagram (Fig.4B),theicestalagmiteandstalactiteplotnearthe predictedequilibriumfreezingslopeof5.9,althoughthey haveaweakcorrelationcoefficient( R 2 0.58).However, theicecurtainplotsalongamuchlowerslope( d D 2.7 d 18 O 37.6).Inthe d d Ddiagram,thethreeicetypesdo notshowaclearrelation,astheyarescatteredacrossbroad horizontalbands(Fig.10).Theseobservations,andthe factthatnotrendisdisplayedintheverticaldistributionof heavyisotopes( 18 OandD)duringtheaccretionoftheice stalagmiteandstalactite(Fig.5),indicatethattheygrew fromthesuccessivefreezingofwaterdrippingfromthe ceilingundernon-equilibriumconditions.Thesublimation oftheiceformationsbyaircirculationwouldnotaffectthe stableO-Hisotopeoftheremainingice,assublimationisa physicalsurfacephenomenon. F ORMATIONOF C RYOGENIC C AVE C ALCITE P OWDERS :A C OMPARISONWITH C RYOGENIC A UFEIS C ALCITE P OWDERS Cryogeniccavecalciteandcryogenicaufeiscalciteare theonlytwoknowntypesofcalcitepowdersformedbythe freezingofcalcium-bicarbonatewaters.Inthesimplest form,theformation(anddissolution)ofcalcitecanbe expressedbythisreaction: CO 2 aq H 2 O CaCO 3 < Ca 2 2HCO 3 1 Duringtheformationofcalcitemineralsunderequilibriumconditions,theheavyisotopes( 18 Oand 13 C)are preferentiallyincorporatedfromtheaqueousphaseinto themineralsinaproportiongovernedbytheisotope fractionationfactor,andbysuch,theC-Oisotope compositionofcalciteiscontrolledbythe d 18 Oand d 13 C DIC compositionoftheparentwaterandtemperature atwhichprecipitationoccurs.However,itiswellknown thatfreezingcanmodifytheisotopecompositionofthe aqueousphaseastheproductsarebeingisolatedimmediatelyaftertheirformation(Rayleighdistillation),leading toaprogressivedepletionin d 18 Oandenrichmentin d 13 C DIC intheresidualwater(Lacelleetal.,2006). However,itwasshownintheprevioussectionthatthe seasonalcaveiceformationsexaminedinthisstudyformed throughnon-equilibrium(rapid)freezingofwaters.Therefore,thekineticisotopeeffectduringtheformationofthe icestalagmitesandstalactitesbywaterfreezingrapidly shouldalsocreateisotopicdisequilibriumbetweenthe aqueousphaseandtheprecipitatingcalcite,thelatter showinganincreaseinheavyisotopes.ClarkandLauriol (1992)demonstratedthattherapidfreezingofacalcium bicarbonatesolutionthatisatornearcalcitesaturationled tostrongCisotopedisequilibriumbetweentheprecipitatingcalciteandtheescapingCO 2 ( e 13 CCaCO 3 -CO 2 31.2 6 3.1 % )(ClarkandLauriol,1992).Bycontrast,thestable Oisotopefractionationbetweenwaterandcalciteduring kineticfreezing( e 18 OCaCO 3 -H 2 O 36.7 6 1.3 % VSMOW)(ClarkandLauriol1992)isonlyslightlygreater thanequilibriumvalues( e 18 OCaCO 3 -H 2 O 33.6 % VSMOW)(KimandOÂ’Neil,1997).Accordingtothese kineticisotopeenrichmentfactorsandtheaverage d 13 C DIC ofdripwaters( 11.9 6 1.3 % ;unpublisheddata)and d 18 O ofthevariousiceformations( 8.2 6 0.2 % VSMOW),the Figure11. d D d 18 Osystematicduringcondensationof secondaryevaporatedwater(i.e.,smallstreamonthefloorof theopencavity).Whenhumidityconditionsarenear100%, precipitationplotsclosetothelocalmeteoricwaterline, however,underdecreasingrelativehumidityconditions,the vapor(evaporatedwater)becomesstronglydepletedand precipitationformedbyequilibriumcondensationplots furtherabovetheGMWL(globalmeteoricwaterline)along acondensationlinewithaslopeverysimilartotheGMWL. D.L ACELLE ,B.L AURIOL AND I.D.C LARK JournalofCaveandKarstStudies, April2009 N 59

PAGE 13

d 13 C(9.4 6 2.6 % )and d 18 Ovalues( 8.9 6 1.6 % VPDB) ofthecryogeniccalcitepowdersinCavernedelÂ’Oursare withintherangeexpectedofaformationbykinetic freezing. Cryogeniccavecalcitepowdershavehighlydistinct d 13 Cand d 18 Ovaluesfromthosemeasuredinthecryogenic calcitepowdersformedinassociationwithaufeis,thelatter havingmuchlower d 13 Cand d 18 Ovalues(Fig.6).Aufeis (icings)aresheet-likemassesofhorizontallylayeredice thatformonriverchannelsbysuccessivefreezingof overflowofperennialgroundwater-fedspringsupon exposuretocoldair.Stableisotopeprofileswithintheice layersindicatethatthegrowthofaufeisoccursunder closed-systemequilibriumconditions(Elver,1994;Clark andLauriol,1997).Asaresult,unliketheseasonalcaveice formationsdiscussedinthisstudy,theresidualwateris coveredbyasheetoficethatgrowsdownwards,leadingto saturationintheresidualwaterofvariousminerals, includingcalcite(CaCO 3 )andikaite(CaCO 3 6H 2 O), andtheirprecipitationwithintheice(Hall,1980;Clarkand Lauriol,1997;Heldmannetal.,2005;Lacelleetal.,2006). Accordingly,aclosedsystemRayleigh-typefractionation processoccursduringtheaggradationofaufeis.Although JouzelandSouchez(1982)andSouchezanddeGrotte (1985)demonstratedtheOisotopesystematicsassociated withtheslowfreezingoflowionicstrengthwatersunder equilibriumconditions,muchremainstobeknownabout theCisotope.ItcanbeexpectedthattheCisotope fractionationduringfreezingunderequilibriumconditions involvesaCisotopemassbalanceintheCO 2 -HCO 3 -CO 3 CaCO 3 system.ConsideringthattheDICspecies(CO 2aq HCO 3 -CO 3 )fractionatedifferentlyamongthem,andthat theirrelativeconcentrationissetbypH,the d 13 C DIC and d 13 CCaCO 3 duringequilibriumfreezingisnotsimply controlledbytheinitialgeochemicalandisotopiccompositionofthesourcewater,butalsobychangingphysical andgeochemicalconditionsasfreezingprogresses(i.e.,pH andexolvedCO 2 duringtheprecipitationofcalcite).Given thedifferentconditionsofformationfortheaufeis(closed systemequilibriumfreezing),thecryogenicaufeiscalcite powdersarecharacterizedby d 13 Cand d 18 Ovaluesthatare inequilibriumwiththatofthewaterfromwhichthey precipitated(Lacelleetal.,2006).Thesecharacteristics indicatethatthelimitingprocessaffectingthedegreeof deviationbetweenthestableC-Oisotopecompositionof thecryogenicaufeiscalcite(formedunderequilibrium conditions)andwaterfromwhichtheyprecipitatedisthe attainmentofcalcitesaturation,whereasforthecryogenic cavecalcitepowders(formedunderkineticconditions),itis therateofprecipitationofcalcite. M ICROMORPHOLOGIESOF C RYOGENIC C AVE C ALCITES : A BIOTICAND P SEUDO -B IOGENIC S IGNATURES Innaturalenvironments,suchasincaves,calcite crystalscommonlydisplayawiderangeofcrystal micromorphologies.Althoughrhombohedralhabitis highlysuggestiveofaninorganicprecipitationofcalcite, theformationofsphericalcrystalaggregatesisusually attributedtoanorganicinfluence(Folk,1993;Braissantet al.,2003).Inthisstudy,itwasshownthatthecryogenic cave-calcitepowders,whichareproducedbyfreezing,an abioticprocess,revealedhighlydifferentcrystalhabits dependingontheproceduresusedtocollectthesample, andtheconditionsunderwhichitwasexaminedunder SEM.Thesamplesanalyzedunderroomtemperatureall producedrhombohedralcrystals(eithersingleorstacked) (Fig.7),whereastheonesexaminedundercryogenic conditionsshowedsmallspheres( 2 m)andthickneedle structures(Fig.8).Needle-likecrystalsarealsothe dominanttypeofcrystalhabitobservedincryogenicaufeis calcite(e.g.,Lacelle,2007).Thesamplesexaminedunder cryogenicconditionsprobablypreservedtheundisturbed natureofthecalcitepowders,whereasthoseexaminedat roomtemperaturewereprobablyalteredandrecrystallized beforeanalysis.Thelattereffectwouldnotgreatlyaffect thestableC-Oisotopecompositionofthecalcite. Consideringthattheformationofcryogeniccarbonate powdersispurelyabioticandthatvateriteisstableatlow temperature( 10 u Cand1atm;Albright,1971)and precipitatesinasphericalshapewithcrystalsrangingfrom 0.05to2 m(Kraljetal.,1990;VechtandIreland,2000), themostplausibleexplanationfortheobservationofsmall spheres( 2 m)inthecrystalhabitsofthecryogenic powdersisthattheyconsistofvaterite.Thismetastable polymorphofcalciterecrystallizestocalcitewhenin contactwithwater(Silk,1970),whichwouldexplainwhy sphereswerenotobservedinthesamplesexaminedat roomtemperaturebecausetheywereexposedtowater duringmeltingoftheiceformations.Theformationof vateriteduringfreezingisnotunusualandwasalso observedbyFairchildetal.(1996)duringlaboratory experiments,andbyGrasby(2003)inspringdepositsinthe highArctic.AlthoughhigherpHvalues(between9.3and 10.0;Kraljetal.,1990)thanwhatwasmeasuredintheice formations(Fig.5)arenecessarytoprecipitatevaterite,it ispossiblethatfavorablemicroenvironmentswerecreated withintheaccretionofthevariousannualiceformations thatallowedvateritetoprecipitate.Vateritewasnot identifiedinourXRDmeasurementsbecausethesamples werenotkeptfrozenfortheseanalyses.Itistherefore suggestedthatcareshouldbetakenbeforesuggesting biologicaloriginofcalciteprecipitatesbasedsolelyon crystalhabitsbecausetheymightrepresentpseudobiogenicstructuresformedthroughabioticprocesses. C ONCLUSIONS Basedontheresultspresentedinthisstudy,the followingconclusionscanbereachedregardingthe formationofseasonalcaveicesandtheassociated cryogeniccalcitepowdersinCavernedelÂ’Ours,QC, Canada: F ORMATIONOFSEASONALICEBODIESANDASSOCIATEDCRYOGENICCARBONATESIN C AVERNEDEL Â’O URS ,Q UE BEC ,C ANADA :K INETICISOTOPEEFFECTS ANDPSEUDO BIOGENICCRYSTALSTRUCTURES 60 N JournalofCaveandKarstStudies, April2009

PAGE 14

1.Theseasonaliceformations,whicheitherformedby(1) freezingofdrippingwater(icestalagmiteandstalactite);(2)freezingofstagnantorslowmovingwater (flooriceandcurtainice)and;(3)condensationof watervapor(hoarice),all(exceptfloorice)showed kineticisotopeeffectsduringthefreezingofwater.This ismadeevidentinthe d d Ddiagramwheretheice formationsshownorelationbecausetheyarescattered acrossbroadhorizontalbands. 2.Thecryogeniccalcitepowders,whichprecipitateduring theformationoftheseasonaliceformations,alsoshow kineticisotopeeffects.Their d 13 Cvaluesareamongthe highestmeasuredincold-temperaturecarbonatesand arecausedbytherapidrateoffreezing,whichresultsin strongCdisequilibriumbetweenthewaterand precipitatingcalcite,thelattershowinganincreasein heavyisotopes.The d 18 Ocompositionofthecryogenic calcitepowdersalsoshowelevatedvaluesassociated withkineticfreezing. 3.Thecryogeniccalcitepowdersshowedvariedcrystal habits.Rhombs,aggregatedrhombs,spheresand needleswereallobservedunderSEM.Therhombs crystalhabitwasobservedatroomtemperature whereasthespheresandneedleswereobservedat sub-freezingtemperatureswithcryogenicstorageofthe samples.Thissuggeststhatthespherestructuresmight representvaterite,apolymorphofcalcitestableonlyat lowtemperature.Thisindicatesthatnotallsphere crystalhabitscanbeattributedtobiogenicoriginfor calcite,asinCavernedel’Ours,theformationof cryogeniccalciteispurelyabiotic. A CKNOWLEDGEMENTS ThisworkwassupportedbyaCanadianSpaceAgency internalresearchfundtoDLandNaturalSciencesand EngineeringResearchCouncilofCanada(NSERC)grants toBLandIDC.WewouldliketothankR.Mineau,W. AbdyandP.Middlesteadfortheirtechnicalassistancein thelaboratories.M.S.Field(editor),I.Sasowsky(associate editor),K.Zak,andananonymousrefereeprovided helpfulreviewsofthemanuscript. R EFERENCES Albright,J.N.,1971,Vateritestability:AmericanMineralogy,v.56, p.620–624. Braissant,O.,Cailleau,G.,Dupraz,C.,andVerrecchia,E.P.,2003, Bacteriallyinducedmineralizationofcalciumcarbonateinterrestrial environments:Theroleofexopolysaccharidesandaminoacids: JournalofSedimentaryResearch,v.73,p.485–490. Clark,I.D.,andLauriol,B.,1992,Kineticenrichmentofstableisotopes in cryogeniccalcite:ChemicalGeology,v.102,p.217–228. Clark,I.D.,andLauriol,B.,1997,AufeisoftheFirthRiverbasin, NorthernYukon,Canada:Insightsintopermafrosthydrogeologyand karst:ArcticandAlpineResearch,v.29,p.240–252. Courty,M.A.,Marlin,C.,Dever,L.,Tremblay,P.,andVachier,P.,1994, Theproperties,genesisandenvironmentalsignificanceofcalcitic pendantsfromthehighArctic(Spitsbergen):Geoderma,v.61, p.71–102. Dansgaard,W.,1961,Theisotopiccompositionofnaturalwaters: MeddelelseromGronland,v.165,p.1–120. Dondale,C.D.,Redner,J.H.,Paquin,P.,andLevi,H.W.,2003,The InsectsandArachnidsofCanada:Part23,TheOrb-weavingSpiders ofCanadaandAlaska.Ottawa,Canada,NationalResearchCouncil ofCanada,371p. Dresser,J.A.,andDenis,T.C.,1946,Lage ologiedeQue bec,ge ologie descriptive:Quebec,Canada,Ministe `redesMines,Que bec,Rapport ge ologique20,647p. Elver,M.S.,1994,Thestratigraphicandisotopiccharacteristicsofan ArcticIcing,BylotIsland,N.W.T.[M.Sc.thesis]:Ottawa,Canada, CarletonUniversity,http://www.climate.weatheroffice.ec.gc.ca/cl imate_ normals/ Fairchild,I.J.,Bradly,L.,andSpiro,B.,1993,Carbonatediagenesisin ice: Geology,v.21,p.901–904. Fairchild,I.J.,Killawee,J.A.,Spiro,B.,andTison,J.-L.,1996,Calci te precipitatesformedbyfreezingprocesses:kineticcontrolson morphologyandgeochemistry, in Bottrell,S.,ed.,Proceedingsof thefourthInternationalSymposiumontheGeochemistryofthe Earth’sSurface,Ilkley,Yorkshire,Wiley,p.178–183. Folk,R.L.,1993,SEMimagingofbacteriaandnanobacteriaincarbonate sedimentsandrocks:JournalofSedimentaryPetrology,v.63, p.990–999. Ford,D.C.,andWilliams,P.,2007,Karsthydrogeologyandgeomorphology:WestSussex,U.K.,Wiley,576p. Grasby,S.E.,2003,Naturallyprecipitatingvaterite( `-CaCO 3 )spheres: unusualcarbonatesformedinanextremeenvironment:Geochimicaet CosmochimicaActa,v.67,p.1659–1666. Hagen-Thorn,A.,Callesen,I.,Armolaitis,K.,andNihlga rd,B.,2004, TheimpactofsixEuropeantreespeciesonthechemistryofmineral topsoilinforestplantationsonformeragriculturalland:Forest, EcologyandManagement,v.195,p.373–384. Hall,D.K.,1980,MineralprecipitationinNorthSloperivericings:Arct ic, v.33,p.343–348. Hallet,B.,1976,DepositsformedbysubglacialprecipitationofCaCO 3 : GeologicalSocietyofAmericaBulletin,v.87,p.1003–1015. Heldmann,J.L.,Pollard,W.H.,McKay,C.P.,Andersen,D.T.,andToon, O.B.,2005,Annualdevelopmentcycleofanicingdepositand associatedperennialspringactivityonAxelHeibergIsland,Canadian HighArctic:Arctic,AntarcticandAlpineResearch,v.37,p.127–135. IAEA/WMO(InternationalAtomicEnergyAgency/WorldMeterology Organization),2004,GlobalNetworkofIsotopesinPrecipitation. TheGNIPDatabase,http://isohis.iaea.org Jouzel,J.,andSouchez,R.A.,1982,Melting—refreezingattheglacierso le andtheisotopiccompositionoftheice:JournalofGlaciology,v.28, p.35–42. Killawee,J.A.,Fairchild,I.J.,Tison,J.L.,Janssens,L.,andLorrain, R., 1998,Segregationofsolutesandgasesinexperimentalfreezingof dilutesolutions:implicationfornaturalglaciersystems:Geochimicae t CosmochimicaActa,v.62,p.3637–3655. Kim,S.T.,andO’Neil,J.R.,1997,Equilibriumandnonequilibrium oxygenisotopeeffectsinsyntheticcarbonates:Geochimicaet CosmochimicaActa,v.61,p.3461–3475. Kralj,D.,Brecevic,L.,andNielson,A.E.,1990,Vateritegrowthand dissolutioninaqueoussolution:1.Kineticsofcrystalgrowth:Journal ofCrystalGrowth,v.104,p.793–800. Kretz,R.,1980,Occurrence,mineralchemistryandmetamorphismof PrecambriancarbonaterocksinaportionoftheGrenvilleProvince: JournalofPetrology,v.21,p.573–620. Kretz,R.,2001,OxygenandcarbonisotopiccompositionofGrenville marble,andanappraisalofequilibriuminthedistributionofisotopes betweencalciteandassociatedminerals,OtterLake,Quebec,Canada: TheCanadianMineralogist,v.39,p.1455–1472. Lacelle,D.,2007,Environmentalsetting,(micro)morphologies,andsta ble C-Oisotopecompositionofcoldclimatecarbonateprecipitates–a reviewandevaluationoftheirpotentialaspaleoclimaticindicators: QuaternaryScienceReviews,v.26,p.1670–1989. Lacelle,D.,Lauriol,B.,andClark,I.D.,2004,Seasonalisotopicimprin t inmoonmilkfromCavernedel’Ours:implicationsforclimatic reconstruction:CanadianJournalofEarthSciences,v.41, p.1411–1423. D.L ACELLE ,B.L AURIOL AND I.D.C LARK JournalofCaveandKarstStudies, April2009 N 61

PAGE 15

Lacelle,D.,Lauriol,B.,andClark,I.D.,2006,Effectofchemical compositionofwaterontheoxygen-18andcarbon-13signature preservedincryogeniccarbonates,ArcticCanada:implicationsin paleoclimaticstudies:ChemicalGeology,v.234,p.1–16. Majoube,M.,1971,Fractionnementenoxygene-18etendeuteriumentre l’eauetsavapeur.JournalofChemicalPhysics,v.197,p.1423–1436. Marion,G.M.,2001,Carbonatemineralsolubilityatlowtemperaturesin theNa-K-Mg-Ca-H-Cl-SO 4 -OH-HCO 3 -CO 3 -CO 2 -H 2 Osystem:GeochimicaetCosmochimicaActa,v.65,p.1883–1896. Marlin,C.,Dever,L.,Vachier,P.,andCourty,M-A.,1993,Variations chimiquesetisotopiquesdel’eaudusollorsdelarepriseengeld’une coucheactivesurperge lisolcontinu(Presqu’iledeBrogger,Svalbard): CanadianJournalofEarthSciences,v.30,p.806–813. Murase,N.,Ruike,H.,Matsunaya,N.,Hagakawa,M.,Kaneko,Y.,and Ono,Y.,2001,Spidersilkhasanicenucleationactivity:Naturwissenschaften,v.88,p.117–118. Omelon,C.R.,Pollard,W.H.,andMarion,G.M.,2001,Seasonal formationofikaite(CaCO 3 6H 2 O)insalinespringdischargeat ExpeditionFjord,CanadianHighArctic:assessingconditional constraintfornaturalcrystalgrowth:GeochimicaetCosmochimica Acta,v.65,p.1429–1437. Pauly,H.,1963,Ikaite,anewmineralfromGreenland:Arctic,v.16, p.263–264. Pre vost,C.,andLauriol,B.,1994,Variabilite del’e rosionactuelleet Holocene:lecasdesmarbresdeGrenvilleenOutaouaisQue be cois: Ge ographiePhysiqueetQuaternaire,v.48,p.297–303. Pollard,W.,1983,Astudyofseasonalfrostmounds,NorthForkPass, northerninteriorYukonTerritory[Ph.D.thesis]:Ottawa,Canada, UniversityofOttawa. Richter,D.K.,andRiechelmann,D.F.C.,2008,LatePleistocenecryogeni c calcitespherolitesfromtheMalachitdomCave(NERhenishSlate Mountains,Germany):origin,unusualinternalstructureandstable C-Oisotopecomposition:InternationalJournalofSpeleology,v.37, p.119–129. Rowe,J.S.,1972.ForestRegionsofCanada.EnvironmentofCanada, CanadianForestryService,PublicationNo.1300,172p. Silk,S.T.,1970,Factorsgoverningpolymorphformationincalcium carbonate[PhDthesis]:NewYork,NewYorkUniversity. Souchez,R.A.,andJouzel,J.,1984,Ontheisotopiccompositionin d D and d 18 Oofwaterandiceduringfreezing:JournalofGlaciology, v.30,p.369–372. Souchez,R.A.,anddeGrotte,J.M.,1985, d Dd 18 Orelationshipsinice formedbysubglacialfreezing:paleoclimaticimplications:Journalof Glaciology,v.109,p.599–602. Souchez,R.A.,Jouzel,J.,Lorrain,R.,Sleewaegen,S.,Stie venard,M.,and Verbeke,V.,2000,Akineticisotopeeffectduringiceformationby waterfreezing:GeophysicalResearchLetters,v.27,p.1923–1926. Suess,E.,Balzer,W.,Hesse,K.-F.,Muu ¨ller,P.J.,Ungerer,C.A.,and Wefer,G.,1982,Calciumcarbonatehexahydratefromorganic-rich sedimentsoftheantarcticshelf:Precursorsofglendonites:Science, v.216,p.1128–1131. Turnbull,A.G.,1973,Athermodynamicstudyofvaterite:Geochimicaet CosmochimicaActa,v.37,p.1593–1601. Vecht,A.,andIreland,T.G.,2000,Theroleofvateriteandaragonitein theformationofpseudo-biogeniccarbonatestructures:implications formartianexobiology:GeochimicaetCosmochimicaActa,v.64, p.2719–2725. White,A.F.,Bullen,T.D.,Vivit,D.V.,Schultz,M.S.,andClow,D.W., 1999,Theroleofdisseminatedcalciteinthechemicalweathering ofgranitoidrocks:GeochimicaetCosmochimicaActa,v.63, p.1939–1953. Zak,K.,Urban,J.,Cilek,V.,andHercman,H.,2004,Cryogenic cavecalcitefromseveralCentralEuropeancaves:age,carbonand oxygenisotopesandageneticmodel:ChemicalGeology,v.206, p.119–136. Zak,K.,Onac,B.P.,andPersoiu,A.,2008,Cryogeniccarbonatesincave environments:areview:QuaternaryInternational,v.187,p.84–96. F ORMATIONOFSEASONALICEBODIESANDASSOCIATEDCRYOGENICCARBONATESIN C AVERNEDEL ’O URS ,Q UE BEC ,C ANADA :K INETICISOTOPEEFFECTS ANDPSEUDO BIOGENICCRYSTALSTRUCTURES 62 N JournalofCaveandKarstStudies, April2009



PAGE 1

WAVELETANALYSISOFLATEHOLOCENESTALAGMITE RECORDSFROMORTIGOSACAVESINNORTHERNSPAIN A.M UN OZ 1 ,A.K.S EN 2 ,C.S ANCHO 1 AND D.G ENTY 3 Abstract: Wehavededucedshort-termclimaticchangesfrommillennialtoannual scalesfromthestudyoflaminaethicknessandradiocarbonanalysisofHol ocene stalagmiterecordsfromtwocavesinOrtigosadeCameros(IberianRange,n orthern Spain).Speleothemsaremadeupofdarkcompactlaminae(DCL)andwhitepor ous laminae(WPL)ofseasonalorigin.Coupletsseasonalityisdeducedfrommo nitoring calcitelaminaegrowth,dripwaterrates,andsoilorganicmatterflushed intothecaves. Thethicknessvariationsofthecoupletsareanalyzedusingacontinuousw avelet transformandthevariousperiodicitiesatinterannual,decadal,multid ecadal,and centennialscalesarerevealedfromthewaveletpowerspectrum.Theperio dicitiesat decadal,multidecadalandcentennialscales,withperiodsaround9.7,10 .4,14,16,22,43, 73,83and180years,aremainlyrelatedtosolaractivity.Amongtheintera nnual periodicities,oscillationsaroundthe2.4-yr-periodmaybelinkedtoth eQuasi-Biennial Oscillation(QBO),whereasperiodsrangingfrom4to7yearsmaybeassocia tedwiththe ElNin o-SouthernOscillation(ENSO)and/ortheNorthAtlanticOscillation(N AO). I NTRODUCTION TheunderstandingofHolocenepaleoclimaticevolution inthenorth-centralIberianPeninsulahassignificantly improvedduringthepastfewyearsfromthestudiesof lacustrine(Luque,2003),tufaceous(Sanchoetal.,1997), fluvial(ThorndycraftandBenito,2006),alluvial(Sanchoet al.,2008),andslope(Gutie rrezetal.,2006)records.In general,fromtheanalysisofthereporteddata,ithasbeen possibletodeduceaprevailingclimatewithhighvariability andmillennialscalecycles.Asaconsequence,alink betweenNorthAtlanticcirculationandweatherinIberia duringtheHoloceneperiodhasbeenestablished.However, thecontributiontoclimatevariabilitymadefromspeleothemrecordsstillremainsatapreliminarylevel(Dura n etal.,2000;Labonneetal.,2002;Mun oz-Garc aetal., 2004;2007;Mart n-Chiveletetal.,2006). Speleothemsarewidelyusedforpaleoclimaticand paleoenvironmentalreconstructionsessentiallybecause theycanbewelldatedandtheirisotopiccompositions recordchangesintemperature,rainfall,andvegetation-soil activity(McDermott,2004).Inaddition,growthpatterns ofspeleothemscanbeusedtoestablishpaleoclimatic sequencesmadeupofcycleswithvariablefrequency.From radiometricagesandstableisotopeanalysisofspeleothems,severalstudieshavedetectedshort-termorhigh frequencyclimaticchangesatregional/localscale(Dorale etal.,1992;Frumkinetal.,1999;Burnsetal.,2001). Furthermore,veryhighfrequencyclimaticcycleshavebeen identifiedbyperformingstatisticalanalysisof(1)laminaetedstructureofthestalagmites(Quinifetal.,1994; GentyandQuinif,1996;Mingetal.,1998;Qinetal.,1999; Frisiaetal.,2003;Soubie `setal.,2005),and(2) d 18 Ohigh resolutionrecords(Niggemannetal.,2003;Holzka ¨mperet al.,2004;Dykoskietal.,2005). Inthispaperweexaminestalagmiterecordsfromtwo cavesinOrtigosadeCameros(LaRioja,Spain).Weused radiometricchronologicaldatatodeduceshort-term environmentalchanges.Wealsousewaveletanalysisof thelaminaetedstructureofthestalagmitestodeducevery highfrequencyperiodicities.Asaresultofthisanalysis,the presentstateofknowledgeofenvironmentalchanges duringHolocenetimes,inthenorthernsectorofthe IberianPeninsula,canbesignificantlyimproved.Preliminaryinformationaboutpaleoenvironmentalmeaningof thespeleothemrecordsfromOrtigosaCaveshasbeen reportedbyMun ozetal.(2001,2004). S TUDY A REA Thestudyareaislocatedinthewesternmostsectorof theCamerosRange(IberianMountainSystem,Northern Spain(Fig.1a).Thegeologicalframeworkismadeupofa PaleozoicbasementsurroundedbyaMesozoicstratigraphicsequencedippingtotheS-SE.Specifically,theOrtigosa cavesystemiscomposedof185-m-thickMiddleJurassic limestonesofahighenergyshallowshelfsequence(ITGE, 1990). LaPazandLaVin aCavesarelocatedintheEncinedo MountainnearOrtigosadeCamerosvillage(LaRioja). Themeanannualtemperatureis9 u Candthemeanannual precipitationis630mm.TheOrtigosacavesystemisoneof themostimportantendokarsticfeaturesintheIberian Range.LaVin aCaveis114-m-longandislocatedata 1 DepartamentodeCienciasdelaTierra,UniversidaddeZaragoza,50009Zar agoza, Spain,armunoz@unizar.escsancho@unizar.es 2 DepartmentofMathematicalSciences,IndianaUniversity,402N,Blackfo rd Street,Indianapolis,IN46202,USA,asen@iupui.edu 3 LaboratoiredesSciencesduClimatetdel’Environnement,LSCEUMRCEA/ CNRS,91191GifsurYvettecedex,France,dominique.genty@lsce.ipsl.fr A.Mun oz,A.K.Sen,C.Sancho,andD.Genty–WaveletanalysisofLateHolocenest alagmiterecordsfromOrtigosaCavesin NorthernSpain. JournalofCaveandKarstStudies, v.71,no.1,p.63–72. JournalofCaveandKarstStudies, April2009 N 63

PAGE 2

lowerelevationwithanaltitudeofabout1080mabovethe sealevel.LaPazCaveislonger(236m),anditislocated 20mabovetheformer.BothLaPazCaveandLaVin a Caveexhibithorizontaldevelopmentandageometrythat iscontrolledbythedirectionofthemainregionalNE-SW faultsystem(Fig.1b). M ATERIALAND M ETHODOLOGY TheOrtigosaCavescontainalargevarietyof speleothems.Threespeleothemdevelopmentstageshave beendifferentiatedafteradetailedmorphostratigraphic andchronologicalanalysis(Mun ozetal.,2001).Theoldest periodbeganmorethan400,000Maandisrepresentedby flowstonesthatareverywellexposedinLaVin aCave.The intermediatestageisassociatedwithagreaterdevelopment ofstalagmitesinbothcaves.ItincludesIsotopicStages7, 5,and3separatedbyperiodsofinactivity.Thefinal developmentstageisrelatedtosmallerstalagmitesand correspondstoIsotopicStage1.Inthisstudy,weusetwo stalagmitedepositsfromLaPazCave(LP-4)andLaVin a Cave(LV-1)relatedtotheyoungeststage,Holoceneinage. ThestalagmitefromtheLaVin aCavewasstillactivewhen samplesweretaken(Fig.2). Thesampledstalagmiteswerecutalongtheirgrowth axisshowingaverywell-markedinternallybanded structurecharacterizedbyanalternationofwhiteporous laminae(WPL)anddarkcompactlaminae(DCL)(Figs.2 and3a,b,c).Thisterminology(WPL/DCL)hasbeenused byQuinifetal.(1994)andGentyetal.(1997b).An Figure1.(a)Locationandgeologicalsettingofthe OrtigosaCaves.(b)NE-SWsectionoftheEncinedo MountainshowingthemorphologicalfeaturesoftheCaves. TheasteriskinLaPazCaveillustratesthepositionofthe OR-P1artificialcarbonateplate(Fig.3b,c,d).Theasterisk inLaVin aCaveshowsthelocationoftheHOBORG3Data LoggingRainGaugetomeasurethedripwaterflowrates (seeFig.4). Figure2.LaminatedstructureoftheLV-1stalagmitefrom LaVin aCave,andoftheLP-4stalagmitefromLaPaz Cave.LocationsofsamplesusedinU/Thdatingand radiocarbon-AMSanalysisaredenotedbyrectanglesand dots,respectively. W AVELETANALYSISOF L ATE H OLOCENESTALAGMITERECORDSFROM O RTIGOSA C AVESIN N ORTHERN S PAIN 64 N JournalofCaveandKarstStudies, April2009

PAGE 3

elongatedcolumnarorfibrousfabricwasthemost prevalentfabricidentifiedinthesestalagmitesaccording tothedifferentiationmadebyFrisiaetal.(2000). Oneofthetwosymmetricalportionswasusedtoselect samplesinordertocarryoutmineralogicalstudies,aswell asradiometricdating.X-raydiffractionanalysisdata indicatethatthestalagmitedepositsarelargelymadeup oflow-Mgcalcite.TheaveragemolarratioofMgCO 3 in calciteis0.51%anditsmaximumvaluereaches1.05%. SeveralsamplesweredesignedtoestimatetheradiometricagesbyusingU/ThisotopicratiosandradiocarbonAMStechniques.TheUandThisotopicratioswere determinedbyalphaspectrometryandtheactivitieswere calibratedbyadditionofknownquantitiesofartificial radioactivespikes( 232 U228 Thinradioactiveequilibrium). ChemicalpreparationwascarriedoutattheIsotopic GeochemistryLaboratoryoftheCentredÂ’Etudesetde RecherchesApplique esauKarstdelaFaculte PolytechniquedeMons(Belgium).The 14 C-AMSageshavebeen correctedforanarbitrarydeadcarbonproportion(also calleddcpwherethecarboncomesmainlyfromthe limestonedissolutionandis 14 Cfree)of10%,whichisan averagevalueforseveralEuropeansites(Gentyetal., 1997b,1998,2001;GentyandMassault,1999).Theages havebeencalibratedusingdendrochronologicalandcoral curves(StuiverandKra,1986;INTCAL04,Reimeretal., 2004).ThesecalibrationswerecarriedoutattheHydrology andIsotopicGeochemistryLaboratoryoftheUniversite deParis-SudandTANDETRON,CNRS-CEA,UMS T2004,GifsurYvette(France). Thesecondsymmetricalportionofthestalagmiteswas carefullypolishedtocarryoutaspectral-temporalanalysis oftheinternallaminaetionusingwavelets.First,the sampleswerephotographedandtheimageswerestored inadigitalformat.Thenthelaminaethicknesswas measuredusingthesoftwareOPTIMAV5attheInstituto JaumeAlmera(Barcelona).Subsequently,weapplied waveletanalysisonthethicknessvariationdatatodetect veryhighfrequencyperiodicities.Inaddition,todetectthe variousperiodicitiesintheinternallaminaetions,wavelet analysiscandelineatethetimeintervalsoverwhichthese periodicitiespersist.Wavelet-basedmethodshavebeen usedforsignalanalysisinawidevarietyofapplications (Addison,2001;SenandDostrovsky,2007;Senetal., 2008a,b)includinganalysisofspeleothemrecords (Holmgrenetal.,2003;Lachnietetal.,2004;Tanetal., 2006).Thevariousperiodicitiesarediscernedfromthe waveletpowerspectrumofthethicknessvariations.Abrief descriptionofthewaveletanalysismethodologyisgivenin theAppendix. Inordertotesttheseasonalgrowthpatternofthe laminae,onMarch27,2003acarbonateplate(25 3 15 3 2cm)(OR-P1)wasplacedunderadripwaterpointto samplethepresentdaystalagmitegrowthinLaPazCave. ItwasremovedonNovember8,2007andwasstudiedby usingapetrographicmicroscopeincorporatingfluoresFigure3.(a)AnnualmicrosequencefromtheLV-1stalagmiteshowingthewhiteporouslamina(WPL)anddark compactlamina(DCL).(b)Generalviewofstalagmite growingovertheOR-P1carbonateplatelocatedinLaPaz Cave(betweenMarch27,2003andNovember8,2007). Darkcompactlamina(DCL)wouldcorrespondtothe increaseofwaterdriprateinwinters(seeFig.4).Nonpolarized(c)andfluorescence(d)photographsofthe stalagmiteareaindicatedinb).Theluminescencelines (DCL)correlatewiththetrendsinseasonalwaterexcess whichflushesorganicmaterialsthroughtheaquifertothe speleothem(Bakeretal.,1997,1999;Gentyetal.,1997a). A.M UN OZ ,A.K.S EN ,C.S ANCHO AND D.G ENTY JournalofCaveandKarstStudies, April2009 N 65

PAGE 4

cenceanalysis.Theimagesobtainedwithnon-polarized lightshowcoupletsofdarkandclearlaminaerelatedto DCL/WPLpairs(Fig.3b).Infact,fiveclearlaminae (WPL)andfourdarklaminae(DCL)wereidentified. Luminescenceintensityreflectsorganicmatterconcentration(Bakeretal.,1997)andwasobservedtocorrelate withthetrendsinseasonalwaterexcess(Gentyetal., 1997a).Inordertoobtainafluorescenceimagefromthe stalagmitegrowingoverthecarbonateplate(Fig.3d),we usedanexcitationwavelengthof450–480nm.Itwas filteredbyU-MWBfilterandthefluorescenceemission wasdetectedat515–700nm.Finally,dripwaterflowinside thecaveswascontinuouslyrecordedbyusingaHOBO RG3DataLoggingRainGaugefromSeptember22,2005 inordertoknowthehydrologicalresponseofthekarstic systemtoseasonallycontrolledexternalrainfallevents. R ESULTS C HRONOLOGY BasedonpreliminarymorphostratigraphicarrangementofspeleothemsintheOrtigosaCaves,aswellason radiometricages(Mun ozetal.,2001),stalagmitesLP-4 andLV-1havebeenassociatedwiththeIsotopicStage1. TheU/Thagesarefoundtobe23.2( 3.0/ 2.9)kyBP (bottomLV-1)and7.6( 0.7/ 0.7)kyBP(middleLP-4) (Table1).However,allthesamplesshowaverylow 230 Th/ 232 Thisotopicratio(3 6 0.5forLP-4and4 6 1for LV-1)whichmeansthattheyarecontaminatedwith detritalTh( 232 Th).ThereforetheU/Thagesobtainedare uncertainandanycorrectionisunreliableaswedon’t knowtheinitial 230 Th/ 232 Thratio.Moreover,the23.2ky BPdatingfromLV-1stalagmiteoccursduringaverycold period(LastGlacialMaximum)whenthegrowthof speleothemswasnotlikely.Asaconsequence,this chronologicalresultshouldbeusedwithcaution. InordertoimprovetheestimationofU/Thageofthe stalagmiteLV-1,threeadditionalanalysesbyradiocarbonAMSwerecarriedout.Theobtainedages(Table2)ofthe differentsamplesfrombottomtotopare3.4 6 0.6kyBP, 1.2 6 0.4kyBP,and0.7 6 0.35kyBP,respectively. Despitetheuncertaintiesduetotheunknowndcp,these resultsdemonstratethatthestalagmiteisLateHolocenein age. S TALAGMITE L AMINATION F REQUENCY A NALYSIS Theinternalalternatingstructureofwhiteporous laminae(WPL)anddarkcompactlaminae(DCL)inthe stalagmitesamples(Figs.2and3)maybeseasonally controlledand,asaconsequence,thepairedmicrosequence maybeannuallydeveloped(Bakeretal.,1993;Quinifet al.,1994;Railsbacketal.,1994;Shopovetal.,1994;Genty andQuinif,1996;Gentyetal.,1997b;Bakeretal.,1998). Inaddition,Mitchell(1976)deducedthatthemagnitudeof theseasonalcyclepowerisoneordergreaterthanany othercycle-generatingmechanism. Inordertovalidatetheseasonalcontrolonstalagmite laminaetioninOrtigosaCaves,weusepresentdaykarstic activityincaves,basedonboththehydrologicalresponse modelandthestalagmitegrowthpattern.Byusingthe stalagmitegrownontheartificialtablet(OR-P1,Fig.3b), itcanbededucedthatthefirstclearlaminaedevelopedon thesamplewouldcorrespondto2003spring/summerand thefirstdarklaminaeto2004winter.Thelastlaminae correspondsto2007spring/summerperiod.Onthebasisof thewater-dripratesinsidethecave(Fig.4),weinterpret thedarkcompactlaminae(DCL)wouldformatthe momentofamoreintensedripping(winter),whereasthe clear(white)porouslaminae(WPL)wouldformduringthe slowdrippingthattakesplaceduringspringandsummer. GentyandQuinif(1996)indicatethattheprecipitationof theWPLinthemicrosequencemostlikelytookplace duringsummerandisrelatedtoalowwaterexcessanda moreregularandchemicallymoreefficientflowrate,while theDCLwasprobablyformedduringthewinterseason,is relatedtoahighwaterexcess,andachemicallyless efficientwaterflow. Ontheotherhand,themostprominentfluorescence emissionintheOR-P1sampleoccursinthedarkcompact laminae(Fig.3c,d).Thiswouldalsoindicatethattheyhave beenformedduringwinterswhensoilorganicmatteris flushedintothecave.Thewhiteporouslaminaewouldform laterinthehydrologicalcyclewhenthedriprateislower. Withinthispremise,thestalagmitesamplesLV-1and LP-4aremadeupof1276and638annualcycles, respectively.InthesampleLV-1,chronologicallyanalyzed byradiocarbon,900annualcycleshavebeencounted between3400 6 600yrBPand1200 6 400yrBP,326 cyclesbetween1200 6 400yrBPand700 6 350yrBP,and 50cyclesbetween700 6 350yrBPandpresenttime. Waveletanalysisofthecoupletthicknessvariationsof theLP-4stalagmitelaminaetionsfromtheLaPazcave indicatestheoccurrenceofveryhighfrequencyclimatic cycles.Figure5adepictsthetimeseriesofthickness variationsintheLP-4stalagmitelaminaetions.Thewavelet powerspectrum(WPS)ofthistimeseriesisshownin Table1.AnalyticaldataofU/Thdating. Sample No.Lab.No.[U] ppm 234 U/ 238 U 230 Th/ 234 U 230 Th/ 232 Th[ 234 U/ 238 U] t 0 Age(kyBP) LP-461730.043 6 0.0011.100 6 0.0270.068 6 0.0063.0 6 0.51.1027.6[ 0.7/ 0.7] LV-161740.031 6 0.0011.148 6 0.0290.193 6 0.0224 6 11.15823.2[ 3.0/ 2.9] W AVELETANALYSISOF L ATE H OLOCENESTALAGMITERECORDSFROM O RTIGOSA C AVESIN N ORTHERN S PAIN 66 N JournalofCaveandKarstStudies, April2009

PAGE 5

Figure5b.InFigures5aand5bthehorizontalaxisis labeledasthenumberofannualcycles.InFigure5b,the darkcontourlinesrepresent95%confidencelevelwith respecttorednoisespectrum,andtheareabelowthethin U-shapedcurvedenotestheconeofinfluence(see Appendixfordetails).Severalperiodicities,andthetime intervals(i.e.,thenumberofannualcycles)overwhichthe periodicitiespersist,canbediscernedfromFigure5b.For example,thereisastrongperiodicbandaroundthe9.7-yrperiod.Thisbandpersistsoverthetimeintervalbetween 250and285annualcycles.Asimilarperiodicbandis presentaroundthe10.4-yrperiod.Anotherperiodicband isfoundaroundthe43-yrperiodspanningapproximately thetimeintervalbetween160to260annualcycles.Wealso observetime-varyingperiodicitieswithperiodsfrom22to 31years.Inaddition,severalveryhighfrequencycyclesin the2–4-yrbandareseeninFigure5b.Theseperiodicities appearinanintermittentfashion.Figure5cdepictsthe globalwaveletspectrum(GWS)oftheLP-4stalagmite thicknesstimeseries.Thedominantspectralmodescanbe identifiedfromthevariouspeaksinFigure5c(see Appendixfordetails). TheresultsofwaveletanalysisoftheLV-1stalagmite fromLaVin acaveareshowninFigure6.Figure6a,b, andcillustrate,respectively,thetimeseriesofthickness variations,waveletpowerspectrum,andglobalwavelet spectrum.Thefollowingperiodicitiescanbeobservedin Figure6b.Thereisastrongperiodicbandaroundthe180yr-periodpersistingcontinuouslyovertheintervalfrom 260to775annualcycles.Therearealsostrongperiodic bandsaroundthe73-yrand83-yrperiodspersisting continuouslyoverseveralannualcycles.Inaddition,we observeoscillationsaroundthe14-yrand16-yrperiods. Thesebandspersistapproximatelyover46annualcycles. Severalveryhigh-frequencyperiodicitieswithpeaksatthe 2.4-yr,4.0-yr,and5.6-yrperiodsarealsoseeninFigure6. Theseperiodicitiesappearinanintermittentpattern.The dominantspectralmodescanalsobeidentifiedfromthe globalwaveletspectrumofthetimeseriesshownin Figure6c. Table2.Analyticaldataofradiocarbon-AMSdating. SampleNo.Lab.No.DistancetoTop(mm)Age(yearsBP) LV1-4PA764/H241020700 LV1-3PA763/H24092101200 LV1-1PA762/H24015403400 Figure4.Drip-waterflowinLaVin acave(September 2005–September2007).Thedripwatershowsaninitial intenseflowgeneratingthedarksheetoftheseasonalcycle andanotherstageofmoreslowandeffectivedrip-waterflow thatgeneratestheclearsheet. Figure5.(a)TimeseriesofthethicknessvariationsinLP-4 stalagmitefromLaPazcave.(b)Waveletpowerspectrumof thetimeseriesofLP-4stalagmitethicknessvariationsshown in(a).Thedarkcontourlinesrepresent95%confidencelevel withrespecttoarednoisebackgroundandtheareabelowthe thinU-shapedcurvedenotestheconeofinfluence(COI). InsidetheCOI,theedgeeffectsmaybecomeimportantand theresultsshouldbeusedwithcaution(TorrenceandCompo, 1998).(c)Globalwaveletspectrumofthetimeseriesshown in(a). A.M UN OZ ,A.K.S EN ,C.S ANCHO AND D.G ENTY JournalofCaveandKarstStudies, April2009 N 67

PAGE 6

D ISCUSSION S HORT -T ERM E NVIRONMENTAL C HANGES Takingintoaccountthechronologicaldataobtained fromLP-4andLV-1stalagmites,weproposethatthe youngestmorphostratigraphicstalagmitegrowthperiodin OrtigosaCavesmaybeLateHoloceneinageandisrelated tothewarmIsotopicStage1atglobalscale(Henningetal., 1983).TheoccurrenceofthisperiodontheIberian PeninsulaandBalearsIslandsisshowninthegeochronologicaldatingscenariocompiledbyDura n(1989).This Holocenewarmperioddeducedfromstalagmitescanbe relatedtoothermorphosedimentaryrecordsintheIberian Range.Radiometricagesofregionalfluvialtufadeposits indicateanextensiveperiodoftufabuildingduringwarm IsotopicStage1(Mart nez-Tudelaetal.,1986;Ordo n ezet al.,1990;Sanchoetal.,1997;Pen aetal.,2000). Thisclimaticperiodbasedonspeleothemgrowthcould becorrelatedwiththepaleoenvironmentalevolutionin highlatituderegions(seeforexample,Gordonetal.,1989). However,inadditiontothetemperature,rainfallchanges mustbeconsideredasanimportantfactorinthe speleothemdevelopmentstagesduetothelocationofthe IberianPeninsulanearthelowlatitudearidbelts(Brooket al.,1990;Bar-Matthewsetal.,1996,1997). Fromtheradiocarbonages,itispossibletoestablish thattheLV-1stalagmitestartedtogrow3400 6 600Ma andfinishedatpresent-daytime.Ontheotherhand,1276 annualcycleshavebeenrecordedfromtheinternal laminaetionsofthestalagmite.Asaconsequence,itis likelythateitherwemissedalotoflaminaeduring countingbecausetheywerenotvisible,ortherearegrowth hiatuses.AdescriptionoftheLateHolocenespeleosequencecanbeproposedconsideringthat900annualcycles havebeencountedbetween3400 6 600yrBPand1200 6 400yrBP,326cyclesbetween1200 6 400yrBPand700 6 350yrBP,and50cyclesbetween700 6 350yrBPand presenttime.Inspiteofthehighchronologicaluncertainty, thesedataindicateanimportantlackofcyclesbetween 3400 6 600yrBPand1200 6 400yrBP,andalsobetween 700 6 350yrBPandpresenttime.Tentatively,wepropose tocorrelatetheoccurrenceofgrowthhiatusesinstalagmitesfromtheOrtigosakarstsystemwiththeIronAge ColdPhase(thecoldestmaximumisat2700–2500yrBP) andwiththeLittleIceAge(XVI-XIXcenturies)(Mun ozet al.,2001;Pen aetal.,2004).However,atthistimeandwith thecurrentlyavailablechronologicaldata,itisnotpossible tolocateexactlybothactivityandinactivityperiods. Similarshort-termclimaticrecords(10 2 –10 3 years)have beendeducedusingHolocenestalagmitesinIsrael(Frumkinetal.,1999),SouthAfrica(Repinskietal.,1999),Oman (Burnsetal.,2001),andGermany(Niggemannetal., 2003). HighclimatevariabilityduringLateHolocenehasbeen proposedbydifferentauthorsatregionalscalebyusing differentmorphosedimentaryrecordsandmultiproxydata (Pen albaetal.,1997;Perrette,1999;Sa nchezGon iand Hannon,1999;Luque,2003;Gonza lez-Sempe rizetal., 2006;Mart n-Chiveletetal.,2006;Thorndycraftand Benito,2006;Luzo netal.,2007;Vegas,2007;Sanchoet al.,2008).HoloceneclimaticchangesinIberia,and particularlyvariationsinrainfall,areconnectedwith large-scaleatmosphericprocessessuchastheNorth AtlanticOscillation(NAO)(Zoritaetal.,1992;Trigoet al.,2004).Acloserelationshiphasalsobeenproposed betweenNorthAtlanticOscillationandsolaractivity (Luque,2003). V ERY H IGH F REQUENCY C LIMATIC C YCLES Someoftheshorttermandveryhighfrequency periodicitiesrevealedbywaveletanalysiscanberelated tosolaractivitycycles,aswellastonaturalclimatic oscillations,suchasElNin o-SouthernOscillation(ENSO), NorthAtlanticOscillation(NAO),Quasi-BiennialOscillation(QBO),etc.(O’Sullivanetal.,2002;Burroughs,2003). Climaticcyclescorrespondingtothesolaractivity(1–100 years)havebeenrecognizedinHoloceneaswellas Pleistocenestalagmitesbyseveralauthorsandindifferent partsoftheworld(Bakeretal.,1993;Shopovetal.,1994; Gentyetal.,1994;GentyandQuinif,1996;Qinetal.,1999; Niggemannetal.,2003;Frisiaetal.,2003;Holzka ¨mperet al.,2004;Dykoskietal.,2005;Soubie `setal.,2005). First,cycleswithstrongdecadalscaleperiodicities(e.g., 9.7-yrand10.4-yr-periodsinLP-4,and14-yrperiodinLV1stalagmites)canbeassociatedwiththesunspotcycle(9– 14years).The14-yr-periodisrelatedtolakedryingphases inGallocanta(IberianRange)andisinterpretedas influencingofthelow-frequencycomponentofENSO (Rodo etal.,1997).Gentyetal.(1994)andGentyand Figure6.(a)TimeseriesofthethicknessvariationsinLV-1 stalagmitefromLaVin acave.(b)Waveletpowerspectrum ofthetimeseriesofLV-1stalagmitethicknessvariations shownin(a).Thedarkcontourlinesandtheconeofinfluence havethesamemeaningasinFigure5(b).(c)Globalwavelet spectrumofthetimeseriesshownin(a). W AVELETANALYSISOF L ATE H OLOCENESTALAGMITERECORDSFROM O RTIGOSA C AVESIN N ORTHERN S PAIN 68 N JournalofCaveandKarstStudies, April2009

PAGE 7

Quinif(1996)foundasimilarperiodicityinaPleistocene stalagmitefromBelgium.Thisperiodicityisalsoobserved byBakeretal.(1993)andShopovetal.(1994)intheirstudies ofthelaminaetedstructureinstalagmitesusingultraviolet light.Spectralanalysisperformedonhighresolution d 18 O recordsbyNiggemannetal.(2003),Holzka ¨mperetal.(2004) andDykoskietal.(2005),andonstalagmitelaminae thicknessrecords(Qinetal.,1999;Frisiaetal.,2003; Soubie `setal.,2005)showstrongperiodicitiesatsubdecadal anddecadalscales.The22-yr-periodmaybeassociatedwith theHalecycleofsunspotactivity. Ontheotherhand,the4–7-year-periodicbandcan berelatedtoENSOaswellasNAO.Similarperiodicities inHolocenestalagmitesaredescribedbyGentyetal. (1994),Frisiaetal.(2003),Dykoskietal.(2005),Rasbury andAharon(2006),andSoubie `setal.(2005),among others. Thepresenceofveryhighfrequencyperiodicities aroundthe2.4-yr-periodcanberelatedtotheQuasiBiennialOscillation(QBO).Thisclimaticcycleismainly observedinrainfallandtemperaturerecordsandhasbeen identifiedindifferentvarvedsediments(Sonettetal.,1992; Mun ozetal.,2002).Oscillationswiththisperiodhavebeen foundinstalagmitelaminaebyQinetal.(1999),andFrisia etal.(2003). Finally,periodicitiesatdecadal,multidecadal,and centennialscalesthatareobservedinOrtigosaCave stalagmitesandrelatedtosolaractivity(suchasthe Gleissbergcycle)havealsobeenfoundinothercave deposits(seeQinetal.,1999;Niggemannetal.,2003; Holzka ¨mperetal.,2004,andDykoskietal.,2005). ThewaythatthesolarforcingdrivesEarth’sclimate systemhasbeenfullydiscussed(Friis-Christensenand Lassen,1991;HoytandSchatten,1997;KellyandWigley, 1992;SchlesingerandRamankutty,1992;Schonwieseet al.,1994;VanGeeletal.,1999;ScafettaandWest2006, 2007).Theclimaticinformationprovidedbythespeleothemcycleswouldbetranslatedintoavariationofthe cyclethicknessduetoamoreorlessintensedevelopment ofthesoil-vegetationsystemoverthecave.Duringwarm andhumidclimaticperiods,theintensesoilactivityfavors agreateramountofpercolatingwaterwithhighconcentrationinbiogenicCO 2 inthekarstsystem.Theresultisa thickerannualcyclethroughyearswithhighertemperature (morebiogenicCO 2 andrainfall)andthinnerinyearswith lowertemperature. Inthesameway,acloserelationhasbeenobserved betweenNAOandthedistributionofwinterrainfallinthe IberianPeninsula.PeriodswiththeNAOinanegative phaseareassociatedwithwetconditionsinthewestern MediterraneanandnorthernAfrica(Wanneretal.,1994) whereasperiodswithpositiveNAOarerelatedtodroughts intheIberianPeninsula.Moreover,Rodo etal.(1997) relatewarmperiodsofENSOtoperiodsofreducedrainfall andhighertemperaturesintheeasternhalfofSpain. AlthoughtherelationshipbetweenNAOandENSOevents andatmosphericteleconnectionsinEurope,andspecificallySpain,isnotclear,itispossibletocorrelatehigher growthratesintheOrtigosaspeleothemswithNAO negativephases.Thesimplerelationshipbetweengrowth ratesintheOrtigosaCavesandsolarforcingandNAO phasesmaybecomemorecomplexwiththeoccurrenceof ENSOteleconnections. C ONCLUSIONS Basedonthestudiesofchronologicalcontrolandthe waveletanalysisofthicknessvariationsofstalagmite depositsintheOrtigosaCaves(LaPazandLaVin a),we havededucedenvironmentalandclimaticchangeswith multipleperiodicitiesduringLateHolocene.Fromour analysis,thefollowingconclusionscanbemade. 1.LateHoloceneperiod(from4000yrBPtothepresent) isaveryactivestageofspeleothemgrowthinthe northernsectoroftheIberianPeninsula. 2.Short-term(10 2 –10 3 years)climaticchangescanbe deducedfromradiocarbon-AMSdatingcombined withannuallydevelopedlaminaenumber.However, withtheavailablechronologicaldata,itisnot possibletoestablishaprecisesequenceofactivity andinactivityperiodsoncentennial-millennialtime scales. 3.ThestalagmitesbelongingtotheLateHolocenestage havedevelopedabandedstructurecharacterizedbyan alternatingsequenceoflightanddarklaminae,which constituteaseasonallycontrolledmicrosequence.The waveletpowerspectrumanalysisofthethickness variationsinthestalagmitesindicatesclimaticcyclesat centennial,decadal,multidecadal,andinterannual scaleswithperiodsaround2.4,4–7,9.7,10.4,14,16, 22,43,73,83and180years. 4.Veryhighfrequencyclimaticchangesobtainedfrom waveletanalysisarerelatedtoastronomicaland atmosphericcontrols.Highergrowthratesinthe Ortigosaspeleothemscouldberelatedtowarmand humidclimaticperiodsasaconsequenceofhighsolar activityandnegativephasesofNAO. A CKNOWLEDGEMENTS ThisworkhasbeensupportedbytheLaRioja GovernmentandOrtigosadeCamerosTownCouncil. WethankProfessorIvesQuinif(Faculte Polytechniquede Mons)forhiscollaborationonU/Thdating,andDr. AntonioVa zquez(InstitutoJaumeAlmeraC.S.I.C., Barcelona)forhissupportonimageanalysis.Thisisa contributionbythePaleoQandCuencasSedimentarias Continentalesgroups(Arago nRegionalGovernment).We thankDr.IraD.Sasowsky(AssociateEditor)andtwo anonymousreviewersforvaluablesuggestions. A.M UN OZ ,A.K.S EN ,C.S ANCHO AND D.G ENTY JournalofCaveandKarstStudies, April2009 N 69

PAGE 8

R EFERENCES Addison,P.S.,2001,Theillustratedwavelettransformhandbook,Bristo l, U.K.,InstituteofPhysicsPublishing,353p. Baker,A.,Smart,P.L.,Edwards,R.L.,andRichards,D.A.,1993,Annual growthbandinginacavestalagmite:Nature,v.364,p.518–520,doi: 10.1038/364518a0. Baker,A.,Barnes,W.L.,andSmart,P.L.,1997,Variationsinthe dischargeandorganicmattercontentofstalagmitedripwatersin LowerCave:HydrologicalProcesses,v.11,p.1541–1555. Baker,A.,Genty,D.,Dreybrodt,W.,Barnes,W.L.,Mockler,N.J.,and Grapes,J.,1998,Testingtheoreticallypredictedstalagmitegrowthrat e withrecentannuallylaminaetedsamples:implicationsforpaststalagmi te deposition:GeochimicaetCosmochimicaActa,v.62,p.393–404. Baker,A.,Mockler,N.J.,andBarnes,W.L.,1999,Fluorescenceintensity variationsofspeleothem-forminggroundwaters:Implicationsfor paleoclimatereconstruction:WaterResourcesResearch,v.35, no.2,p.407–413. Bar-Matthews,M.,Ayalon,A.,Matthews,A.,Sass,E.,andHalicz,L., 1996,Carbonandoxygenisotopestudyoftheactivewater-carbonate systeminakarsticMediterraneancave:implicationsforpaleoclimate researchinseimiaridregions:GeochimicaetCosmochimicaActa, v.60,p.337–347. Bar-Matthews,M.,Ayalon,A.,andKaufman,A.,1997,LateQuaternary paleoclimateintheEasternMediterraneanregionfromstableisotope analysisofspeleothemsatSoreqCave,Israel:QuaternaryResearch, v.47,p.155–168. Brook,G.A.,Burney,D.A.,andCowart,J.B.,1990,Desertpaleoenvironmentaldatafromcavespelothemswithexamplesfromthe Chihuahuan,Somali-Chalbi,andKalaharideserts:Palaeogeography, Palaeoclimatology,Palaeoecology,v.76,p.311–329. Burns,S.J.,Fleitmann,D.,Matter,A.,Neff,U.,andMangini,A.,2001, SpeleothemevidencefromOmanforcontinentalpluvialeventsduring interglacialperiods:Geology,v.29,p.623–626. Burroughs,W.,2003,Weathercycles:Realorimaginary?:Cambridge, U.K.,CambridgeUniversityPress,317p. Dorale,J.A.,Gonzalez,L.A.,Regan,M.K.,Pickett,D.A.,Murrel,M.T., andBaker,R.G.,1992,AhighresolutionrecordofHoloceneclimate changeinspeleothemcalcitefromColdWaterCave,NortheastIowa: Science,v.258,p.1626–1630. Dura n,J.J.,1989,Geocronolog adelosdepo sitosasociadosalkarsten Espan a, in Dura n,J.J.,andLo pez,J.,eds.,ElkarstenEspan a. Monograf a,4,SociedadEspan oladeGeomorfolog a,p.243–256. Dura n,J.J.,Lo pez,J.,Dallai,L.,Bruschi,G.,Caballero,E.,Jime nez,C., andJulia,R.,2000,Paleoenvironmentalreconstructionbasedona detailedstableisotopeanalysisanddatingofaHolocenespeleothem fromValporqueroCave,NorthernSpain:Geogaceta,v.27,p.63–67. Dykoski,C.A.,Edwards,R.L.,Cheng,H.,Yuan,D.,Cai,Y.,Zhang,M., Lin,Y.,Qing,J.,An,Z.,andRevenaugh,J.,2005,Ahigh-resolution, absolute-datedHoloceneanddeglacialAsianmonsoonrecordfrom DonggeCave,China:EarthandPlanetaryScienceLetters,v.233, p.71–86. Friis-Christensen,E.,andLassen,K.,1991,Lengthofthesolarcycle:an indicatorofsolaractivitycloselyassociatedwithclimate:Science, v.254,p.698–700. Frisia,S.,Borsato,A.,Fairchild,I.J.,andMcdermott,F.,2000,Calcit e fabrics,growthmechanismsandenvironmentsofformationin speleothemsfromtheItalianAlpsandSouthwesternIreland:Journal ofSedimentaryResearch,v.70,p.1183–1196. Frisia,S.,Borsato,A.,Preto,N.,andMcDermott,F.,2003,Late HoloceneannualgrowthinthreeAlpinestalagmitesrecordsthe influenceofsolaractivityandtheNorthAtlanticOscillationonwinter climate:EarthandPlanetaryScienceLetters,v.216,p.411–424. Frumkin,A.,Ford,D.C.,andSchwarcz,P.,1999,Continentaloxygen isotopicrecordofthelast170,000yearsinJerusalem:Quaternary Research,v.51,p.317–327. Genty,D.,andMassault,M.,1999,Carbontransferdynamicsfrom bomb14 Cand d 13 CtimeseriesofalaminaetedstalagmitefromSWFrance:Modellingandcomparisonwithotherstalagmite:Geochimica etCosmochimicaActa,v.63,p.1537–1548. Genty,D.,andQuinif,Y.,1996,Annuallylaminaetedsequencesinthe internalstructureofsomeBelgianstalagmites-Importancefor paleoclimatology:JournalofSedimentaryResearch,v.66, p.275–288. Genty,D.,Quinif,Y.,andDeflandre,G.,1994,Microsequencesdes laminesannuellesdansdeuxstalagmitesdumassifdeHan-Sur-Lesse (Belgique):Spelochronos,v.6,p.9–22. Genty,D.,Baker,A.,andBarnes,W.,1997a,Comparaisonentreles laminesluminescentesetleslaminesvisiblesannuellesdestalagmites: Comptesrendusdel’Acade miedessciences,Se rie2,v.325,no.3, p.193–200. Genty,D.,Deflandre,G.,Quinif,Y.,andVerheyden,S.,1997b,Les laminesdecroissancedesspeleothemes:origineetinteretpaleoclimatique:BulletindelaSocie te BelgedeGe ologie,v.106,p.63–77. Genty,D.,Vokal,B.,Obelic,B.,andMassault,M.,1998,Bomb14Ctime historyrecordedintwomodernstalagmites:Importanceforsoil organicmatterdynamicsandbomb14Cdistributionovercontinents: EarthandPlanetaryScienceLetters,v.160,p.795–809. Genty,D.,Baker,A.,Massault,M.,Proctor,Ch.,Gilmour,M.,PonsBranchu,E.,andHamelin,B.,2001,Deadcarboninstalagmites: Carbonatebedrockpaleodissolutionvs.ageingofsoilorganicmatter: Implicationsfor13Cvariationsinspeleothems:Geochimicaet CosmochimicaActa,v.65,no.15,p.3443–3457. Gonza lez-Sempe riz,P.,Valero-Garce s,B.L.,Moreno,A.,Jalut,G., Garc a-Ruiz,J.M.,Mart -Bono,C.,Delgado-Huertas,A.,Navas,A., Otto,T.,andDedoubat,J.J.,2006,ClimatevariabilityintheSpanish Pyreneesduringthelast30,000yrrevealedbytheElPortalet sequence:QuaternaryResearch,v.66,p.38–52. Gordon,D.,Smart,P.L.,Ford,D.C.,Andrews,J.N.,Atkinson,T.C., Rowe,P.,andChristopher,N.S.J.,1989,DatingoflatePleistocene InterglacialandInterstadialperiodsintheUnitedKingdomfrom speleothemsgrowthfrequency:QuaternaryResearch,v.31,p.14–26. Gutie rrez,M.,Gutie rrez,F.,andDesir,G.,2006,Considerationsonthe chronologicalandcausalrelationshipsbetweentalusflatironsand paleoclimaticchangesincentralandnortheasternSpain:Geomorphology,v.73,p.60–63. Henning,G.J.,Grun,R.,andBrunnacker,K.,1983,Speleothems, TravertinesandPaleoclimates:QuaternaryResearch,v.20,p.1–29. Holmgren,K.,Lee-Thorp,J.A.,Cooper,G.R.J.,Lundblad,K.,Partridge, T.C.,Scott,L.,Sithaldeen,R.,Talma,A.S.,andTyson,P.D.,2003, Persistentmillennial-scaleclimatevariabilityoverthepast25,000 yearsinSouthernAfrica:QuaternaryScienceReviews,v.22, p.2311–2326. Holzka ¨mper,S.,Mangini,A.,Spo ¨tl,C.,andMudelsee,M.,2004,Timing andprogressionoftheLastInterglacialderivedfromahighalpine stalagmite:GeophysicalResearchLetters,v.31,L07201,doi:10.1029/ 2003GL019112. Hoyt,D.V.,andSchatten,K.H.,1997,TheRoleoftheSuninClimate Change,Oxford,OxfordUniversityPress,397p. ITGE,1990,Mapageolo gicodeEspan a.Escala1:50.000.2 a Serie(241) Anguiano,Madrid,ServiciodePublicacionesdelMinisteriode IndustriayEnerg a. Kelly,P.M.,andWigley,T.M.L.,1992,Solarcyclelength,green-house forcingandglobalclimate:Nature,v.360,p.328–330. Labonne,M.,Hillaire-Marcel,C.,Ghaleb,B.,andGoy,J.L.,2002,Multiisotopicageassessmentofdirtyspeleothemcalcite:anexamplefrom AltamiraCave,Spain:QuaternaryScienceReviews,v.21,p.1099–1110. Lachniet,M.,Burns,S.J.,Piperno,D.R.,Asmerom,Y.,Polyak,V.J., Moy,C.M.,andChristenson,K.,2004,A1500-yearElNino/Southern oscillationandrainfallhistoryfortheisthmusofPanamafrom speleothemcalcite:JournalofGeophysicalResearch,v.109,D20117, doi:10.1029/2004JD004694. Luque,J.A.,2003,LagodeSanabria:unsensordelasoscilaciones clima ticasdelAtla nticoNortedurantelosu ltimos6.000an os[PhD thesis],Barcelona,UniversitatdeBarcelona,384p. Luzo n,A.,Pe rez,A.,Mayayo,M.J.,Soria,A.R.,Sa nchez-Gon i,M.F., andRoc,A.C.,2007,Holoceneenvironmentalchangesinthe Gallocantalacustrinebasin,IberianRange,NESpain:Holocene, v.17,p.649–663. Mart n-Chivelet,J.,Mun oz-Garc a,M.B.,Dom nguez-Villar,D.,Turrero, M.J.,andOrtega,A.I.,2006,Comparativeanalysisofstalagmites fromtwocavesofnorthernSpain:ImplicationsforHolocene paleoclimatestudies:GeologicaBelgica,v.9,no.3–4,p.323–335. Mart nez-Tudela,A.,Cuenca,F.,Santisteban,C.,Grun,R.,and Hentzsch,B.,1986,LostravertinosdelR oMatarran a,Beceite (Teruel)comoindicadorespaleoclima ticosdelCuaternario, in Lo pezVera,F.,ed.,QuaternaryClimateinWesternMediterranean, p.307–324. W AVELETANALYSISOF L ATE H OLOCENESTALAGMITERECORDSFROM O RTIGOSA C AVESIN N ORTHERN S PAIN 70 N JournalofCaveandKarstStudies, April2009

PAGE 9

McDermott,F.,2004,Paleoclimatereconstructionfromstableisotope variationsinspeleothems:Areview:QuaternaryScienceReviews, v.23,p.901–918. Ming,T.,Tungsheng,L.,Xiaoguang,Q.,andXianfeng,W.,1998, Significationchronoclimatiquedespe le othe `meslamine sdeChinedu Nord:Karstologia,v.32,p.1–6. Mitchell,J.M.,1976,Anoverviewofclimaticvariabilityanditscausal mechanism:QuaternaryResearch,v.6,p.481–493. Mun oz,A.,Pen a,J.L.,Sancho,C.,andMart nez,M.A.,2001,Losespeleotemas delascuevasdeOrtigosadeCameros(LaRioja):Datoscronolo gicosy consideracionespaleoambientales:Geogaceta,v.30,p.95–98. Mun oz,A.,Ojeda,J.,andSanchez-Valvede,B.,2002,Sunspots-likeand ENSO/NAO-likeperiodicitiesinlacustrinelaminaetedsedimentsof thePlioceneVillarroyaBasin(LaRioja,Spain):Journalof Paleolimnology,v.27,no.4,p.453–463. Mun oz,A.,Sancho,C.,Pen a,J.L.,Sa nchez-Valverde,B.,Valero-Garce s, B.,Dura n,J.J.,andGenty,D.,2004,Ciclosclima ticosdealta frecuentaenlosespeleotemasdelasCuevasdeOrtigosadeCameros (LaRioja), in Liesa,C.,Pocov ,A.,Sancho,C.,Colombo,F., Gonza lez,A.,andSoria,A.R.,eds.,Geo-Temas,v.6,no.5, p.141–144(VICongresoGeolo gicodeEspan a). Mun oz-Garc a,M.B.,Mart n-Chivelet,J.,andRossi,C.,2004,Implicacionespaleoclima ticasdeladistribucio ngeocronolo gicadeespeleotemasenlaCuevadelCobre(Palencia):Geogaceta,v.35,p.179–182. Mun oz-Garc a,M.B.,Mart n-Chivelet,J.,Rossi,C.,Ford,D.C.,and Schwarcz,H.P.,2007,ChronologyofTerminationIIandtheLast InterglacialPeriodinNorthSpainbasedonstableisotoperecordsof stalagmitesfromCuevadelCobre(Palencia):JournalofIberian Geology,v.33,no.1,p.17–30. Niggemann,S.,Mangini,A.,Mudelsee,M.,Richter,D.K.,andWurth, G.,2003,Sub-MilankovitchclimaticcyclesinHolocenestalagmites fromSauerland,Germany:EarthandPlanetaryScienceLetters, v.216,p.539–547. O’Sullivan,P.E.,Moyeed,R.,Cooper,M.C.,andNicholson,M.J.,2002, Comparisonbetweeninstrumental,observationalandhighresolution proxysedimentaryrecordsofLateHoloceneclimaticchange—a discussionofpossibilities:QuaternaryInternational,v.88,p.27–44. Ordo n ez,S.,Gonza lez,J.A.,andGarc aDelCura,M.A.,1990,Datacio n radioge nica(U-234/U-238yTh-230/U-234)desistemastravert nicos delAltoTajo(Guadalajara):Geogaceta,v.8,p.53–56. Pen a,J.L.,Sancho,C.,andLozano,M.V.,2000,Climaticandtectonic significanceofLatePleistoceneandHolocenetufadepositsinthe Mijaresrivercanyon(EasternIberianRange,NESpain):Earth SurfaceProcessesandLandforms,v.25,p.1403–1417. Pen a,J.L.,Julia n,A.,Chueca,J.,Echeverr a,M.T.,andA ngeles,G.,2004, Etapasdeevolucio nholocenaenelvalledelr oHuerva:Geomorfolog a yGeoarqueolog a, in Pen a,J.L.,Longares,L.A.,andSa nchez,M.,eds., Geograf aF sicadeArago n:aspectosgeneralesytema ticos,UniversidaddeZaragoza-Institucio nFernandoelCato lico,p.289–302. Pen alba,M.C.,Arnold,M.,Guiot,J.,Duplessy,J.C.,anddeBeaulieu, J.L.,1997,TerminationoftheLastGlaciationintheIberianPen nsula inferredfromthepollensequenceofQuintanardelaSierra: QuaternaryResearch,v.48,p.205–214. Perrette,Y.,1999,Lesstalagmites:archivesenvironnementaleset climatiquesa `hautere solution.Pre sentationdesprotocolesd’e tude etpremiersre sultatssurlesspe le othe `mesduVercors:Karstologia, v.34,no.2,p.23–44. Qin,X.,Tan,M.,Liu,T.,Wang,X.,Li,T.,andLu,J.,1999,Spectral analysisofa1000-yearstalagmitelaminae-thicknessrecordfrom ShihuaCavern,Beijing,China,anditsclimaticsignificance:Holocene, v.9,p.689–694. Quinif,Y.,Genty,D.,andMaire,R.,1994,Lesspe le othe `mes:unoutil perfomantpourlese tudespale oclimatiques:BulletinSociete Ge ologiqueFrance,v.165,p.603–612. Railsback,L.B.,Brook,G.A.,Chen,J.,Kalin,R.,andFleisher,J.,1994, EnvironmentalcontrolsonthepetrologyofalateHolocene speleothemfromBotswanawithannuallayersofaragoniteand calcite:JournalofSedimentaryResearch,v.A64,p.147–155. Rasbury,M.,andAharon,A.,2006,ENSO-controlledrainfallvariability recordsarchivedintropicalstalagmitesfromthemid-oceanislandofNiu e, SouthPacific:Geochemistry,Geophysics,Geosystems,v.7,p.1–15. Reimer,P.J.,Baillie,M.G.L.,Bard,E.,Bayliss,A.,Beck,J.W.,Bertran d, C.,Blackwell,P.G.,Buck,C.E.,Burr,G.,Cutler,K.B.,Damon,P.E., Edwards,R.L.,Fairbanks,R.G.,Friedrich,M.,Guilderson,T.P., Hughen,K.A.,Kromer,B.,McCormac,F.G.,Manning,S.,Bronk Ramsey,C.,Reimer,R.W.,Remmele,S.,Southon,J.R.,Stuiver,M., Talamo,S.,Taylor,F.W.,VanderPlicht,J.,andWeyhenmeyer,C.E., 2004,IntCal04terrestrialradiocarbonagecalibration,26-0kaBP: Radiocarbon,v.46,p.1029–1058. Repinski,P.,Holmgren,K.,Lauritzen,S.E.,andLee-Thorp,J.A.,1999,A lateHoloceneclimaterecordfromastalagmite,ColdAirCave, Northernprovince,SouthAfrica:Palaeogeography,Palaeoclimatology,Palaeoecology,v.150,p.269–277. Rodo ,X.,Baert,E.,andCom n,F.A.,1997,Variationsinseasonal rainfallinSouthernEuropeduringthepresentcentury:relationships withtheNorthAtlanticOscillationandtheElNin o-Southern Oscillation:ClimateDynamics,v.13,p.275–284. Sa nchezGon i,M.F.,andHannon,G.E.,1999,High-altitudevegetational patternontheIberianMountainChain(north-centralSpain)during theHolocene:Holocene,v.9,p.39–57. Sancho,C.,Pen a,J.L.,andMele ndez,A.,1997,ControlsonHoloceneand present-daytravertineformationintheGuadalaviarRiver(Iberian Chain,NESpain):ZeistchriftfurGeomorphologie,v.41,p.289–307. Sancho,C.,Pen a,J.L.,Mun oz,A.,Benito,G.,McDonald,E.,Rhodes,E., andLongares,L.A.,2008,Holocenealluvialmorphopedosedimentary recordandenvironmentalchangesintheBardenasRealesNatural Park(NESpain):Catena,v.73,p.225–238. Scafetta,N.,andWest,B.J.,2006,Phenomenologicalsolarsignaturein4 00 yearsofreconstructedNorthernHemispheretemperaturerecord:GeophysicalResearchLetters,v.33,L17718,doi:10.1029/2006GL027142. Scafetta,N.,andWest,B.J.,2007,Phenomenologicalreconstructionsof thesolarsignatureintheNorthernHemispheresurfacetemperature recordssince1600:JournalofGeophysicalResearch,v.112,D24S03, doi:10.1029/2007JD008437. Schlesinger,M.E.,andRamankutty,N.,1992,Implicationsforglobal warmingofintercyclesolarirradiancevariations:Nature,v.360, p.330–333. Schonwiese,C.D.,Ullrich,R.,Beck,F.,andRapp,J.,1994,Solarsignals inglobalclimaticchange:ClimateChange,v.27,p.259–281. Sen,A.K.,andDostrovsky,J.O.,2007,Evidenceofintermittencyinthe localfieldpotentialsrecordedfrompatientswithParkinson’sdisease —Awavelet-basedapproach:ComputationalandMathematical MethodsinMedicine,v.8,p.165–171. Sen,A.K.,Longwic,R.,Litak,G.,andGorski,K.,2008a,Analysisof cycle-to-cyclepressureoscillationsinadieselengine:Mechanical SystemsandSignalProcessing,v.22,p.362–373. Sen,A.K.,Filippelli,G.,andFlores,J.,2008b,Anapplicationofwavele t analysistopaleoproductivityrecordsfromtheSouthernOcean: ComputersandGeosciences(inpress). Shopov,Y.Y.,Ford,D.C.,andSchwarcz,H.P.,1994,Luminescent microbandinginspeleothems:high-resolutionchronologyandpaleoclimate:Geology,v.22,p.402–410. Sonett,C.P.,Williams,C.R.,andMo ¨rner,N.A.,1992,TheFourier spectrumofSwedishriverinevarves:evidenceofsub-arcticquasibiennial(QBO)oscillations:Palaeogeography,Palaeoclimatology, Palaeoecology,v.98,p.57–65. Soubie `s,F.,Seidel,A.,Mangin,A.,Genty,D.,Ronchail,J.,Plagnes,V., Hirooka,S.,andSantos,R.,2005,Afifty-yearclimaticsignalinthree HolocenestalagmiterecordsfromMatoGrosso,Brazil:Quaternary International,v.135,p.115–129. Stuiver,M.,Kra,R.S.,andeditors,1986,Calibrationissue,Proceeding softhe 12thInternational14CConference:Radiocarbon,v.28,p.805–1030. Tan,M.,Baker,A.,Genty,D.,Smith,C.,Esper,J.,andCai,B.,2006, Applicationsofstalagmitelaminaetopaleoclimatereconstructions: Comparisonwithdendrochronology/climatology:QuaternaryScience Reviews,v.25,p.2103–2117. Thorndycraft,V.R.,andBenito,G.,2006,TheHolocenefluvial chronologyofSpain:Evidencefromanewlycompiledradiocarbon database:QuaternaryScienceReviews,v.25,p.223–234. Torrence,C.,andCompo,G.P.,1998,Apracticalguidetowaveletanalysis : BulletinoftheAmericanMeteorologicalSociety,v.79,p.61–78. Trigo,R.M.,Pozo,D.,Osborne,T.,Castro,Y.,Ga miz,S.,andEsteban, M.J.,2004,NorthAtlanticOscillationinfluenceonprecipitation,rive r flowandwaterresourcesintheIberianPeninsula:International JournalofClimatology,v.24,p.925–944. VanGeel,B.,Raspopov,O.M.,Renssen,H.,VanderPlicht,J.R., Dergachev,V.A.,andMeijer,H.A.J.,1999,Theroleofsolarforcing uponclimatechange:QuaternaryScienceReview,v.18,p.331–338. A.M UN OZ ,A.K.S EN ,C.S ANCHO AND D.G ENTY JournalofCaveandKarstStudies, April2009 N 71

PAGE 10

Vegas,J.,2007,Caracterizacio ndeeventosclima ticosdelPleistoceno superior-Holocenomedianteelestudiosedimentolo gicodelaLaguna Grande(SierradeNeila,NOSistemaIbe rico):RevistadelaSociedad Geolo gicadeEspan a,v.20,p.53–70. Wanner,H.,Brazdil,R.,Frich,P.,Frydendahl,K.,Jonsson,T.,Kington, J.A.,Pfister,C.,Rosenorn,S.,andWishman,E.,1994,Synoptic interpretationofmonthlyweathermapsforthelateMaunder Minimum(1675–1704), in Frenzel,B.,Pfister,C.,andGlaser,B., eds.,ClimaticTrendsandAnomaliesinEurope,Stuttgart,Gustav FischerVerlag,p.401–424. Zorita,E.,Kharin,V.,andvonStorch,H.,1992,Theatmospheric circulationandseasurfacetemperatureintheNorthAtlanticareain winter:TheirinteractionandrelevanceforIberianprecipitation: JournalofClimate,v.5,p.1097–1108. A PPENDIX W AVELET A NALYSIS M ETHODOLOGY Awaveletisasmallwavewithzeromeanandfinite energy.Thecontinuouswavelettransform(CWT)ofa function x ( t )withrespecttoawavelet y ( t )isgivenbythe convolutionofthefunctionwithascaledandtranslated versionof y ( t ).Thefunction y ( t )isreferredtoasan analyzingwaveletoramotherwavelet.Theconvolutionis expressedbytheintegral: Ws t ? ? xt y s t t dt 1 where y s t ( t ) 1 s y t t s 2 isascaledandtranslatedversionofthemotherwavelet y ( t )andanasteriskon y denotesitscomplexconjugate. Thesymbols s and t arethescaleparameterand translationparameters,respectively.Thescaleparameter controlsthedilation( s 1)andcontraction( s 1)ofthe motherwavelet.Thefactor1 s isintroducedinEquation (2)sothatthefunction y s, t ( t )hasunitenergyateveryscale. Thetranslationparameter t indicatesthelocationofthe waveletintime;inotherwords,as t varies,thesignalis analyzedinthevicinityofthispoint.Theamountofsignal energycontainedataspecificscale s andlocation t isgiven bythesquaredmodulusoftheCWT, P ( s t ) W ( s t ) 2 Foratimeseries x i with i 1,2,3,…, N ,theintegral formulationshowninEquation(1)canbediscretizedas (TorrenceandCompo,1998): W n s N n 1 d t s 1 2 x n y n n s 3 Here n isthetimeindex, s isthewaveletscale,and d t isthe samplinginterval.Thewaveletpowerspectrum(WPS)of thetimeseriesisdefinedby W n ( s ) 2 ,whichisameasureof thefluctuationofthevarianceatdifferentscalesor frequencies.Thispowerspectrum,whichdependsonboth scaleandtime,isrepresentedbyasurface.Byplotting contoursofthissurfaceonaplane,atime-scalerepresentationofthespectrummaybederived. Atime-scalerepresentationisfoundtobeusefulfor extractingimportantfeaturesofsignalsarisinginmany applications.Analternaterepresentation,namely,atimefrequencyrepresentation,hasalsobeenused.Ascale-tofrequencyconversion,whichfollowsareciprocalrelationship,canbeeasilymadebyuseoftheformula, f f 0 f s where f istheinstantaneousfrequencyofthesignal, f is thesamplingfrequency,and f 0 isthecenterfrequencyof themotherwavelet(seebelow).Inouranalysis,weuseda complexMorletwaveletasthemotherwavelet.Acomplex Morletwaveletconsistsofaplanewavemodulatedbya Gaussianfunctionandisdescribedby yg p 1 4 e i v 0 g e g 2 2 4 where v 0 2 p f 0 ,istheorderofthewavelet,with f 0 beingthecenterfrequency.Inourcomputationswe haveusedaMorletwaveletoforder6asthemother wavelet.Thischoiceprovidesagoodbalancebetween timeandfrequencylocalizations.Forthischoice,the scaleisalsoapproximatelyequaltotheFourierperiod andthusthetermsscaleandperiodcanbeused interchangeably. Thewaveletpowerspectrum(WPS)displaysacontour plotofpowerasafunctionofscale(period)andtimeandis sometimesreferredtoasalocalwaveletspectrum. Additionalinformationaboutthespectralpropertiesof thetimeseriescanbeobtainedbyaveragingtheWPSat eachscaleoveralltime,andtherebycalculatingtheglobal waveletspectrum(GWS).TheGWSisgivenby W 2 S 1 N N n 1 W n ( s ) 2 5 (TorrenceandCompo,1998).TheGWSdisplayspoweras afunctionofperiodorfrequencyandissimilartoa smoothedFourierpowerspectrum.Thedominantspectral modesofthetimeseriescanbeidentifiedfromthevarious peaksintheGWS. W AVELETANALYSISOF L ATE H OLOCENESTALAGMITERECORDSFROM O RTIGOSA C AVESIN N ORTHERN S PAIN 72 N JournalofCaveandKarstStudies, April2009



PAGE 1

LIMITATIONSOFHENDYTESTCRITERIAINJUDGING THEPALEOCLIMATICSUITABILITYOFSPELEOTHEMS ANDTHENEEDFORREPLICATION J EFFREY A.D ORALE 1 AND Z AIHUA L IU 2,3, Abstract: Carbonandoxygenisotopesincalcitespeleothemsarepowerfulproxiesfo r understandingpastclimatechange.Forcalcitedepositedunderisotopic equilibrium conditions,variationsin d 18 Ovaluesdirectlyreflectchangesincavetemperatureandthe isotopiccompositionofmeteoricwater.Speleothem d 13 Cvalueshavebedrock, atmospheric,andsoilgassources.Soilgasescanbetracedtotheoverlyin gvegetation, whichisrelatedtoclimate.Both d 13 Cand d 18 Ovaluesarethereforepotentiallypowerful tracersofclimatechange.Processesthatcouldalterspeleothem d 13 Cand/or d 18 Ovalues, andtherebymaskprimaryenvironmentalsignals,fallinthecategoriesof 1)kinetic processes,includingdepositionofcalciteoutofisotopicequilibrium, and2)vadose processes,includingevaporationofwateratornearthelandsurface.Int ruth,thereisno absolutetestfortheabsenceofthesekinetic/vadose-zoneprocesses.Ho wever,theHendy Testiswidelyusedforassessingwhetherisotopicequilibriumexisteddu ringthetimeof calcitedeposition.Criterion(1)oftheHendyTest(i.e.,that d 18 Ovaluesremainconstant alongasinglegrowthlayer)maynotbeavalidcontrolofequilibriumcondi tionsbecause isotopicequilibriumcouldtheoreticallyoccurinthecenterofthespele othematthesame timethatkineticfractionationoccursattheflanks.Moreover,theconce ptofsampling alongasinglegrowthlayerisflawedinboththeoryandpractice.Criterio n(2)ofthe HendyTest(i.e.,thatthereisnorelationshipbetween d 13 Cand d 18 O)isbasedonthe assumptionthatspeleothem d 13 Cvaluesarenotlinkedtoclimate.However,speleothem d 13 Cvaluesmaywellbelinkedtoclimatebecauseclimateprovidesafirst-ord ercontrol onsoilproductivityandthetypeofvegetation.Therefore,HendyTestcri terion(2)isnot aprerequisitetoisotopicequilibriuminallcases.Weproposeinsteadth eReplication Test(i.e.,thedemonstrationofsimilarisotopicprofilesamongtwoormo respeleothems) forevaluatingthelikelihoodofcalcitedepositionunderisotopicequil ibriumconditions. Replicationofisotopicprofilesamongtwoormorespeleothemsispossibl eonlyif kinetic/vadose-zoneprocessesareeither:1)absentor2)haveaffecteds patiallyseparated speleothemsinexactlythesameway.Becausethesecondscenarioishighly unlikely,we proposethattheReplicationTestiseffectivelysufficientinrulingout kinetic/vadosezoneoverprintingprocesses.WefurthersuggestthattheReplicationTes tisfarmore robustintestingfortheabsenceofthewiderangeofprocessesdescribeda bovethanis thetraditionalHendyTest. I NTRODUCTION Carbon-andoxygen-isotopicvariationsinspeleothems havebeenstudiedformorethan40years(Galimovand Grinenko,1965;HendyandWilson,1968;Duplessyetal., 1970),commonlywiththecentralgoalofreconstructing pastenvironmentaland/orclimaticconditions.Inaseminal paper,Hendy(1971)outlinedthevariousequilibriumand nonequilibriumprocessesthatgovern d 18 Oand d 13 Cvalues duringcalcitespeleothemformationanddiscussedthe meansofrecognizingwhetherspeleothemstableisotopic signaturesmightserveasappropriatepaleoclimaticindicators. ContinuedrefinementsinU/Thdatingtechniques (Edwardsetal.,1987;Shenetal.,2002;Doraleetal., 2004;Hellstrom,2006)andconstructionofhigh-resolution d 18 Otimeseriescomparabletoicecorerecords(BarMatthewsetal.,1997;Doraleetal.,1998;Wangetal., 2001;Yuanetal.,2004;Spo ¨tletal.,2006;Wangetal., 2008;Zhangetal.,2008)havegeneratedasurgeofrenewed interestinusingspeleothemsashigh-resolutionpaleoclimaticarchives.Thus,thefidelitywithwhich d 18 Oand d 13 C variationsinspeleothemsmightbeinterpretedaspaleoclimaticindicatorsremainsabasicandcriticalissue.This *Correspondingauthor.Tel:86-851-5892338;Fax:86-851-5891721;E-ma il addresses:liuzaihua@vip.gyig.ac.cn(Z.Liu) 1 DepartmentofGeoscience,UniversityofIowa,Iowa52242-1379,USA 2 TheStateKeyLaboratoryofEnvironmentalGeochemistry,Instituteof Geochemistry,ChineseAcademyofSciences,550002Guiyang,China 3 TheKeyLaboratoryofKarstDynamics,MinistryofLandandResources, InstituteofKarstGeology,ChineseAcademyofGeologicalSciences,5410 04Guilin, China J.A.DoraleandZ.Liu–Limitationsofhendytestcriteriainjudgingthepa leoclimaticsuitabilityofspeleothemsandtheneedfor replication. JournalofCaveandKarstStudies, v.71,no.1,p.73–80. JournalofCaveandKarstStudies, April2009 N 73

PAGE 2

paperconsidersthewidelyusedHendyTestmethodfor evaluatingthepresenceorabsenceofisotopicequilibrium inspeleothems.Wereachthelongoverdueconclusionthat theHendyTestcriteriaforjudgingspeleothems(Hendy, 1971)arenotreliablyeffectiveatscreeningstalagmitesfor paleoclimaticsuitability.Wethereforesuggestthatthe widespreadapplicationofHendyTestshouldbediscontinued.WeproposeinsteadtheReplicationTest(i.e.,the demonstrationofsimilarisotopicprofilesamongtwoor morespeleothems)forevaluatingthelikelihoodofcalcite depositionunderisotopicequilibriumconditions(e.g. Doraleetal.,1998;Wangetal.,2001;Constantinetal., 2007;Dennistonetal.,2007;Wangetal.,2008). Replicationofisotopicprofilesamongtwoormore speleothemsispossibleonlyifkinetic/vadose-zoneprocessesareeither:1)absentor2)haveaffectedspatially separatedspeleothemsinexactlythesameway.Asthe secondscenarioishighlyunlikelybecauseofthespatial heterogeneityofthekarstaquifer(Perrinetal.,2003),we proposethattheReplicationTestiseffectivelysufficientin rulingoutkinetic/vadose-zone-overprintingprocesses. O XYGENAND C ARBON I SOTOPESIN S PELEOTHEMS The d 18 Oand d 13 CvaluesofspeleothemCaCO 3 are relatedtotheprimarysourcesofoxygenandcarboninthe caveseepagewater.Inthecaseofoxygen,thisismeteoric water.Inthecaseofcarbon,thisissoilCO 2 ,atmospheric CO 2 ,andcarbonatebedrock.Inmostcases,theprocessof speleothemdepositioncanbetracedbacktothesoillayer wherebiologicalactivityproduceshighlevelsofCO 2 .This soilCO 2 acidifiesseepagewaters,whichinturndissolve carbonatebedrockenroutetounderlyingcaves.Upon enteringacavepassageoflowerCO 2 partialpressure (relativetothesoilatmosphere),theseepagewaterreleases CO 2 andCaCO 3 depositiontakesplaceinaccordancewith theequationbelow(Hollandetal.,1964): Ca 2 2HCO 3 ? CO 2 CaCO 3 H 2 O 1 BecauseHCO 3 concentrationsinkarstgroundwaters aretypicallyinthepartsperthousandrange, d 18 Ovaluesof thewaterandthedissolvedcarbonatespeciesare dominatedbythewatermoleculesthemselves,which originatedasmeteoricprecipitation.Therefore,the d 18 O valuesofspeleothemsaregenerallynotsignificantly influencedbybedrock d 18 Ovalues(Harmon,1979). Paleoclimaticinterpretationofspeleothem d 18 Ovariationsrequiresknowledgeandquantitativeestimationof processesthatmayaffecttheisotopiccompositionofwater duringthecourseofthehydrologiccycleandduringcalcite deposition.ThecavetemperatureeffectofWilliamsetal. (1999)representsisotopicfractionationbetweenwaterand calciteduringcalcitedeposition.Thetemperaturedependenceofthefractionationhasbeenexperimentally determinedas 0.24 % per u C(OÂ’Neil,1969),meaning thatthereisgreaterfractionationatlowtemperatures relativetohightemperatures.Therainwatercomposition effectrepresentsestablishedempiricalrelationships betweenprecipitation d 18 Ovaluesandcertainclimatic parameterssuchastemperatureandrainfallintensity (Dansgaard,1964).Differentpartsoftheglobeare representedbyfundamentallydifferentrelationships,dependingontheprevailingmeteorologicalpatternof precipitation.IntropicalregionsandregionsthatexperienceastrongmonsoonoraMediterraneanclimate,for example,therainwatercompositioneffectmaynothavea stronglinktotemperature.Otherfactors,suchasthe intensityoftherainfall(DansgaardÂ’s(1964)Amount Effect),orthepartitioningofmultiplemoisturesources overthecourseoftheyear,maydominatevariationsin the d 18 Ovaluesofprecipitation(Wangetal.,2001; Bar-Matthewsetal.,2003;Wangetal.,2008).However, atmanymiddleandhigh-latitudesites,therainwatercompositioneffectislinkedtotemperature.The moderndayspatialslopeforGreenland,forexample,is 0.67 % per u C(Johnsenetal.,1989).Formanymidlatituderegions,the d 18 O MAPmeanannualprecipitation MAT meanannualtemperature slopeisgenerallyalsopositive, buttheslopeistypicallysmallerthan0.67 % per u C. Cavesareexcellentenvironmentsfortemperature reconstructionsbecausetheambienttemperaturewithin sufficientlydeepcavesisconstantyear-round,andreflects themeansurfacetemperatureaveragedoverseveralyears (WigleyandBrown,1976).The 0.24 % per u C fractionationthatdefinesthecavetemperatureeffect, therefore,reflectsthemeanannualtemperatureofthearea. Ifthecavedripwatercanbeassumedtoapproximate d 18 O MAP (Yongeetal.,1985),therainwatercomposition effectandthecavetemperatureeffectmaybecombinedto yieldan d 18 O speleothem MATslopethatiseitherpositive, negative,orzero,dependingontheexactsignandslopeof the d 18 O MAP MATrelationship(HendyandWilson, 1968;Harmonetal.,1978;Gascoyneetal.,1980;Dorale etal.,1992;Winogradetal.,1992;Doraleetal.,1998; LauritzenandLundberg,1999b;Williamsetal.,1999).If the d 18 O MAP MATslopehasavaluelargerthan 0.24 % per u C,thenthe d 18 O speleothem MATslopewillbepositive (i.e.morepositive d 18 O speleothem valuesindicatewarmer temperatures).Ifthe d 18 O MAP MATslopeislessthan 0.24 % per u C,the d 18 O speleothem MATslopewillbe negative,inwhichcasemorepositive d 18 O speleothem values indicatelowertemperatures. Whatmostlylimitsrobustquantitativetemperature reconstructionsbackthroughtimeistheincomplete knowledgeofhowtheslopefor d 18 O MAP MAThas variedthroughtime.Possiblereasonsfor d 18 O MAP MAT slopevariationsincludefundamentalchangesinthe seasonalityorthemoisturesourceoftheprecipitation (Jouzeletal.,1997).Inthecaseofspeleothems,the problemsofanunknown,butpresumablyvariable rainwatercompositioneffectcouldbecircumventedif L IMITATIONSOFHENDYTESTCRITERIAINJUDGINGTHEPALEOCLIMATICSUITABILI TYOFSPELEOTHEMSANDTHENEEDFORREPLICATION 74 N JournalofCaveandKarstStudies, April2009

PAGE 3

therewassomewayofaccessingdirectlythewaterfrom whichthecalcitewasprecipitated.Fluidinclusionsin speleothemsaretinytrappedparcelsoftheoriginaldrip waterfromwhichthecalcitewasdeposited.Henry Schwarczandhisstudentspioneeredtheearlyideasof fluidinclusionwork(Schwarczetal.,1976;Harmonetal., 1979;Yonge,1982),buttechnicaldifficultiesinextracting anunfractionatedsampleofthewaterstymiedthe approachformanyyears.Recentworkshowsconsiderable promiseinhavingovercomethetechnicalchallenges(Rowe etal.,1998;Matthewsetal.,2000;Vonhofetal.,2006; Zhangetal.,2008). Mostrecently,theclumpedisotopetechniquehasbeen developedasatypeofquantitativepaleothermometerfor carbonatesthatrequiresnoinformationorassumptions aboutthe d 18 Ovaluesofwatersfromwhichcarbonatesgrew (Eiler,2007;Affeketal.,2008).Thisisanexciting development,althoughapplicationoftheclumpedisotope paleothermometertospeleothemsisinitsinfancyandwe wouldthinktherewillbesignificantchallenges(although notinsurmountable)inaccuratelyapplyingthetechniqueto speleothemsbecauseofissueswithsamplesizeandprecision. Unlike d 18 Ovalues,speleothem d 13 Cvaluesaresignificantlyinfluencedbytheisotopiccompositionofthe bedrockandthesoilCO 2 (Deinesetal.,1974).Vegetation isamajorcontrollingfactorofspeleothem d 13 Cvalues becausesoilCO 2 isgeneratedlargelybyrootrespirationand themicrobialoxidationofsoilorganicmatter,whichis derivedfromvegetation.Whereasbedrockcharacteristics aregenerallystableoverthetimescalesofU/Thdating, vegetationcanbedynamic.Mostoftheworld’splantsutilize eithertheC 3 photosyntheticpathwayortheC 4 pathway(a minorgroupistheCAMpathway).TheC 3 andC 4 photosyntheticpathwaysproducelargedifferencesin d 13 C values.C 3 plantshave d 13 Cvaluesaveragingca. 26 % whereasC 4 plantsaverageca. 12 % (Deines,1986).C 4 plantsaretypicallywarm-seasongrassesandafewherbs foundintropicalandtemperategrasslands,whereasC 3 plantsincludemosttrees,shrubs,cool-seasongrasses,and mostherbs.WhereC 3 andC 4 planttypescoexist,their biomassratiowillbereflectedinthesoilCO 2 d 13 Cvalues. Proxiescapableofpreservingsoil d 13 Csignaturessuchas speleothems(Doraleetal.,1992;1998;Holmgrenetal., 1995)arethuscapableofrecordingtheproportionofC 3 to C 4 plantbiomassthroughtime,andareindirectlylinkedto climate.Thus,wedonotnecessarilyagreewiththe generalizedassertionofDreybrodt(2008),that‘‘...oxygen isotoperecordsofstalagmitearemorepromisingfor extractingpaleo-climaticsignalsthanthoseofcarbon’’. L IMITATIONSOFTHE H ENDY -T EST C RITERIA E QUILIBRIUM V ERSUS N ON -E QUILIBRIUM D EPOSITION OF C ALCITE Althoughitisclearthatspeleothem d 18 Ovaluesare controlledbytemperatureandtheisotopiccompositionof meteoricwaterandthat d 13 Cvaluesarecontrolledbythe isotopiccompositionofbedrockandsoilorganicmatter, thedegreetowhichthe d 18 Oand d 13 Cvaluesofspeleothem calcitereflectonlytheseprimaryenvironmentalvariablesis possiblyamorecomplexissue.Itispossiblethatother processesmaymasktheseprimaryenvironmentalvariables (Hendy,1971).Theseprocessesmayfallintothecategory ofkineticprocessesduringcalcitecrystallization.Hendy (1971)distinguishedsuchkineticeffectsfromthemore desirableisotopic-equilibriumcondition.Thatisthe conditiondefinedbysufficientlyslowdegassingratesand withnoevaporation,suchthatthefractionationofheavy andlightisotopesbetweenaqueousandsolidphasesisonly afunctionoftemperature.Althoughusefulpaleoenvironmentalinformationmayberecordedinspeleothems depositedoutofisotopicequilibrium(TalmaandVogel, 1992;Spo ¨tlandMangini,2002),equilibriumdepositionis generallymoredesirablebecause,undersuchconditions, temporalvariationsin d 18 Oand d 13 Cvaluesmostlikely reflectchangesintheprimaryenvironmentalvariablesof interestdirectly,notindirectlythroughcomplicatedcontrolsonvaryingdegreesofnon-equilibriumeffects. Recently,anumberoffield,experimental,andmodelingstudieshavefocuseddirectlyontheissueofisotopic equilibriuminstalagmites(e.g.,Mickleretal.,2004; Mickleretal.,2006;Mu ¨hlinghausetal.,2007;Romanov etal.,2008;Dreybrodt,2008).Allhaveprovidedvaluable insightintothisimportantissueofequilibriumversusnonequilibriumdeposition,butnonehaveseriouslychallenged thegeneralapplicabilityoftheHendyTestinscreening stalagmitesforpaleoclimaticsuitability. T HE H ENDY -T EST C RITERIAAND T HEIR L IMITATIONS Inmanyrecentstudies,researchershaveappliedHendy Testcriteriatoasinglestalagmitetoascertainwhetherthe calcitewasdepositedunderisotopicequilibriumconditions (e.g.,Fleitmannetal.,2004;Dykoskietal.,2005;Vaccoet al.,2005;Williamsetal.,2005;Johnsonetal,2006a;Spo ¨tl etal.,2006;Manginietal.,2007;Huetal.,2008;Zhanget al.,2008;Zhouetal.,2008).TheHendyTestapproach requiresdrillingsubsamplesalongagrowthlayerinthe stalagmite,fromthecenteroutwardanddowntheflanks. TheHendyTestcriteriaare:(1) d 18 Ovaluesremain constantalongasinglegrowthlayer;(2)thereareno simultaneousenrichmentsof 13 Cand 18 Ointhespeleothem calcite(i.e.,thereisnocorrelationbetween d 13 Cand d 18 O values).Thebasisfor 13 Cenrichmentsisthoughttobe linkedtoRayleighdistillationoftheHCO 3 aq reservoir duringdegassingof d 13 C-depletedCO 2 (Hendy,1971; Mickleretal.,2006;Romanovetal.,2008).Incontrast, 18 OmaynotbenearlyasaffectedbyRayleighdistillation duringdegassingbecauseCO 2 hydrationandhydroxylationreactionswillbuffertheoxygenisotopiccomposition oftheHCO 3 aq reservoir(Mickleretal.,2006). However,iftheeffectsofRayleighdistillationmanifest themselvesintheoxygenisotopicsystem,theywillresultin J.A.D ORALEAND Z.L IU JournalofCaveandKarstStudies, April2009 N 75

PAGE 4

d 18 OenrichmentintheHCO 3 aq reservoirandultimately intheprecipitatedCaCO 3 Basedontheaboveargument,criterion(1)mayatface valueappeartohaveasoundtheoreticalbasis,butitmay notbeaperfecttestforthepresenceofequilibrium conditionsbecauseisotopicequilibriumfractionation couldoccurinthecenterofastalagmiteatthesametime thatkineticfractionationoccursattheflanks(Talmaand Vogel,1992;Spo ¨tlandMangini,2002;Dreybodt,2008).In thisparticularcase,criterion(1)oftheHendyTestcould failtoindicatethatisotopicequilibriumwasmaintainedin anysectorofthestudiedstalagmite.Anadditional,and largelyunderappreciated,problemwithcriterion(1) revolvesaroundtheissuesofsamplingresolutionand samplingerrors.Inpractice,criterion(1)isdifficult(ifnot impossible)toapply,inpart,becauseevensmallsampling errorscouldinvalidatethewithingrowthlayercomparison, andinpart,becausethegeometryofatypicalstalagmiteis notwell-suitedforthetask.Atypicalgrowthlayeris commonlythickestalongthetopsurfaceofthestalagmite andbecomesprogressivelythinneralongthesides.If growthlayerswereofuniformthickness,stalagmiteswould betwiceaswideastheyaretall,whichisseldom,ifever, thecase.Therefore,whenaconstantdiameterdrillbitis used,drillingsfromthesidesofthestalagmite(wherethe targetedlayeristhinner)willlikelyincludecalcitethatis youngerand/orolderrelativetodrillingsfromthetop surface.Thiserrantlysampledyoungandoldcalcite (possiblyevenincludingsmallgrowthhiatusesnotpresent inthestalagmitecenter)mayhaveasignificantlydifferent isotopiccompositionthanthatoftargetedgrowthlayer. Thus,acomparisonofcoevalsub-samplesfromthetop andthesidesisnottrulyvalid,andanegativeresultofthe HendyTestdoesnot,therefore,necessarilyindicatekinetic fractionationduringdeposition(LauritzenandLundberg, 1999a).Atrulyvalidattemptatexecutingcriterion(1) wouldbynecessityusemultipledrillbitsofvariable diameter.Thisapproachhasnever,toourknowledge,been reportedintheliterature. Another,possiblymorecriticalissuewithregardto samplingresolution,involvesthelargetemporalmismatch betweendrilledsubsamplesandtheactualtimescaleof calcite-formationprocesses.Subsamplesdrilledusing moderntechniquesaretimeintegrationstypicallyonthe orderofmonthstocenturies,dependingoncalcite depositionrates.Eventheshortendofthistimescaleis clearlyincompatiblewiththesecondstohourstimescaleof driprate,CO 2 degassing,calciteprecipitation,andother reactionprocessesbeingevaluatedasequilibriumversus non-equilibriumbyHendyTestcriterion(1).Inother words,sincedrilledsub-samplesoftenincorporateyearsto centuriesofmaterialintoasingleisotopicvalue,criterion (1)implicitlyassumesthatanylevelofapparentisotopic disequilibriumdetectedbycriterion(1)wasmaintainedat someconstantlevelfortheentiretimeofintegration.This ishighlyunlikely,asnumerousstudiesofcavedripwaters haveshownhighvariabilityinseasonaldripandrelated isotopicphenomena(e.g.Bar-Matthewsetal.,1996;Treble etal.,2005;Johnsonetal.,2006b;Cobbetal.,2007). Thesametemporalmismatchisproblematicforthe samereasontostudiesofmoderndripwatersand associatedspeleothems,whichhasbeenclaimedbysome toofferthemostconvincingcaseforassessingwhether isotopicequilibriumexistsinagivencavesetting(e.g.,BarMatthewsetal.,1996;Mickleretal.,2006;Harmonetal., 2004).Theproblemagainisthatdripwaterscollectedin modernsettingsrepresentisotopicsnapshotsofseconds-tominuteswhilecalcitecollectedforcomparison(dueto samplesizeconstraints)mayrepresentmuchlongertime integrations.Furthermore,thismismatchinwatervolume versuscalcitevolume(normalizedtotime)toenable measurementsonbothprobablyresultsinsignificantbias inmoderndrip/calcitestudiestowardfast-growingcalcites. Unfortunately,thismaywellbethetypeofcalcitemost likelytohavebeendepositedoutofisotopicequilibrium. HendyTestcriterion(2)isbasedontheassumptionthat thereisnosystematicchangeinthe d 13 CvalueofsoilCO 2 withachangeinclimateasreflectedby d 18 Ovalues (Hendy,1971).Whilethisscenarioistheoreticallypossible, itwouldappeartobeunlikelyinthemajorityofcases.In fact,theconcentrationand d 13 CvaluesofsoilCO 2 aretied toclimate,whichcontrolsthesoilbio-productivity(Cerling andQuade,1993;Hellstrometal.,1998;Hellstrometal, 2000;Gentyetal.,2006)andthetypeofvegetation (Cerling,1984;Deines,1986).Asoneexample,vegetation couldchangewithachangeinclimate(Cerling,1984; Doraleetal.,1992;1998).Clearlyinthiscase,simultaneous changesinspeleothem d 13 Cand d 18 Ovaluescouldbe expected(Doraleetal.,1992;Bakeretal.,1998;Holmgren etal.,1995;Bar-Matthewsetal.,1997;Doraleetal.,1998; Dennistonetal.,2000;Frumkinetal.,2000;Dennistonet al.,2001;Xiaetal.,2001;Burnsetal.,2002;Vaksetal., 2003,McDermottetal.,2004;Zhangetal.,2004,2006; Frisiaetal.,2005;Vaccoetal.,2005;CruzJr.etal.,2006; Vaksetal.,2006;Manginietal.,2007).Thus,insome cases,suchacouplingof d 13 Cand d 18 Ovaluesisnot relatedtothekineticisotopeeffects,butisinsteadan indicationofclimaticchange.Severalexamplesfollow below. TheworkofDoraleetal.(1998)providesahighly illustrativeexampleofthistypeofcouplingintheirstudy ofseveralstalagmitesfromCreviceCave,Missouri,USA. Theyfoundthatmultiplestalagmitesyieldedhighly replicating d 13 Cand d 18 Oprofilesthroughoutpartofthe lastglacialperiod(Fig.1).Temperaturesinterpretedfrom d 18 OvaluesandC 4 plantbiomassinterpretedfrom d 13 C valueswerebothrelativelyhighfrom59–55ka(Fig.1). Declining d 13 Cand d 18 Ovalueswereinterpretedasa coolingclimatewithforestreplacinggrasslandaround 55ka.Thekeyelementininterpretingthecoupledchanges in d 13 Cand d 18 Ovaluesasclimate-drivenversuskineticdriven(asmightbeconcludedfromaHendyTest L IMITATIONSOFHENDYTESTCRITERIAINJUDGINGTHEPALEOCLIMATICSUITABILI TYOFSPELEOTHEMSANDTHENEEDFORREPLICATION 76 N JournalofCaveandKarstStudies, April2009

PAGE 5

approach)wasthereplication.Thestalagmitesdifferedin growthrates,diameters,andanumberofothercharacteristicsthatcouldbeconsideredreflectiveofdriprates,CO 2 degassingrates,andcalciteprecipitationrates(allthought toinfluenceequilibriumversusnon-equilibriumdeposition),yet d 13 Cand d 18 Ovaluesreplicated.Thereplication impliesthatthedifferencesinhydrologiccharacteristics betweenspecificdrippathwayscannotaccountforthe long-termchangesinspeleothem d 13 Cand d 18 Ovalues. Instead,these d 13 Cand d 18 Otrendslikelyreflectamore pervasiveinfluence,suchastheclimatedrivenchangesin d 13 Cvaluesoftheoverlyingsoilandvegetation. Hellstrometal.(1998)producedadetailedrecordof climateandvegetationchangefromthe d 13 Cand d 18 O profilesoftwoNewZealandspeleothems(Fig.2).The d 13 Crecord(Fig.2)isinterpretedintermsofinorganic processesactingoncarbonderivedpurelyfromC 3 plants (asC 4 plantsdonotexistinNewZealand),andappearsto becloselyrelatedtochangesinsoilCO 2 concentration abovethecave.Assuch,itisconsideredaproxyforforest productivityabovethecave,aconclusionstronglysupportedbyitssimilaritytopollen-basedrecordsofforest extentintheregion.Notably,anextremelyrapidincrease inforestcover(indicatedbysteepdecreasein d 13 Cvalues (Fig.2))coincidentwithasignificantincreaseinspeleothemgrowthrate,wascenteredon15kaandmarking theendofglacialmaximumconditions(indicatedbythe highestspeleothem d 18 Ovaluesandhighestspeleothem d 13 Cvaluesat 19ka(Fig.2))intheregion.Close examinationofFigure2revealsashortdelayinthedecline of d 13 Cvaluesrelativeto d 18 Ovalues,whichmaycapturea smalllagbetweenthetimeofclimatechangeandsoil microbere-activationinducedbytheclimatechange. SimilartotheexamplefromCreviceCave,thesystematic co-variationbetween d 13 Cand d 18 Ovaluesisagainnot necessarilyrelatedtokineticisotopicfractionation,but instead,islikelyforcedbyclimatechange. Inourlastexample,Baldinietal(2005)foundthat increasesincalcitedepositionrates,combinedwith decreasesin d 13 Cand d 18 Ovaluesinthreemodern stalagmitesfromBrownÂ’sFollyMine,Wiltshire,England, arecorrelativewithawell-documentedre-vegetationabove themine.Increasedsoil p CO 2 resultedingreateramounts ofdissolvedCaCO 3 inthedripwaters,whichconsequently increasedannualcalcitedepositionrates.Lower d 13 C valuesthroughtimemayreflecttheincreasedinputof isotopicallylightbiogeniccarbontothetotaldissolved inorganiccarbon(DIC).Inthiscase, d 18 Ovaluesdecreased synchronouslywith d 13 Cvalues,reflectingtheincreased importanceofisotopicallylightwinterrechargedueto greaterbiomass-inducedsummerevaportranspiration. Finally,inviewofhisrecentmodelingresultsDreybrodt(2008)concludedthatstalagmitescollectedfor paleoclimaticstudiesshouldhavegrownwithdripintervals oflessthan50seconds,whichresultsindiametersofat least10cmandpredictsoxygenisotopicequilibriuminthe calcite.Wefindthisresultintriguingandperhapsreflective ofthesituationinseveralrecentlypublishedChinesecave d 18 Orecords,whicharederivedfromfast-growing,large diameter,replicatingstalagmites(e.g.,Wangetal.,2001; Yuanetal.,2004).Ontheotherhand,thereplicatingfossil stalagmitesfromCreviceCavearecharacteristicallyslowFigure1.The d 13 Cand d 18 Ovaluesvs.time,forstalagmites CC94-DBL-LandCC94-E(afterDoraleetal.,1998), showingthatthesimultaneousenrichmentof 13 Cand 18 O(or anegativeresultoftheHendyTest)istheindicationof changeinbothclimateandvegetationtype,butnottheresult ofnonequilibriumisotopicfractionation.Thatmeansa negativeresultoftheHendyTestdoesnotnecessarily indicatekineticfractionation. Figure2.The d 13 Cand d 18 Ovaluesvs.time,foradjacent incrementsofcoresMD3andED1(afterHellstrometal., 1998),showingthatthesimultaneousenrichmentof 13 Cand 18 O(oranegativeresultoftheHendyTest)isnotrelatedto thekineticisotopeeffectsinthiscase,butratheran indicationofchangesinbothclimateandforestproductivity. J.A.D ORALEAND Z.L IU JournalofCaveandKarstStudies, April2009 N 77

PAGE 6

growing(typicalrange:1–10 myr 1 )andwithdiameters typicallyintherangeof4–7cm.Therefore,ourfield evidencefromCreviceCavewouldnotappeartosupport thepredictionsofDreybrodt’s(2008)model.Wesuspect insteadthatslowdegassingandcalciteprecipitationrates, asreflectedintheslowgrowthratesoftheseparticular speleothems,likelycontributedtotheirreplicatingprofiles, whichweinterpretasequilibriumdeposition. C ONCLUSIONS Theargumentforreplicationissimple,straightforward, andpowerful.Imaginethescenariowheretwocoeval stalagmitesfromagivencavecontaindisparate d 13 Cand d 18 Oprofiles.Whichstalagmitecontainsthecorrect climaticsignal,stalagmiteA,stalagmiteB,orneither? TraditionallytheHendyTestwouldbeemployedtosort outthismystery,butwehaveshownherethattheHendy Testispronetoproducinginvalidcomparisonsandfalse negativeresults;andtherefore,cannotbeconsidered accurateinitsapproach.Intruthandinpractice,the HendyTestisaneasilyexecutedexercisewidelyusedinthe reviewandeditorialprocesstovalidatespeleothemclimate records,but,unfortunately,itisnotareliableexercise. TohelpresolvetheconundrumofourstalagmiteA versusBscenario,shouldweinsteadadopttheadvicefrom modelingstudiessuchasDreybrodt(2008)andchoosethe largerdiameterstalagmiteoverthesmallerdiameterone? Shouldwechoosethestalagmitewiththelessporous fabric,ortheonelesscolored?Shouldwechoosethe stalagmitewithflatlayeringortheonewitharcedlayering? Intruth,wecannotknowfromthesetypesofbasic observationswhichone,ifeither,reflectsequilibrium deposition.Hadasinglestalagmitebeenusedinthestudy, theentireissueofdisparitywouldhavebeenavoided,but wewouldbenoclosertothetruthindecipheringthestatus ofequilibriumdeposition. Wewouldarguethatonlyinthecasewheretwoor morestalagmitesreplicate d 18 Oand d 13 Cvaluescanwe haveanysignificantmeasureofconfidencethatisotopic equilibriumwasmaintained.Othershavearguedthatthe onlyrealtestofequilibriumdepositionistodirectly measureandmonitortheisotopiccompositionofdrip watersandcalcites,alongwiththeenvironmentalconditionsthecalciteisprecipitatedunder,inmodernsystems, andthenusethisinformationtohelpinterprettheancient speleothemrecord(e.g.,Bar-Matthewsetal.,1996;Mickler etal.,2004;2006;Harmonetal.,2004).Whilewedonot necessarilydisagreewiththewisdomofthisapproach,it maynotbefeasibleinmanycases,andalso,thereisno guaranteethatmodernconditionsmimicancientones.In fact,withlargedifferencesinclimateandthewholerange ofcarbonatechemistryandwaterflowparameterslinked toclimate,itmayevenbelikelythatmodernsystemsare pooranalogsforancientones.Wewouldargue,again,that replicationisakey,asisahighlevelofagreementwith independentindicatorsofpaleoclimaticconditions,suchas paleovegetationoricecores. Finally,ouradvocacyforreplicationhasimplications forcaveconservation.Speleothemcollectionshouldalways becarriedoutwiththeveryhigheststandardsof conservation.Insomecases,thecollectionofmultiple speleothemssimplymaynotbejustifiable.Instead,we shouldfocusoureffortsonthosesystemsthathavean adequateresourcetoallowforsamplingwithminimal impact.InthecaseofCreviceCave,forexample,literally hundredsofspeleothems,naturallybroken,existinthe cave.Wemightalsoconsiderincreasingourdepthof understandingofthosesystemsthathavealreadydisplayed promiseoffaithfullycapturingpaleoclimaticconditions, versusracingaroundtheglobeattemptingtocollect samplesfrompristinesettings. A CKNOWLEDGMENTS ThisworkwassupportedbytheU.S.NationalScience FoundationGrantATM-0402482toJ.A.D.Supportto Z.L.includestheHundredTalentsProgramofChinese AcademyofSciences,thefoundationoftheChinese AcademyofSciencesforInnovation(GrantNo.kzcx2yw-306),NationalNaturalScienceFoundationofChina (GrantNos.40572017and40872168),andMinistryof ScienceandTechnologyofChina(GrantNo. 2005DIB3J067).WethankP.J.Micklerandananonymous reviewerforcommentsthathelpedimprovethemanuscript. R EFERENCES Affek,H.P.,Bar-Matthews,M.,Ayalon,A.,Matthews,A.,andEiler, J.M.,2008,Glacial/interglacialtemperaturevariationsinSoreqcave speleothemsasrecordedby‘clumpedisotope’thermometry:GeochimicaetCosmochimicaActa,v.72,p.5351–5360. Baker,R.G.,Gonza lez,L.A.,Raymo,M.,Bettis,E.A.,Reagan,M.K., andDorale,J.A.,1998,Comparisonofmultipleproxyrecordsof HoloceneenvironmentsintheMidwesternUnitedStates:Geology, v.26,p.1131–1134. Baldini,J.U.L.,McDermott,F.,Baker,A.,Baldini,L.M.,Mattey,D.P., andRailsback,L.B.,2005,Biomasseffectsonstalagmitegrowthand isotoperatios:A20thcenturyanaloguefromWiltshire,England: EarthandPlanetaryScienceLetters,v.240,p.486–494. Bar-Matthews,M.,Ayalon,A.,Matthews,A.,Sass,E.,andHalicz,L., 1996,Carbonandoxygenisotopestudyoftheactivewater-carbonate systeminakarsticMediterraneancave:Implicationsforpaleoclimate researchinsemiaridregions:GeochimicaetCosmochimicaActa, v.60,p.337–347. Bar-Matthews,M.,Ayalon,A.,andKaufman,A.,1997,LateQuaternary paleoclimateintheEasternMediterraneanregionfromstableisotope analysisofspeleothemsatSoreqCave,Israel:QuaternaryResearch, v.47,p.155–168. Bar-Matthews,M.,Ayalon,A.,Gilmour,M.,Matthews,A.,and Hawkesworth,C.J.,2003,Sea–landoxygenisotopicrelationships fromplanktonicforaminiferaandspeleothemsintheEastern Mediterraneanregionandtheirimplicationforpaleorainfallduring interglacialintervals:GeochimicaetCosmochimicaActa,v.67, p.3181–3199. Burns,S.J.,Fleitmann,D.,Mudelsee,M.,Neff,U.,Matter,A.,and Mangini,A.,2002,A780-yearannuallyresolvedrecordofIndian OceanmonsoonprecipitationfromaspeleothemfromSouthOman: L IMITATIONSOFHENDYTESTCRITERIAINJUDGINGTHEPALEOCLIMATICSUITABILI TYOFSPELEOTHEMSANDTHENEEDFORREPLICATION 78 N JournalofCaveandKarstStudies, April2009

PAGE 7

JournalofGeophysicalResearch—Atmospheres,v.107,doi:10.1029/ 2001JD001281. Cerling,T.E.,1984,Thestableisotopecompositionofmodernsoil carbonateanditsrelationshiptoclimate:EarthandPlanetaryScience Letters,v.71,p.229–240. Cerling,T.E.,andJ.Quade,1993,Stablecarbonandoxygenisotopesin soilcarbonates, in Swart,P.K.,Lohmann,K.C.,McKenzie,J.A.,and Savin,S.M.,eds.,ClimateChangeinContinentalIsotopicRecords, Washington,D.C.,AmericanGeophysicalUnionGeophysicalMonograph,v.78,p.217–231. Cobb,K.M.,Adkins,J.F.,Partin,J.W.,andClark,B.,2007,Regionalscaleclimateinfluencesontemporalvariationsofrainwaterandcave dripwateroxygenisotopesinnorthernBorneo:EarthandPlanetary ScienceLetters,v.263,p.207–220. Constantin,S.,Bojar,A.,Lauritzen,S.,andLundberg,J.,2007,Holocen e andLatePleistoceneclimateinthesub-Mediterraneancontinental environment:AspeleothemrecordfromPolevaCave(Southern Carpathians,Romania):Palaeogeography,Palaeoclimatology,Palaeoecology,v.243,p.322–338. Cruz,Jr.,F.W.,Burnsa,S.J.,Karmann,I.,Sharp,W.D.,Vuillea,M.,and Ferrari,J.A.,2006,Astalagmiterecordofchangesinatmospheric circulationandsoilprocessesintheBraziliansubtropicsduringthe LatePleistocene:QuaternaryScienceReviews,v.25,p.2749–2761. Dansgaard,W.,1964,Stableisotopesinprecipitation:Tellus,v.16, p.436–468. Deines,P.,Langmuir,D.,andHarmon,R.S.,1974,Stablecarbonisotope ratiosandtheexistenceofagasphaseintheevolutionofcarbonate waters:GeochimicaetCosmochimicaActa,v.38,p.1147–1164. Deines,P.,1986,Theisotopiccompositionofreducedorganiccarbon, in Fritz,P.,andFontes,J.C.,eds.,HandbookofEnvironmental Geochemistry:TheTerrestrialEnvironment,Amsterdam,Elsevier, p.331–406. Denniston,R.F.,Gonzalez,L.A.,Asmerom,Y.,Reagan,M.K.,and Recelli-Snyder,H.,2000,Speleothemcarbonisotopicrecordsof HoloceneenvironmentsintheOzarkHighlands,USA:Quaternary International,v.67,p.21–27. Denniston,R.F.,Gonzalez,L.A.,Asmerom,Y.,Polyak,V.,Reagan, M.K.,andSaltzman,M.R.,2001,Ahigh-resolutionspeleothem recordofclimaticvariabilityattheAllerod-YoungerDryastransition inMissouri,centralUnitedStates:Palaeogeography,Palaeoclimatology,Palaeoecology,v.176,p.147–155. Denniston,R.F.,DuPree,M.,Dorale,J.A.,Asmerom,Y.,Polyak,V.J., andCarpenter,S.J.,2007,EpisodesoflateHolocenearidityrecorded bystalagmitesfromDevil’sIceboxCave,centralMissouri,USA: QuaternaryResearch,v.68,p.45–52. Dorale,J.A.,Gonza lez,L.A.,Reagan,M.K.,Pickett,D.A.,Murrell, M.T.,andBaker,R.G.,1992,Ahigh–resolutionrecordofHolocene climatechangeinspeleothemcalcitefromColdWaterCave,northeast Iowa:Science,v.258,p.1626–1630. Dorale,J.A.,Edwards,R.L.,Ito,E.,andGonzalez,L.A.,1998,Climate andvegetationhistoryofthemid-continentfrom75to25ka:A speleothemrecordfromCreviceCave,Missouri,USA:Science, v.282,p.1871–1874. Dorale,J.A.,Edwards,R.L.,Alexander,Jr.,E.C.,Shen,C.-C.,Richards D.A.,andCheng,H.,2004,Uranium-seriesdisequilibriumdatingof speleothems:Currenttechniques,limits,andapplications, in Mylroie, J.,andSasowsky,I.D.,eds.,StudiesofCaveSediments:Physicaland ChemicalRecordsofPaleoclimate,NewYork,KluwerAcademic/ PlenumPublishers,p.177–197. Dreybrodt,W.,2008,Evolutionoftheisotopiccompositionofcarbonand oxygeninacalciteprecipitatingH 2 O-CO 2 -CaCO 3 solutionandthe relatedisotopiccompositionofcalciteinstalagmites:Geochimicaet CosmochimicaActa,v.72,p.4712–4724. Duplessy,J.C.,Labeyrie,J.,Lalou,C.,andNguyen,H.V.,1970, Continentalclimaticvariationsbetween130,000and90,000years B.P.:Nature,v.226,p.631–633. Dykoski,C.A.,Edwards,R.L.,Cheng,H.,Yuan,D.,Cai,Y.,Zhang,M., Lin,Y.,Qing,J.,An,Z.,andRevenaugh,J.,2005,Ahigh-resolution, absolute-datedHoloceneanddeglacialAsianmonsoonrecordfrom DonggeCave,China:EarthandPlanetaryScienceLetters,v.233, p.71–86. Edwards,R.L.,Chen,J.H.,andWasserburg,G.J.,1987, 238 U234 U230 Th232 Thsystematicsandtheprecisemeasurementof timeoverthepast500,000y:EarthandPlanetaryScienceLetters, v.81,p.175–192. Eiler,J.M.,2007,‘‘Clumped-isotope’’geochemistry—Thestudyof naturally-occurring,multiply-substitutedisotopologues:Earthand PlanetaryScienceLetters,v.262,p.309–327. Fleitmann,D.,Burns,S.J.,Neff,U.,Mudelsee,M.,Mangini,A.,and Matter,A.,2004,Paleoclimaticinterpretationofhigh-resolution oxygenisotopeprofilesderivedfromannuallylaminatedspeleothems fromSouthernOman:QuaternaryScienceReviews,v.23,p.935–945. Frisia,S.,Borsato,A.,Spo `tl,C.,Villa,I.M.,andCucchi,F.,2005,Climate variabilityintheSEAlpsofItalyoverthepast17000years reconstructedfromastalagmiterecord:Boreas,v.34,p.445–455. Frumkin,A.,Ford,D.C.,andSchwarcz,H.P.,2000,Paleoclimateand vegetationofthelastglacialcyclesinJerusalemfromaspeleothem record:GlobalBiogeochemicalCycles,v.14,p.863–870. Galimov,E.M.,andGrinenko,V.A.,1965,Theeffectofleachingunder surfaceconditionsontheisotopiccompositionofcarboninsecondary calcite:GeochemistryInternational,v.1,p.79–82. Gascoyne,M.,Schwarcz,H.P.,andFord,D.C.,1980,Apaleotemperature recordforthemid-WisconsininVancouverIsland:Nature,v.285, p.474–476. Genty,D.,Blamart,D.,Ghaleb,B.,Plagnes,V.,Causse,C.h., Bakalowicz,M.,Zouari,K.,Chkir,N.,Hellstrom,J.,Wainer,K., andBourge,F.,2006,Timinganddynamicsofthelastdeglaciation fromEuropeanandNorthAfrican d 13 Cstalagmiteprofiles— comparisonwithChineseandSouthHemispherestalagmites: QuaternaryScienceReviews,v.25,p.2118–2142. Harmon,R.S.,Thompson,P.,Schwarcz,H.P.,andFord,D.C.,1978,Late PleistocenepaleoclimatesofNorthAmericaasinferredfromstable isotopestudiesofspeleothems:QuaternaryResearch,v.9,p.54–70. Harmon,R.S.,1979,Anisotopicstudyofgroundwaterseepageinthe centralKentuckykarst:WaterResourcesResearch,v.15,p.476–480. Harmon,R.S.,Schwarcz,H.P.,andO’Neil,J.R.,1979,D/Hratiosin speleothemfluidinclusions:Aguidetovariationsintheisotopic compositionofmeteoricprecipitation:EarthandPlanetaryScience Letters,v.42,p.254–266. Harmon,R.S.,Schwarcz,H.P.,Gascoyne,M.,Hess,J.W.,andFord, D.C.,2004,Paleoclimateinformationfromspeleothems:Thepresent asaguidetothepast, in Mylroie,J.,andSasowsky,I.D.,eds.,Studies ofCaveSediments:PhysicalandChemicalRecordsofPaleoclimate, NewYork,KluwerAcademic/PlenumPublishers,p.199–226. Hellstrom,J.,McCulloch,M.,andStone,J.,1998,Adetailed31,000-year recordofclimateandvegetationchangefromtheisotopegeochemistryoftwoNewZealandspeleothems:QuaternaryResearch,v.50, p.167–178. Hellstrom,J.,andMcCulloch,M.T.,2000,Multi-proxyconstraintsonthe climaticsignificanceoftraceelementrecordsfromaNewZealand speleothem:EarthandPlanetaryScienceLetters,v.179,p.287–297. Hellstrom,J.,2006,U-ThdatingofspeleothernswithhighinitialTh-230 usingstratigraphicalconstraint:QuaternaryGeochronology,v.1, p.289–295. Hendy,C.H.,andWilson,A.T.,1968,Palaeoclimaticdatafrom speleothems:Nature,v.216,p.48–51. Hendy,C.H.,1971,Theisotopicgeochemistryofspeleothems-I:The calculationsoftheeffectsofdifferentmodesofformationonthe isotopiccompositionofspeleothemsandtheirapplicabilityas paleoclimateindicators:GeochimicaetCosmochimicaActa,v.35, p.801–824. Holland,H.D.,Kirsipu,T.V.,Huebner,J.S.,andOxburgh,U.M.,1964, Onsomeaspectsofthechemicalevolutionofcavewaters:Journalof Geology,v.72,p.36–67. Holmgren,K.,Karlen,W.,andShaw,P.A.,1995,Palaeoclimatic significanceofthestableisotopiccompositionandpetrologyofa LatePleistocenestalagmitefromBotswana:QuaternaryResearch, v.43,p.320–328. Hu,C.,Henderson,G.M.,Huang,J.,Xie,S.,Sun,Y.,andJohnson,K.R., 2008,QuantificationofHoloceneAsianmonsoonrainfallfrom spatiallyseparatedcaverecords:EarthandPlanetaryScienceLetters, v.266,p.221–232. Johnsen,S.J.,Dansgaard,W.,andWhite,J.W.C.,1989,Theoriginof Arcticprecipitationunderpresentandglacialconditions:Tellus,v.41 p.452–468. Johnson,K.R.,Ingram,B.L.,Sharp,W.D.,andZhang,P.,2006a,East AsiansummermonsoonvariabilityduringMarineIsotopeStage5 J.A.D ORALEAND Z.L IU JournalofCaveandKarstStudies, April2009 N 79

PAGE 8

basedonspeleothem d 18 OrecordsfromWanxiangCave,central China:Palaeogeography,Palaeoclimatology,Palaeoecology,v.236, p.5–19. Johnson,K.R.,Hu,C.,Belshaw,N.S.,andHenderson,G.M.,2006b, Seasonaltrace-elementandstable-isotopevariationsinaChinese speleothem:Thepotentialforhigh-resolution:EarthandPlanetary ScienceLetters,v.244,p.394–407. Jouzel,J.,Alley,R.B.,Cuffey,K.M.,Dansgaard,W.,Grootes,P., Hoffman,G.,Johnsen,S.J.,Koster,R.D.,Peel,D.,Shuman,C.A., Stievenard,M.,Stuiver,M.,andWhite,J.,1997,Validityofthe temperaturereconstructionfromwaterisotopesinicecores:Journal ofGeophysicalResearch,v.102,p.26,471–26,487. Lauritzen,S.-E.,andLundberg,J.,1999a,Speleothemsandclimate:A specialissueofTheHolocene:TheHolocene,v.9,p.643–647. Lauritzen,S.-E.,andLundberg,J.,1999b,Calibrationofthespeleothem deltafunction:AnabsolutetemperaturerecordfortheHolocenein northernNorway:TheHolocene,v.9,p.659–669. Mangini,A.,Blumbacha,P.,Verdesa,P.,Spotl,C.,Scholz,D.,Machel, H.,andMahon,S.,2007,Combinedrecordsfromastalagmitefrom BarbadosandfromlakesedimentsinHaitirevealvariableseasonality intheCaribbeanbetween6.7and3kaBP.:QuaternaryScience Reviews,v.26,p.1332–1343. Matthews,A.,Ayalon,A.,andBar-Matthews,M.,2000, D / H ratiosof fluidinclusionsofSoreqcaveIsraelspeleothemsasaguidetothe EasternMediterraneanMeteoricLinerelationshipsinthelast120ky: ChemicalGeology,v.166,p.183–191. McDermott,F.,2004,Paleoclimatereconstructionfromstableisotope variationsinspeleothems:areview:QuaternaryScienceReviews, v.23,p.901–918. Mickler,P.J.,Banner,J.L.,Stern,L.,Asmerom,Y.,Edwards,R.L.,and Ito,E.,2004,Stableisotopevariationsinmoderntropicalspeleothems:Evaluatingequilibriumvs.kineticisotopeeffects:GeochimicaetCosmochimicaActa,v.68,p.4381–4393. Mickler,P.J.,Stern,L.A.,andBanner,J.L.,2006,Largekineticisotope effectsinmodernspeleothems:GeologicalSocietyofAmerica Bulletin,v.118,p.65–81. Mu ¨hlinghaus,C.,Scholz,D.,andMangini,A.,2007,Modelling stalagmitegrowthand d 13 Casafunctionofdripintervaland temperature:GeochimicaetCosmochimicaActa,v.71,p.2780–2790. O’Neil,J.R.,Clayton,R.N.,andMayeda,T.K.,1969,Oxygenisotope fractionationindivalentmetalcarbonates:JournalofChemical Physics,v.51,p.5547–5559. Perrin,J.,Jeannin,P.Y.,andZwahlen,F.,2003,Implicationsofthe spatialvariabilityofinfiltration-waterchemistryfortheinvestigat ion ofakarstaquifer:afieldstudyatMilandretestsite,SwissJura: HydrogeologyJournal,v.11,p.673–686. Romanov,D.,Kaufmann,G.,andDreybrodt,W.,2008, d 13 Cprofiles alonggrowthlayersofstalagmites:comparingtheoreticaland experimentalresults:GeochimicaetCosmochimicaActa,v.72, p.438–448. Rowe,P.J.,Dennis,P.F.,Atkinson,T.C.,Lauritzen,S.-E.,andLundberg J.,1998,Ahighresolutiondeuteriumrecordfromfluidinclusionsina lateHolocenespeleothemfromS.W.Britain, in Proceedings,PAGES OpenScienceMeeting,London,UniversityofLondon,112p. Schwarcz,H.P.,Harmon,R.S.,Thompson,P.,andFord,D.C.,1976, Stableisotopestudiesoffluidinclusionsinspeleothemsandtheir paleoclimaticsignificance:GeochimicaetCosmochimicaActa,v.40, p.657–665. Shen,C.,Edwards,R.L.,Cheng,H.,Dorale,J.A.,Thomas,R.B.,Moran, S.B.,Weinstein,S.E.,andEdmonds,H.N.,2002,Uraniumand thoriumisotopicandconcentrationmeasurementsbymagneticsector inductivelycoupledplasmamassspectrometry:ChemicalGeology, v.185,p.165–178. Spo ¨tl,C.,andMangini,A.,2002,StalagmitefromtheAustrianAlps revealsDansgaard-Oeschgereventsduringisotopestage3:ImplicationsfortheabsolutechronologyofGreenlandicecores:Earthand PlanetaryScienceLetters,v.203,p.507–518. Spo ¨tl,C.,Mangini,A.,andRichards,D.A.,2006,Chronologyand paleoenvironmentofMarineIsotopeStage3fromtwohigh-elevation speleothems,AustrianAlps:QuaternaryScienceReviews,v.25, p.1127–1136. Talma,A.S.,andVogel,J.C.,1992,LateQuaternarypaleotemperatures derivedfromaspeleothemfromCangoCaves,CapeProvince,South Africa:QuaternaryResearch,v.37,p.203–213. Treble,P.C.,Chappell,J.,Gagan,M.K.,McKeegan,K.D.,andHarrison, T.M.,2005,Insitumeasurementofseasonal d 18 Ovariationsand analysisofisotopictrendsinamodernspeleothemfromsouthwest Australia:EarthandPlanetaryScienceLetters,v.233,p.17–32. Vacco,D.A.,Clark,P.U.,Mix,A.C.,Cheng,H.,andEdwards,R.L., 2005,AspeleothemrecordofYoungerDryascooling,Klamath Mountains,Oregon,USA:QuaternaryResearch,v.64,p.249–256. Vaks,A.,Bar-Matthews,M.,Ayalon,A.,Schilman,B.,Gilmour,M., Hawkesworth,C.J.,Frumkin,A.,Kaufman,A.,andMatthews,A., 2003,Paleoclimatereconstructionbasedonthetimingofspeleothem growthandoxygenandcarbonisotopecompositioninacavelocated intherainshadowinIsrael:QuaternaryResearch,v.59,p.182–193. Vaks,A.,Bar-Matthews,M.,Ayalon,A.,Matthews,A.,Frumkin,Dayan, A.U.,Halicz,L.,Almogi-Labin,A.,andSchilman,B.,2006, PaleoclimateandlocationoftheborderbetweenMediterranean climateregionandtheSaharo—ArabianDesertasrevealedby speleothemsfromthenorthernNegevDesert,Israel:Earthand PlanetaryScienceLetters,v.249,p.384–399. Vonhof,H.B.,vanBreukelen,M.R.,Postma,O.,Rowe,P.J.,Atkinson, T.C.,andKroon,D.,2006,Acontinuous-flowcrushingdeviceforonline d 2 Hanalysisoffluidinclusionwaterinspeleothems:Rapid CommunicationsMassSpectrometry,v.20,p.2553–2558. Wang,Y.,Cheng,H.,Edwards,R.L.,An,Z.,Wu,J.,Shen,C.,andDorale, J.A.,2001,Ahigh-resolutionabsolute-datedlatePleistocenemonsoon recordfromHuluCave,China:Science,v.294,p.2345–2348. Wang,Y.,Cheng,H.,Edwards,R.L.,Kong,X.,Shao,X.,Chen,S.,Wu, J.,Jiang,X.,Wang,X.,andAn,Z.,2008,Millennial-andorbital-scale changesintheEastAsianmonsoonoverthepast224,000years: Nature,v.451,p.1090–1093. Wigley,T.M.L.,andBrown,M.C.,1976,Thephysicsofcaves, in Ford, T.D.,andCullingford,C.H.P.,eds.,TheScienceofSpeleology, London,AcademicPress,p.329–350. Williams,P.W.,Marshall,A.,Ford,D.C.,andJenkinson,A.V.,1999, PalaeoclimaticinterpretationofstableisotopedatafromHolocene speleothemsoftheWaitomoDistrict,NorthIsland,NewZealand: TheHolocene,v.9,p.649–657. Williams,P.W.,King,D.N.T.,Zhao,J.-X.,andCollerson,K.D.,2005, LatePleistocenetoHolocenecompositespeleothem 18 Oand 13 C chronologiesfromSouthIsland,NewZealand—didaglobalYounger Dryasreallyexist?:EarthandPlanetaryScienceLetters,v.230, p.301–317. Winograd,I.J.,Coplen,T.B.,Landwehr,J.M.,Riggs,A.C.,Ludwig, K.R.,Szabo,B.J.,Kolesar,P.T.,andRevesz,K.M.,1992,Continuous 500,000-YearClimateRecordfromVeinCalciteinDevilsHole, Nevada:Science,v.258,p.255–260. Xia,Q.,Zhao,J.,andCollerson,K.D.,2001,Early-MidHoloceneclimatic variationsinTasmania,Australia:multi-proxyrecordsinastalagmite fromLyndsCave:EarthandPlanetaryScienceLetters,v.194,p.177–187. Yonge,C.J.,1982,Stableisotopestudiesofwaterextractedfrom speleothems,[Ph.D.thesis],Hamilton,Canada,McMasterUniversity, 298p. Yonge,C.J.,Ford,D.C.,Gray,J.,andSchwarcz,H.P.,1985,Stableisotop e studiesofcaveseepagewater:ChemicalGeology,v.58,p.97–105. Yuan,D.,Cheng,H.,Edwards,R.L.,Dykoski,C.A.,Kelly,M.J.,Zhang, M.,Qing,J.,Lin,Y.,Wang,Y.,Wu,J.,Dorale,J.A.,An,Z.,andCai, Y.,2004,Timing,duration,andtransitionsofthelastinterglacial AsianMonsoon:Science,v.304,p.575–578. Zhang,M.,Yuan,D.,Lin,Y.,Cheng,H.,Qin,J.,andZhang,H.,2004, Therecordofpaleoclimaticchangefromstalagmitesandthe determinationofterminationIIinthesouthofGuizhouProvince, China:ScienceinChina(Ser.D),v.47,p.1–12. Zhang,M.,Cheng,H.,Yuan,D.,Zhu,X.,Lin,Y.,Qin,J.,andEdwards, R.L.,2006,Carbonandoxygenisotoperecordsandpaleoclimate reconstruction(140–250kaB.P.)fromastalagmiteofShuinanCave, Guilin,China:EnvironmentalGeology,v.49,p.752–764. Zhang,R.,Schwarcz,H.P.,Ford,D.C.,Schroeder,F.S.,andBeddows, P.A.,2008,Anabsolutepaleotemperaturerecordfrom10to6Ka inferredfromfluidinclusionD/Hratiosofastalagmitefrom VancouverIsland,BritishColumbia,Canada:GeochimicaetCosmochimicaActa,v.72,p.1014–1026. Zhou,H.,Zhao,J.,Feng,Y.,Gagan,M.K.,Zhou,G.,andYan,J.,2008, DistinctclimatechangesynchronouswithHeinricheventone, recordedbystableoxygenandcarbonisotopiccompositionsin stalagmitesfromChina:QuaternaryResearch,v.69,p.306–315. L IMITATIONSOFHENDYTESTCRITERIAINJUDGINGTHEPALEOCLIMATICSUITABILI TYOFSPELEOTHEMSANDTHENEEDFORREPLICATION 80 N JournalofCaveandKarstStudies, April2009



PAGE 1

SCUTTLEFLIES(DIPTERA:PHORIDAE)FROMCAVESIN MEGHALAYA,INDIA R.H ENRY L.D ISNEY DepartmentofZoology,UniversityofCambridge,DowningStreet,Cambrid geCB23EJ,U.K.,rhld2@hermes.cam.ac.uk Abstract: Fourscuttleflyspecies(familyPhoridae)werecollectedfromcavesin Meghalaya,India,ofwhichonewasfromdeepinsidethecaves,whiletheoth erswere restrictedtothevicinityofthecaveentrances.Illustratednotesarepr ovidedtoaidfuture recognitionofthesespecies. I NTRODUCTION Sincetheearly1990stheMeghalayanAdventurers Association(basedinShillong),inpartnershipwith Europeanspeleologists,hasconductedaseriesofexpeditionswiththeobjectiveofmappinganddocumentingcaves inMeghalaya,India.Todate,over280kmofcave passageshavebeensurveyed,butmuchmoreremainstobe discovered.ThequantityandlengthofcavesinMeghalaya exceedthatofanyotherknownkarstregionofIndia.Due toamajorexpansionofthelimestone-extractionindustry inrecentyearsintheJaintiaHills,thereisastrongcasefor documentingthebiospeleologyoftheregionbefore significantlossordamageoccurs.Duringthecourseof theseinvestigations,anumberofscuttleflies(Diptera: Phoridae)werecollected.Dr.DanHarries(SchoolofLife Sciences,Heriot-WattUniversity,Edinburgh)passedthese samplesontomeforidentification. M ETHODS ThespecimenswerepreservedinalcoholandsubsequentlymountedonslidesinBerleseFluid(Disney,2001). VoucherspecimenshavebeendepositedintheCambridge UniversityMuseumofZoology. T HE S PECIES Conicerakempi B RUNETTI 1924 Thisspecieswasdescribedfromthefemaleonly,butits malehassincebeendescribed(Disney,1982),itscritical featuresasshowninFigures1and2.AkeytotheOriental speciesof Conicera Meigen(Disney,1990b)onlyallows identificationofmalesandlikewiseusingthekeytothe speciesrecordedfromChina(Liu,2001). Femalescannormallyonlybenamedwhenassociated withtheirmales.Threespecieswhosefemaleshavebeen described(Disney,1990b;Ba ¨nzigerandDisney,2006) differfrom C.kempi inoneormoreofthefollowing features.Thelabrumisdevoidoflongitudinalridgesinthe medianbandormore.Thepostpedicelisclearlyshorter, and/ortheabdominaltergite6(T6)isofadifferentshape. Otherspecies,whosefemalesareunknown,canbe excludedbecausethereisahairatthebaseofvein3in theirmales.Thefemalesreportedbelowhavebeen comparedwithafemalefromNepalthathadpreviously beencomparedwiththetypematerialof C.kempi (Disney, 1982).Inthisspecies,whilethefemalepostpedicelisnotas longasinthemale(Fig.3),itisunusuallylongforafemale (Fig.4).ThelabrumisasshowninFigure5andT5andT6 areasshowninFigure6. Material Twentyfemales,6larvae,KremShynrongLabbit (25 u 21 1.2 N,92 u 30 105 E)collected 500mintothecave and200mbelowsurfaceonFebruary17,2001(KSL4and KSL4.2(larvae),CUMZ,38–55).Fivefemales,KremKotsati Lawan(25 u 10 46 N,92 u 22 29 E)collectedabout85mfrom thecaveentranceonFebruary18,2002(KUL2,CUMZ,38– 56).Twofemales,LiatPrah(25 u 22 31.5 N,92 u 32 18.6 E) collectedabout50mintothecave,February26,2002(LP5.4, CUMZ,38–55).Fivefemaleswerecollectedatthesame locality,butabout10mfromthecaveentrance,onFebruary 26,2002(LP5.2,CUMZ,38–56).Twentyfemaleswere collectedatthesamelocality,butabout1000mfromthecave entrance,onFebruary26,2002(LP3.1,CUMZ,38–56). NaturalHistory Brunetti(1924)recordedsixfemalescollected120– 150mfromtheentranceofSijuCave.Thenewmaterial confirmsthatthisspeciesisacavedweller.Furthermore,it wassometimescommonnearbatroosts.Matureeggsare about0.84-mm-longand0.27-mm-widewithasurface microsculptureofnumerousshortridges.Gravidfemales withabout50eggs.Mostspecimenshadaninfestationof wormsintheabdomen.Thesearebluntateachend, colorless,featureless(at240 3 magnification)andaremore than2-mm-longbutonlyabout0.001-mm-wide.Theylook likeminiaturehairworms,ratherthanNematoda. Diplonevra S PECIES IC Speciesrecognitionisprimarilybasedonthemalesexin thegenus Diplonevra Lioy.However,akeycoveringthe Orientalspecies(Disney,1990a)includedsomepoorly knownspeciesonlyknowninthefemalesex.Thespecies belowrunstocouplets11and12withtheformerbasedon speciesonlyknowninthefemalesex,andthelatterbased onlyonthemalesex.Incouplet11,thefemaleof D. R.HenryL.Disney–Scuttleflies(Diptera:Phoridae)fromcavesinMeghal aya,India. JournalofCaveandKarstStudies, v.71,no.1, p.81–85. JournalofCaveandKarstStudies, April2009 N 81

PAGE 2

evanescens (Brues)differsbyhavingamoreyellowishwing membraneandabroaderabdominaltergite3. D. fasciiventris (Brues),hasmuchdarkerhindfemoraand tibiae,andlacksahairatthebaseofvein3.Incouplet12,the femaleof D.assmuthi (Schmitz)wasdescribedunderits synonym D.ater (Brunetti).Itisimmediatelydistinguished byhavingitsdistiproboscisclearlylongerthanthebasiproboscis(seefigureinSchmitz,1931). D.varians Beyer(1958) wasdescribedfromaseriesofmalesfromBurma.The speciesbelowcouldbethehithertounknownfemaleofthis species.Untilthetwosexesareassociated,thiscannotbe resolved.Thespeciescharacterizedbelowcouldnotberun downinLiu’s(2001)keytoChinesespecies. Female Fronsorangebrowntomoreyellowishinthelowerpart withabout40smallhairs(whichareabsentfrommedian band)andminutemicrosetae.Thesupra-antennalbristles almostaslongasantials,whichareclosertoanterolaterals thantomidline,withtheALsbeingslightlyhigheronthe frons.Thebristlesofthemiddlerowalmostequallyspaced butmediolateralsalittleloweronfronsthanpre-ocellars.A short,butrobust,bristleoncheekandamuchlongeroneon jowl.Theyellowishbrownpostpedicelsaresmall(boththe lengthandgreatestbreadthbeingabout0.18mm).Thestraw yellowpalpsappearduskyindistalthirdormorebecauseof thedensedarkpubescence.Theymeasureabout0.41-mmlong,thesecondsegmentbeingabout0.36-mm-long(andits greatestbreadthbeingabout0.11mm),withfivebristlesat tipandupto30hairsbelow.Thebasiproboscisisyellowish brown,butwithnarrowdarkbrownbandsatsidesofthe basalhalf.Thelengthisabout0.59mmandthemaximum breadthisabout0.26mm.Thedistiprobosciscoloredas postpedicelsandabout0.50-mm-longand0.32mmatits greatestwidth.Thecombinedwidthofthepalelabellaisonly slightlygreater.Thoraxorangebrown,butdarkerontop. Eachsideofscutumwithahumeral,anotopleural,aprealar,anintra-alar,apostalar,andapre-scutellardorsocentralbristle.Scutellumwithfourbristles.Abdomendark greyishbrownapartfromthetergitesandtheyellowtipand cerci.Thetergitesaremainlyorangebrownwithminute hairs,butT1ismorestrawyellowinthemiddle.T2,as showninFigure7,withthemissinganteromedianbandgrey andtheanterolateralwingsstrawyellowuntiltheyencounter abrownpatchonthepleuralregion.T3(Fig.7)andT4 greatlyreducedandT5andT6absent.Legswithcoxae,mid andhindfemoraandhindtibialargelyorangebrown;the reststrawyellow.Fronttibiawithaneardorsalbristlenear middleand4–7shortspinesindistalthirdtoquarter.Front Figures3–6. Conicerakempi .3,male,rightpostpedicel;4, female,leftpostpedicel;5,female,labrum;6,female abdominaltergites5and6.Scalebars = 0.1mm. Figures1–2. Conicerakempi male.1,rightclasperof hypopygium;2,posteriorfaceofmidfemur.Scalebars = 0.1mm. S CUTTLEFLIES (D IPTERA :P HORIDAE ) FROMCAVESIN M EGHALAYA ,I NDIA 82 N JournalofCaveandKarstStudies, April2009

PAGE 3

tarsuswithposterodorsalhairpalisadesonsegments1–3and 5longerthan4.Midtibiawiththenormalpairofbristlesat endofbasalquarter,andapartfromthedorsalhairpalisade, thereisananterodorsalonethatendsatthestartofthedistal third,whichhasaseriesofanterodorsaltransversecombs. Hindfemurwithhairsbelowbasalhalfnotdifferentiated fromthoseofanteriorface.Hindtibiawithtwoanterodorsal bristles,onejustbeforetheendofthebasalthirdandthe otheralittlebeforetheendofthemiddlethird.Wingabout 1.8mmlong.Thecostalindexisabout0.6.Thecostalratios about6:3.5:1.Thecostalcilia(ofsection3)are0.07–0.08mm-long.Thethickveinsyellowishbrown.Veins4–6fine andgrey.Vein7verypale.Vein4originateswellbeforefork ofvein3andisevenlyconcavetowardsthefrontinthebasal fourfifthsandthenrunsstraighttothemargininthelast fifth.Screducedandonlyevidentatextremebase.Afinehair atbaseofvein3.Withfivebristlesonaxillaryridge. Membranealmostcolorless.Haltereknobbrown. Material Fivefemales,KremKotsatiLawan(25 u 10 46 N, 92 u 22 29 E)werecollectedatthecaveentrance.Collection wasonFebruary18,2002(KUL1,CUMZ,38–56). NaturalHistory Thisspecieswasonlyprocuredatacave’sentranceand wasprobablyonlyshelteringthere.Itisprobablynota specialistcavedweller.Matureeggswithasmoothsurface andabout0.87-mm-longand0.34-mm-wide.Gravid femaleswithabout22–24eggs. Megaseliamalaisei B EYER 1958 Thisspecieswasdescribedfromaseriesoffemalesfrom BurmaandhassubsequentlybeenreportedfromThailand inaflowerof Aristochiabaenzigeri HansenandPhuphathanaphong(Ba ¨nzigerandDisney,2006).Themale remainsunknown.Beyer’sdescriptionwasnotillustrated, sothedistinctiveabdominaltergites1–6areillustratedhere (Figs.8and10)andalsothepairofretractile,finger-like processesonthesidesofsegment6(Fig.9),whichBeyer failedtoobserve(astheyarelikelytohavebeenwithdrawn inapinnedspecimen). Material Onefemale,KremPyrda(25 u 20 29 N,92 u 29 23 E)was collectedabout50mintothecaveonFebruary10,2001 (KP3,CUMZ,38–55).Onefemale,LiatPrah(25 u 22 31.5 N, 92 u 32 18.6 E)wascollectedabout50mintothecaveon February26,2002(LP5.4,CUMZ,38–55). NaturalHistory Thisspecieswasonlyprocuredclosetotheentranceofthe cavesandwasprobablyonlyshelteringthere.Itisprobably notaspecialistcavedweller.Onefemaleretainedeightmature eggs,butthiswasprobablyanincompletebatchfollowingthe depositionofsomeeggs.Theseeggshaveasmoothsurface andareabout0.58-mm-longand0.22-mm-wide. Megaselia SPECIES IC Apartfromspeciesdescribedfromfemalesonlyinthe past(suchasthedistinctivespeciesabove),therecognition Figure7. Diplonevra speciesICfemale,abdominaltergites 2and3.Scalebar = 0.1mm. Figures8–10. Megaseliamalaisei female,detailsofabdomen;8,tergites5and6;9,rightretractilelobeofsegment6; 10,tergites1–4.Scalebars = 0.1mm. R.H ENRY L.D ISNEY JournalofCaveandKarstStudies, April2009 N 83

PAGE 4

ofspeciesinthegiantgenus Megaselia Rondaniisnow basedonthemalesforthefirsttime.Thespecies characterizedbelowcannotbenameduntilassociatedwith itsmale.InthekeycoveringtherelevantGroupVIIofthe Orientalspecies(Borgmeier,1967)itreadilyrunstocouplet 65,whichisbasedontherelativedarknessoftheantennal postpedicels.Takingthefirstoptionatcouplet66the distinctionisbasedonthemalesexonly.However,the femaleofitsfirstoption, M.palpella Beyer,issmaller(wing length 2mm),andithasawell-elongatedabdominal tergite6.Takingthesecondleadofcouplet66onethen proceedstocouplet70,to M.apposita Brues.Thelatter’s femaleisimmediatelydistinguishedbyitsyellow,as opposedtobrownishgrey,abdominalventer.Ifonetakes thesecondoptionofcouplet65,onereadilyproceedsto couplet89,wherethedistinctionisbasedonthedegreeof darkeningofthethorax.Thefirstoptionleadstocouplet 94to M.tetricifrons Beyer,whichhasthecostalsectionone lessthanthecombinedlengthsof2 3(asopposedtomore) andhalfasmanydifferentiatedposterodorsalhairsonthe hindtibia.Takingthesecondleadofcouplet89,oneruns tocouplet110,whereneitherleadappliesbutthefirst option, M.rutilipes Beyer(onlyknowninthemalesex)is closest,butithasadarkerfronsanddarkerlegs. Female Fronsbrown,butnotdark,andclearlybroaderthan long,with20–24hairsanddense,butveryfine,microsetae. Thesupra-antennalbristlesunequalinlength,thelowerpair beingalittleshorterandlessrobust.Theantialsslightly loweronfronsthananterolaterals,whichareaboutlevelwith theupperSAs,andclosertotheALsthantotheupperSAs. Pre-ocellarsaboutasfarapartaseitherisfroma mediolateralbristle,whichisveryslightlyhigheronfrons. Cheekwithfourbristlesandjowlwithtwolongerbristles. Thesubglobosepostpedicelslightbrown,about0.12-mmwide,andwithaboutthreedozenSPSvesicles,whicharea littlesmallerthansocketsoflowerSAbristles.Palpsare about0.21-mm-longand0.04-mm-wide,palestrawyellow, andwithfivebristlesand7–10hairs(ofwhich1–2arealittle strongerthantherest).Labrumalittledarkerthanpalpsand about0.45-mm-wide.Labellagreyishwhiteandatmostwith onlyadozenshortspinulesbelow.Thoraxbrown.Two notopleuralbristlesandnocleftinfrontofthese.Mesopleuronbare.Scutellumwithananteriorpairofhairs (subequaltothoseinmiddleofscutum)andaposteriorpair ofbristles.Abdominaltergitesbrown.T4–T7asshownin Figure11.Venterbrownishgrey,andwithsmallhairsbelow segments3–6.Sternite7aslongasT7butanarrowisosceles trianglewithtwolongerhairsatitsrearmarginandhalfa dozensmalleronesfurtherforward.Posterolaterallobesat rearofsternumeightshort,butbroadandwiththreelong hairsnearrearmargin.Cerciwhitishgreyandabouttwiceas longasbroad.Withfourrectalpapillae.Furcanotevident. Dufour’scropmechanismabout0.40-mm-long,butthe posteriorpartis0.15mmandcomprisesapairofpale divergentlobes.Thegreatestbreadthofanteriorpartis about0.25mm.Legsstrawyellowapartfrombrownpatch onmidcoxaandalightbrowntiptohindfemur.Foretarsus withposterodorsalhairpalisadeonsegments1–5and5 longerthan4.Dorsalhairpalisadeofmidtibiaextends almosttwothirdsofitslength.Hairsbelowbasalhalfofhind femurclearlylongerthanthoseofanteroventralrowofouter half.Hindtibiawith17–20differentiatedposterodorsalhairs andspinulesofapicalcombssimple.Wings2.0–2.4-mmlong.Costalindex0.54–0.56.Costalratios3.4–3.7:2.1–2.5:1. Costalcilia(ofsection3)0.07-mm-long.Nohairatbaseof vein3.With4–6axillarybristles,allofwhicharelongerthan costalcilia.ScstronganditstipfusedtoR1.Thickveins yellowishgrey,thinveinsgrey,but7paler.Vein4originates beyondforkofvein3andisdistinctlyrecurvedatitsbase. Membranelightlytingedgrey.Halterebrown. Material Onefemale,KremPyrda(25 u 20 29 N,92 u 29 23 E)was collectedabout50mintothecave,onFebruary10,2001 (KP3,CUMZ,38–55).Twofemales,KremKotsatiLawan (25 u 10 46 N,92 u 22 29 E)werecollectedatthecave entranceonFebruary18,2002(KUL1,CUMZ,38–56). Sixfemales,LiatPrah(25 u 22 31.5 N,92 u 32 18.6 E)were collectedabout10mintothecaveonFebruary26,2002 (LP5.4,CUMZ,38–54);16femaleswerecollectedatthe samelocalitybutabout50mintothecaveonFebruary26, 2002(LP5.2,CUMZ,38–56). NaturalHistory Thisspecieswasonlyprocuredclosetotheentranceof cavesandwasprobablyonlyshelteringthere.Itisprobably notaspecialistcavedweller.Matureeggswithasmooth surfaceandabout0.61-mm-longand0.24-mm-wide. Twentytotwenty-foureggsarematuredatatime. A CKNOWLEDGEMENTS ThanksareduetoNigelWyatt(NaturalHistory Museum,London)fortheloanofafemaleof Conicera kempi. MystudiesofPhoridaearecurrentlysupportedbya grantfromtheBalfour-BrowneTrustFund(Universityof Cambridge). Figure11. Megaselia speciesICfemale,abdominaltergites 4–7.Scalebar = 0.1mm. S CUTTLEFLIES (D IPTERA :P HORIDAE ) FROMCAVESIN M EGHALAYA ,I NDIA 84 N JournalofCaveandKarstStudies, April2009

PAGE 5

P OSTSCRIPT Sincetheacceptanceofthismanuscript,ageneral reviewofthesecaveshasbeenpublished(Harriesetal., 2008). R EFERENCES Beyer,E.,1958,DieerstenPhoridenvonBurma(Dipt.Phor.): CommentationesBiologicae,Helsingfors,v.18,p.3–72. Ba ¨nziger,H.,andDisney,R.H.L.,2006,Scuttleflies(Diptera:Phoridae) imprisonedby Aristolochiabaenzigeri (Aristolochiaceae)inThailand: MitteilungenderSchweizerischenEntomologischenGesellschaft, v.79,p.29–61. Borgmeier,T.,1967,StudiesonIndo-Australianphoridflies,basedmain ly onmaterialoftheMuseumofComparativeZoologyandtheUnited StatesNationalMuseum,PartII:StudiaEntomologica,Petropolis, v.10,p.81–276. Brunetti,E.,1924,DipteraoftheSijuCave,GaroHills,Assam.1. Tipulidae,Tabanidae,Anthomyidae,Acalyptratae,Muscidaeand Phoridae:RecordsoftheIndianMuseum,Calcutta,v.26,p.99–106. Disney,R.H.L.,1982,Theundescribedmaleof Conicerakempi Brunetti (Dipt.,Phoridae):Entomologist’sMonthlyMagazine,v.118, p.29–30. Disney,R.H.L.,1990a,Akeyto Diplonevra malesoftheAustralasianand OrientalRegions,includingtwonewspecies(Diptera,Phoridae): EntomologicaFennica,v.1,p.33–39. Disney,R.H.L.,1990b,ArevisedkeytoAustralasianandOriental Conicera (Diptera:Phoridae),withthreenewspecies:Entomologica Scandinavica,v.21,p.339–344. Disney,R.H.L.,2001,ThepreservationofsmallDiptera:Entomologist’s MonthlyMagazine,v.137,p.155–159. Harries,D.B.,Ware,F.J.,Fischer,C.W.,Biswas,J.,andKharpran-Daly, B.D.,2008,AreviewofthebiospeleologyofMegahalaya,India: JournalofCaveandKarstStudies,v.70,p.163–176. Liu,G.,2001,ATaxonomicStudyofChinesePhoridFliesDiptera: Phoridae(part1),China,Neupress. Schmitz,H.,1931,UeberdieGattung Phorynchus Brunetti:NatuurhistorischMaandblad,v.20,p.43–44. R.H ENRY L.D ISNEY JournalofCaveandKarstStudies, April2009 N 85



PAGE 1

PRESENT-DAYSEDIMENTARYFACIESINTHECOASTAL KARSTCAVESOFMALLORCAISLAND (WESTERNMEDITERRANEAN) J OAN J.F ORNO S 1 ,J OAQUI N G INE S 1 AND F RANCESC G RA ` CIA 1,2 Abstract: Inspiteoftheincreasingnumberofpapersoncavesedimentspublished duringthelastfewdecades,noonehasfocusedfromasedimentologicalpoi ntofviewon theprocessesthattakeplacespecificallywithinthecoastalkarstareas ofcarbonate islands.Theobjectiveofthepresentinvestigationsistodealwiththese dimentary processesthattakeplaceinsidetwolittoralcavesofMallorca(westernM editerranean), characterizingthedifferentfaciesexistingintheparticulargeologic al,geochemical,and hydrologicalsettingthatrepresentsthisveryspecificcavesedimentar yenvironment.The recentexplorationofextensiveunderwatergalleriesandchambersintos omeoutstanding coastalcavesoftheisland,haspermittedtherecognitionofimportantac cumulationsof present-daysedimentaryinfillingsintheirdrownedpassages.BoththeP irata-PontPiquetacavesystemandtheCovadesaGledahavefloorscoveredbymuddyand /or sandysedimentswhich,inawidesense,fitintotwowell-differentiatedc ategories.Onone handwehaveallochthonousreddishmudsediments(mainlysiliciclastic) andonthe otherhandautochthonousyellowishcarbonatemudorsands.Themixingofb oth materialsisalsofrequentaswellastheaccumulationoflargeblocksandd ebrisdueto thebreakdownofroofandcavewalls.Aseriesof21manualcoreswasobtaine dby scuba-diversinbothcaves,inordertocollectthefullthicknessofsedim entaryfill.Soil samplesattheentranceofthetwocaves,aswellasrocksamplesofthewalls ofboth sites,werealsoobtainedforalatercomparison.Severalsedimentaryfac iescanbe distinguished,whichincludecoarse-graineddeposits(entrancefacies andbreakdown blocks),fine-grainedsiliceoussediments(siltsandmuddydepositswit hveryvariable organicmattercontent),carbonatedepositscomposedofcalciteraftacc umulationsand/ orweathering-releasedlimestonegrains,andmixedfaciesincludingdiv erseproportions oftheothersedimenttypes.Therearealsosomerelictdepositscomposedo fsiliceousred silts,whichareaffectedbypolygonaldesiccationcracks.Inallthecase s,thesiliciclastic elements(quartzandfeldspars,mainly)arerelatedtoraineventssupply ingdustof Saharanorigin.Thedepositsandfaciesdescribedhereincorrespondtodi fferent sedimentaryenvironmentsthatcanbeindividualizedinsidethecaves(co llapseentrances, breakdownchambers,fullydrownedpassagesandchambers,poolswithfree water surface…),andreflectveryspecifichydrological,geochemical,andmec hanicalprocesses relatedtothecoastalnatureofthestudiedkarstcaves. I NTRODUCTION Duringthelastdecade,therehasbeenagreatincrease inspeleologicalandkarstresearchonMallorcaisland, especiallyinthoseperipheralareaswherethecoastalkarst attainsnoteworthydevelopment.Therecentexplorations ofextensiveunderwatergalleriesandchambersintosome outstandinglittoralcavesoftheislandareparticularly important(Gra `ciaetal.,2007a),allowingforthedetailed observationandsurveyofsometensofkilometersof drownedpassages.Theseinvestigationshavepermittedthe recognitionofthemorphologicalcharacteristicsofthe underwaterpartofthesecaves,whichexhibitanoutstandingvarietyofpresent-daysedimentaryinfillingsinmostof theexploredchambersandpassages(Gra `ciaetal.,2006, 2007b). Cavesedimentshavebeenrecognizedanddescribed sincescientificinterestincavesbegan.Nevertheless,wecan consider,ingeneral,thatin-depthinvestigationsonthe subjecthavenotbeenperformeduntilrecentlybykarst researchers.Whenscientistsrealizedthatsedimentscontain bothhydrogeologicalandpaleoclimaticalrecords,together withthedevelopmentofabsolutedatingtechniques,the studyofcavesedimentsbecomeoneofthemostinteresting topicsinkarstliterature(White,2007).Inspiteofthe increasingnumberofreferencesonthismatter,published duringthelastfewdecades,onlyafewarespecificsynthesis worksdealingwiththetopic(Ford,2001;Sasowskyand 2 GrupNordMallorca,Federacio Baleard’Espeleologia,xescgracia@yahoo.es 1 Dept.Cie `nciesdelaTerra.UniversitatdelesIllesBalears,Crta.Valldemossa,k m 7.5,07122Palma(BalearicIslands),joan.fornos@uib.es;jginesgracia @yahoo.es J.J.Forno s,J.Gine s,andF.Gra `cia–Present-daysedimentaryfaciesinthecoastalkarstcavesofMallor caisland (westernMediterranean). JournalofCaveandKarstStudies, v.71,no.1,p.86–99. 86 N JournalofCaveandKarstStudies, April2009

PAGE 2

Mylroie,2004).Inthemeantime,therehavealsoappeared somegeneraltextsthatincludereviewsofcurrent understandingofitshydrologicalandgeomorphological significance(FordandWilliams,2007;Gillieson,1996; Palmer,2007).However,amongthegreatnumberofrecent papersdevotedtocavesediments,noonehasfocusedfrom asedimentologicalpointofviewontheprocessesthattake placespecificallyinthecoastalkarstareasofcarbonate islands. Theflankmarginmodeloflittoralspeleogenesis, developedintheBahamasfromthe1980sonwards (MylroieandCarew,1990),recognizedtheimportanceof dissolutionprocessesoccurringinthemixingzonebetween freshandseawaters,whosepositioniscomplicatedby Quaternarysea-leveloscillationsduetoglacioeustatic phenomena.Thismodelhasevolvedduringthe1990suntil now(MylroieandMylroie,2007)leadingtotheelaborationofaCarbonateIslandKarstModel(CIKM),inwhich speleogenesisisstronglyconditionedbytheparticular behaviorofdiageneticimmature(eogenetic)carbonate rocks.TheCIKMprogressivelyintegratedmorecomponentsintothemodel,takingintoaccountboththepresence anddispositionofanimperviousbasement,ifitexists,as wellastheoverprintingoftectonicactivityandsea-level changes.Recently,Gine sandGine s(2007)focusedtheir workonthelittoralcavesofMallorcaisland,addingnew morphogeneticinsightstotheevolutionofthiseogenetic coastalkarst.Theseauthorsparticularlyemphasizetherole ofbreakdownprocesses,togetherwiththerecurrent glacioeustaticsea-leveloscillationsandthesubsequentfalls andrisesofthewatertable.Thiscyclicitywouldprovoke duringtheglacialperiodsthetriggeringofcollapsebythe lossofbuoyantsupportofphreaticwatersand,duringthe subsequentdrowningassociatedwithwarmevents,the underwaterdissolutionofbouldersandcollapsedebris. Thesemechanismswillenabletheenlargementofa nonintegratedarrayofcavesandvug-porosityconnected tothesea(developedinthemixingzone)ratherthan conventionalkarsticflowthroughconduits. Framedintothegeomorphologicalcontextabove described,theobjectiveofthepresentpaperistodeal withthesedimentaryprocessesthattakeplaceinsidethe coastalcavesofMallorca,characterizingthedifferent faciesexistingintheparticulargeomorphologicaland hydrologicalsettingsthatrepresentthisveryspecific undergroundsedimentaryenvironment.Inthissense,the peculiaritiesofcoastalkarstincludespecificgeochemical processeslinkedtothemixingzone(dissolution,dolomitizationÂ…)togetherwiththedecisivefactthatdrowningof cavepassagesiscontrolledbythesea-levelposition,instead ofbyalocalbaselevelplusthepossiblefloodsassociated witheventsofintenserechargeoftheaquifer.Asa consequenceofitshydrogeologicalbehaviorandthelow amountofprecipitation,thestabilityofthewater-table positioninMallorcanlittoralareasisremarkableevenata millennialtime-span,althoughatalongertime-scalethere wereimportantfluctuationsrelatedtoglacioeustatic variability.Ingeneraltermswearedealingwithavery lowenergyaquaticenvironment,framedinahigh permeabilitycontext(UpperMiocenecalcarenites)without afullykarstichydrogeologicalbehavior:rechargeislimited anddispersed,withnosinkingsuperficialstreamsorflood highwaters,andconduitflowisrelativelylessimportant thandiffuseflow. R EGIONAL G EOLOGICAL S ETTING Thetwostudiedcaves,Pirata-Pont-Piquetasystemand CovadesaGleda,aredevelopedinUpperMiocene limestonesthatcropoutalongthesouthernandeastern areasofMallorcaformingthebestfeaturedcoastalkarst regionoftheisland,calledMigjorn.Itslittorallandscapes arecharacterizedbysignificantphenomenaincluding differenttypesofkarsticand/ormarinecaves,paleokarst features,littoralkarrenandfluvio-karsticbightsorcoves. TheUpperMiocenelimestoneconstitutesapost-orogenic platformthatsurroundsthemountainranges(Serresde LlevantandSerradeTramuntana)builtupduringthe Alpineorogeny(Fig.1).Showingameanthicknessaround 70m,thatcanoccasionallyexceed120m,theyonlapa veryirregularalpinefoldedandthrustedbasement composedbyMesozoicdolomitesandlimestoneswith minormarlintercalations.TheUpperMiocenecarbonate sequencecorrespondstoanalternanceofsedimentary bodies(Pomar,1991)ofcalcilutitesandveryporous calcarenites,withacomplexgeometry,thatwereformed bytheprogradationofaTortonianreefcomplex.The sequenceendswithaseriesofcarbonatetabulardeposits withooliticandmangrovefacies,Messinianinage. TheUpperMiocenecarbonatesequenceformsaslab thatstretchesasaflatsurfacebehindthecoastal decametric-scalecliffsofthesouthernandeasternshores oftheisland.Thischaracteristictabularlandscapeis interruptedonlybyinciseddryvalleysfilledupby Holocenesediments,endingatlittoralcoveswhose presenceandmorphologyareconditionedbytheextensionalprocessesoccurringfromtheNeogenetothe Quaternary.Furthermore,theywereresponsibleforthe present-daycoastalmorphologythatiscontrolledby recentnormalfaults;alsonotableisasmalltiltingthat affectstheMigjornplateaualonganorth-southprofileand provokesthevariationofsedimentaryfaciescroppingout nearthesealevel. C AVE L OCATIONSAND D ESCRIPTIONS ThePirata-Pont-PiquetacavesystemislocatedatCan Frasquetfarm-house,inthesocalledMarinadeManacor (Migjornregion,easternMallorca),nearCalaFalco small bight(UTMED-50,525590/4373360-33)andsome 0.4kmfromthecoast.Itiscomposedofaseriesof independentlyexploredcavesthatwererecentlyconnected J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 87

PAGE 3

byunderwaterexploration(Gra `ciaetal,2006).Thesystem presentsaseriesofbreakdowndifferentiatedunits,threeof themcorrespondingtotheentrancechambers.Uptodate explorationsincludeaseriesofchambersandgalleries reaching3,020mofdevelopment,1,190mofwhichare underwater(Fig.2);thesurfaceareaoccupiedbysubterraneanpoolswithanairsurfacereaches5,000m 2 .Its maximumdepthis44m,including11mbelowthepresentdaysealevel.Thepassagesshowsolutionmorphologies thatarerestrictedtothosesectionslocatedbeneaththe currentphreaticlevel.Inthatsensethewatercolumnhas fourdifferentwellstratifiedwatermassesaccordingto theirsalinityandtemperaturethatconstitutearather complexmixingzoneshowingtwohaloclines(Gra `ciaetal., 2006).Thelinearpenetrationofthesecavesperpendicularlyfromthecoastlineisabout700m.Thespeleogenesis ofthesystemcorrespondstothemixingprocessesbetween continentalandmarinewatersthataffectedtheMiocene calcarenitesandprovokedtheinitialvoidcreation. Subsequentbreakdownprocesseswereinducedbythe glacioeustaticsea-levelfallsgivinglargeblockaccumulations,togetherwithspectacularspeleothemornamentation thatisobservedthroughmostofthecaves. TheCovadesaGledaisalsolocatedontheMarinade Manacor(Migjornregion)atSonJosepNoufarm-house (UTMED-50,523805/4372315-36),36mabovesealevel andsome1.7kmfromthecoast.Itcorrespondstoa subterraneancomplexofchambersandgalleriesrelatedto somecollapsesinkholesatthesurface,havingtodaya surveyeddevelopmentnear10,500mwithamaximum underwaterdepthof25m(itisthelargestlittoralmixingzonecaveinEurope).Themorphologicalframeofthecave showsaseriesofbreakdownchambers(Fig.2)whichare connectedtoeachotherbyphreaticgalleriesandpassages showingcircular,ellipticorirregularsections.Someofthe galleriesareclearlystructurallycontrolled.Thewaterprofile inthesubmergedpassagesshowsuptofivedifferentsaline layers,withdefinedhaloclinesthatgivevariedtypesof clearlymarkedcorrosionmorphologies(Gra `ciaetal., 2007b),themostinterestingbeingthecorrosionnotches thataffectboththecavewallsandancientspeleothems formedduringprevioussealevellow-stands.Thepresenceof speleothemsisanotableaspectofthiscave.Inparticular, somedrownedchambersandgalleriesshowimpressive precipitationmorphologiesformingbandsofcrystalcoatings onthecavewallsaswellasonpreviousvadosespeleothems. Figure1.GeologicalmapofMallorcadepictingthemaingeologicunitsoft heislandandthelocationofthestudiedcaves. P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 88 N JournalofCaveandKarstStudies, April2009

PAGE 4

Thesebandsaregeneratedbynoticeableepiaquaticcarbonateprecipitation,beingrelatedtopreviousstabilitylevelsof thewatertablewhichwereinturncontrolledbyQuaternary sea-leveloscillations(Tuccimeietal.,2006). BoththePirata-Pont-PiquetacavesystemandtheCova desaGledahaveimportantaccumulationsofsedimentsin theirchambersandpassages,thataretodaydrownedafter theHolocenesea-levelrise.Mostofthecavefloorsare coveredbymuddyand/orsandysediments,whichinawide sense,aremarkedbytwowelldifferentiatedcharacteristics. Ontheonehand,wehaveredmudsediments(mainly siliciclastic)andontheotherhand,ayellowishcarbonate Figure2.CavemapsofthePirata-Pont-PiquetasystemandCovadesaGleda( Manacor,Mallorca)showingthelocationof collectedcoresandsamples.Noticethatscalesaredifferentinbothmaps .SeecaveslocationonFig.1. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 89

PAGE 5

mudorsand.Themixingofbothmaterialsisalsofrequent, asistheaccumulationoflargeblocksanddebrisduetothe collapseandbreakdownofroofandcavewalls. M ETHODS Althoughpracticallyallthecavesknownintheeastern andsoutherncoastsofMallorcaislandshowsedimentsin theirfloodedpassages(CovaGenovesa[Gra `ciaetal., 2003];CovadesColl[Gra `ciaetal.,2005];CovadesÂ’O ` nix [Gine setal.,2007];amongothers),wehavefocusedour workinonlytwocoastalcavesthataretodaythemost representativeandwellknown. F IELD S AMPLE C OLLECTION Aseriesof20manualcores(Fig.2)wasobtainedby scuba-diversintheunderwaterpassagesofbothcaves(10at Pirata-Pont-Piquetasystemand10inCovadesaGleda)by introducinginthefloorsedimentsaPVCpipe,5.1cmin diameterand50-cm-long,untilthefullthicknessof loosesedimentaryfillwasattained(Fig.3).Soilsamplesat theentranceofthecavesaswellasrocksamplesfromthe wallsofbothcaveswerealsoobtainedforsubsequent comparison. Coresobtainedwerebagged,sealed,numbered,and broughtbacktotheEarthSciencesDepartmentofthe UniversitatdelesIllesBalears,wheretheywereopened, longitudinallysectioned,photographed,andsampledin stratigraphicorderaccordingtothedifferentobserved levels.Sampleswerenottakenatregularorfixedintervals duetothefactthatthescopeofthestudywasanapproach todeterminethedifferentsedimentaryfaciesexistingin theselittoralkarstcaves.Presenceofsedimentarystructuressuchaslaminationsandothergeneralobservations wereannotated. L ABORATORY A NALYSES Atotalof136samples(50fromPirata-Pont-Piqueta systemand86fromCovadesaGleda)weresenttothe laboratorywhereeachsedimentsamplewasair-driedfor 24hourspriortoanalysis.Afteracolordescription(dry andhumid)usingtheMUNSELLsoilcolorchart,grainsize,mineralogy,andorganicmatterwereanalyzed. Organicmatter(lostonignition[LOI])wasdetermined byweightlossafterplacingthesamplesinafurnaceat550 u Cforthreehours.ParticlesizedistributionwasdeterminedusingaBeckmanCoulter-LSparticlesizeanalyzer. Cumulativecurves,frequencyhistograms,andsummary statisticswerecalculated. MineralogywasdeterminedwithaSiemensD-5000Xraydiffractometer,usingrandomlyorientedpowdersofthe bulksamples.Sampleswerepre-treatedwithH 2 O 2 to removeorganicmatter.Replicateswereheatedto375or 600 u Cfor1hourortreatedwithethyleneglycolat60 u C todifferentiatebetweenclayminerals.Selectedsamples wereanalyzedbyEDX(BrukerX-FalshDetector4020)or observedbySEM(Hitachi E S-3400N).Semi-quantitative mineralanalyseswerebasedonthepeakareasobtained usingEVAver.7.0software. Radiocarbondateswereobtainedfromtwoorganic debris(seeds)inordertoestablishachronologicalframe forthestudiedcavesedimentaccumulation.Thedating wasperformedattheLaboratoireIRPAKIK(Institut RoyalduPatrimoineArtistique)ofBrussels.Analysesof sampledsedimentsmaybefoundontheNSSwebsite. Figure3.a)Samplingthroughmanualcoringinsidethe underwaterpassagesofthecaves;b)coresofsedimentary infillingfromCovadesaGleda.SeecorelocationonFig.2. P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 90 N JournalofCaveandKarstStudies, April2009

PAGE 6

R ESULTS Present-daysedimentsinthesampledcavesofthe Mallorcancoastalzoneshowaccumulationthicknesses rangingbetween0.5and1.5m,depositedonthefloorsof underwatersolutionalgalleries,smalltomediuminsize,as wellasinsubmergedbreakdownchambersofnotable dimensions.Thesedimentsshowaveryirregulardistributionmainlyrelatedtopresentorrelativelyrecententrances tothecaves,wherealightthinningtrendtowardstheinner partofthepassagescanbeobserved.Theirregular morphologyofthecavities,togetherwiththebreakdown processesthataffectmostofthechambers,controlsthe thicknessofthesediments. S EDIMENTARY F ACIES Inageneralwaywecandescribethepresenceofredto brownsiliciclasticsiltyandclayeysediments,pinkto yellowcarbonatesands,andcoarselithoclasticgravelsand brecciadeposits.Allthesedepositscanbegatheredinto fourdifferentcategoriesofsedimentsaccordingtotheir textural,organicandmineralogicalcompositionaswellas theirgenesis:(1)Gravityandwaterflowcoarse-grained deposits,(2)Waterlainclasticfine-graineddeposits,(3) Carbonatedeposits,and(4)Relictorolderdeposits (Fig.4)mostofwhicharecomposedofseveralsedimentaryfaciesthatareforthwithdescribed. GravityandWaterFlowCoarse-GrainedDeposits BreakdownFacies. Thisfaciesincludesunsortedboulders (ranginginsizefromseveralcentimeterstonear15m)and cobblepileswithnocementbindingtheclasts(Fig.5).Their compositioncorrespondsentirelytocalcarenitesand,in smallamount,calcisiltitescomingfromtheUpperMiocene rocksinwhichthecaveislocated.Grainsizeisrelatedtothe beddingthicknessofthesourcelimestones,beingusually frommediumtothick(decimetrictometric)oreven massive.Surfacetexturesoftheclastsareslightlyweathered dependingontheirdepthinthecavewatersandthediverse geochemicalbehaviorfoundthere.Clastsrangefromsubangulartosub-roundeddependingonthetexturalcharacteristicsoftheoriginalrock.Althoughtheyshowachaotic mixtureregardingsizeanddisposition,somepreferredclast orientationcanbeobserved.Theyareflattenedinshape, accordingtotherockbedding,andthelargestclastsshow theirflatsidesfacingparalleltothesurfaceaccumulationof Figure4.Sketchshowingthesedimentaryfaciesobservedinthestudiedli ttoralcaves. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 91

PAGE 7

thedebriscone.TheclastsdonÂ’tshowanyimbricated structure,oranyothersignoftransport. Thisfaciescorrespondstotheaccumulationofdebris fromroofaswellaswallcollapse,alteredbysolutional processesthattakeplaceinrelationshiptothehalocline. Chip,slaborevenblockbreakdowncanbefound. LooseEntranceFacies. Thesedepositshavesomesimilaritiestothebreakdownmaterialspreviouslydescribed. Locatednearthecurrententrancesorchokedopenings, thisfaciesisabrecciaaccumulationwithclasts,blocksand bouldersveryirregularinsizeandshape,mostlyshowinga sub-roundedmorphologyandhavingaslightinverse grading.Matrixcontentcanbeabundantandismainly formedbysiltsandveryfinesandcomingfromoutside,but withaveryirregulardistribution.Theyformprograding fansusuallywithahighdegreeofslope(Fig.5). Thisfaciesconsistsofamixingbetweenmainlyfinegrainedparticulatematerial,carriedbyexternalcurrents thatdraintotheentrancepoolsofthecaves,andfallen fragmentsofrockfromthecavewalls.Itisoneofthe coarsestfaciespresentinthesecoastalcaves,containingno wellroundedsedimentsduetotheabsenceofcurrentsof sufficientstrengthflowingintothelittoralaquiferandthe lackofallogeniccurrentsdrainingtothecaves.Wehave notobserveddebrisflowdeposits.ThesedepositscorrespondtotheentrancetalusofWhite(2007). WaterlainClasticFine-GrainedDeposits BrownOrganicSilts. Thebrownorganicsiltsfaciesisone ofthemostextensivefoundinthecaves,anditpresents variablethicknessesrangingfrom0.2mtonearly1m.This faciesisespeciallyrepresentedinthechambersandgalleries withopeningstothesurface.Theycorrespondtosiltwith lowproportionsofclay(around10%)andfinesands(less than5%).Withamoderatesorting,theyhaveameangrain sizecorrespondingtocoarsesiltandamedian(D50)offine silt(Fig.6).Samplestakeninthesedimentcoresshowa darkreddishbrowncolor(5YR3/4)whenmoistand reddishyellow(7.5YR6/6)whendriedinthelaboratory. Themostconspicuousfeatureofthesesedimentsisthe presenceofveryfinelaminationsclearlydefinedbythe alternationofreddish-brownandblacklaminae(Fig.7). Thesemillimeter-scaleblacklaminaeshowanotable accumulationofvegetalfibersandseeds.Themedian organiccontentofthefaciesisslightlyhigherthan5% (LOI).Fromthepointofviewoftheirmineralogyitcanbe consideredbasicallysiliceous,withahighproportionof Figure5.Pirata-Pont-Piquetacavesystem,a)geomorphologicalsection ofoneoftheentranceswithrelatedbreakdown processes;b)entrancetalusdeposits;c)breakdownpilesduetocollapse arepresentallaroundthecaves. P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 92 N JournalofCaveandKarstStudies, April2009

PAGE 8

quartz(morethan65%),clayminerals(around20%)and feldspars(near10%).Carbonates(mainlycalcite)are presentinverylowproportions(lessthan5%). Thesedimentsthatcorrespondtothisfaciesare accumulatedmainlyinthosepoolsandunderwater passagesthatarelocatednearestthegreatcollapse entrancestothecaves.Sedimentcompositioncorresponds tosoilmaterialstransportedinsidethecavesbyflowing rainwater,especiallyduringstormevents.Theircompositionhasbeentraditionallyrelatedtoallochthonous siliciclasticcomponentssuppliedbyrainscarryingdown dustparticlesofSaharanorigin(Fioletal.,2005,Goudie andMiddleton,2001).Theclearlaminationsduetothe periodicaccumulationoforganicmatter,includingabundantvegetalfibersandseeds,arerelatedtoeachoneofthe successivestormeventsenteringthecave,andmaynotin facthaveanyseasonalsignificance.Theyalsodonothave anyrelationwiththevarvedepositssoclassicalinglaciated regions.Carbon-14datingofsomeorganiccomponents (seedsandvegetalfibers)fromcoresPP-08andGL-01 Figure6.LOI(lostonignition)percentage,color,mineralogicalcompos ition,andtextureofthesedimentaryfaciesdescribed. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 93

PAGE 9

yieldedagesof220and330yrBP,respectively(Forno s andGra `cia,2007). ReddishSiltsandClays. Thisfaciesistheleastcommonin thestudiedcaves,inextensionaswellasinthickness. Localizedinthecorescollectedintheinnerpassages,it showsathicknessthatscarcelyreaches20cm,showinga darkredcolor(2.5YR4/6)whenmoistthattransformsto yellowishred(5YR5/8)whendriedinthelaboratory (Fig.8).Thepresenceoforganicmatterisreduced(LOI lessthan4%).Texturally(Fig.6)thefaciescorrespondsto silt(81%)withasmallproportionofclayfraction(19%). Thesandfractionisabsent.Itshowsgoodsorting.Their mineralogicalcompositionisformedofsiliceousminerals withquartzrepresentingmorethan75%,feldsparswith morethan10%,andclaymineralsthatpracticallynever reach10%.Presenceofcarbonatemineralsisverylow (mainlycalcitewithlessthan3%). Beingrestrictedtotheinnerpartofthecaves,faraway fromthemainentrancesaswellasinplaceswithlimited connectionswiththesurface,thisfaciesrepresentsthefinegrainedsedimentsthatcanbetransportedassuspended loaddeepintothecavesystem.Asinthecaseofthe organicbrownsiltsfacies,themuddyreddishsiltsare integratedbyabundantallochthonoussilicicsediments relatedtodustrainscomingfromAfrica,whichwere progressivelyaccumulatedintothesoilmantle. CarbonateDeposits CalciteRaftSands. Thethicknessofthisfaciesvariesfrom afewcentimeterstomorethan40cminthesampled chambers.Thesanddepositsshowareddishyellowcolor (7.5YR7/8moistto7.5YR7/6dry).Beddingispoorly definedbytheexistenceofsub-millimeterscalehorizontal Figure7.Brownorganicsilts;a)Generalaspectofcore PP07collectedatGaleriadelLlacRasinPirata-PontPiquetacavesystem;b)conspicuousthin-laminationsdueto grain-sizesortingandorganicaccumulation;c)vegetalfibers arethemainconstituentsoftheblacklaminae. Figure8.Reddishmud;a)corePP06collectedattheendof GaleriadelLlacRasinthePirata-Pont-Piquetacavesystem showingamassiveaspectwithfaintlamination;b)detailed SEMimageofthesedimentsmainlycomposedofquartz grainsandclays. P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 94 N JournalofCaveandKarstStudies, April2009

PAGE 10

laminationresultingfromthesubtlepresenceofredmuddy particles.Textureismoderatelysortedandclearlydominatedbythesands( 78%)withsmallamountsofsilts representingslightlymorethan18%,andsomeclay(less than3%).Thinlimestoneangularparticlesarescarce. Meangrain-sizeaswellasthemedianvaluecorrespondto thetransitionfromfinetomediumsandsize(Fig.6).The organiccontentisverylow(LOIlessthan1%).Grain mineralogyisdominatedbyacarbonatecompositionwith calcite( 85%)andsomedolomite( 2%);therest correspondstoquartz(12%)andscarceclayminerals( 1%). Sandgrainsareconstitutedofcompositerhombohedral calcitecrystalsshowingacleardifferentialgeometric growththatstartswithmicrosparitecrystals,forminga planarsurfacethatcoincideswiththesurfaceofthewater, andevolvestosparitecrystalsinsidethewaterpool.Most ofthecompositecrystalssandgrainswhenfoundonthe bottomofthepoolspresentclearcorrosionmorphologies relatedtothegeochemicalprocesses(aggressivenesslinked tothehaloclines)occurringthroughthewatercolumn (Fig.9). Thiskindofsedimentisespeciallyabundantinthe bottomofthosecavepoolshavingafreewatersurfacethat allowsforCO 2 degassing.Thisprocesscontrolsthe precipitationofcalciteraftsatthesurfaceofthepools wheretheyaremaintainedfloatingbymeansofsurface tension,untilsomeexternalprocessortheirowngrowth triggerstheirsinkingandafinalaccumulationatthe bottomofthepool. CarbonateGrainsReleasedbyPhysico-ChemicalWeatheringofLimestoneWalls. Thicknessofthissedimentfaciesis veryvariable,rangingfromafewcentimeterstonearly20 cm,withdistributionmainlyrelatedtotheproximitytothe baseofcavewallsorrockprotuberancesingeneral.Color isconsistentlyreddishyellow(7.5YR6/6).Theirorganic mattercompositionisvariablewithmeanvaluesof differentcoresrangingbetween3and4%(LOI).The texturalcharacteristicsarealsoquitevariable,usually givingbimodalcurvesandshowingverypoorsorting (Fig.6).Themeangrainsizecorrespondstothefinesand fractionandthemedian(D50)toveryfinesand.Thesand fractionrepresentsnearly60%ofthetotalcomposition, whilethesiltfractionisgreaterthan35%;theclayfraction showvaluesaround5%.Fromamineralogicalpointof view,theircompositionismainlycarbonate.Calcite representsupto80%anddolomitearound8%.Siliceous componentsareformedbyquartzreachingnear7%of meanvaluesandlowpercentagesofclayminerals(lessthan 4%). Thesecarbonateparticulatedepositsarethecoarsest fractioninthecavesystemsapartfromthoseproducedby thegravityandbreakdownprocesses.Theyaremainly composedofcarbonaterockparticlesdetachedfromthe cavewalls(Fig.10),duetothedifferentialresponseto weatheringandcorrosionofthevariegatedbioclastic grains(differingbothinmineralogyandtexture)that composetheenvelopingcalcareniterock.Thisfaciesis particularlyabundantwherethecurrenthaloclinesare presentandcarbonategrainsarereleasedbymixing corrosion. MixedFacies Mixedfaciesarepresentallalongtheentirecave systems.Theyconstituteamixtureoftheabovedescribed facies,includingbothsiliciclasticandcarbonatesediments indifferentproportions(Fig.11).Thequantitativecharacteristicsofthemixturearerelatedtothelocationinthe cave,presenceofsurfaceopenings,depthofthedrowned areas,distancetothesea,presenceofpoolswithfreeair surface,etc.Theresultingtextureandcompositiondiffer fromtheendmemberscorrespondingtothedifferentfacies whichhavebeencharacterizedinthepreviousparagraphs (Fig.6).Fromthemineralogicalpointofview,thesefacies alsocorrespondtoaveryvariablemixtureofsiliceous componentsandcarbonategrains,includingbothcrystal aggregatesaswellasparticulatematerial. Figure9.a)Conicalstructureduetotheaccumulationof sunkencalciterafts;b)calcitecrystalscorrespondingto calciteraftsaffectedbycorrosioninthemixedzoneonce theyaresunktothebottomofthepools. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 95

PAGE 11

Colorisveryvariableembracingshadesofreddish yellow(7.5YR7/8)toduskyred(2.5Y3/3)orevenbrown (10YR5/3).Organicmatteralsohasagreatvariability, rangingfrom1%tomorethan10%.Usuallythetextural curvesshowbimodalityandpoortoverypoordegreesof sorting.Meangrainsizerangesfromfinesandtocoarse silt,whilethemedianvariesbetweenverycoarsesiltto mediumsilt.Texturalfractionsshowfiguresbetween14% and40%forsand,between50%and60%forsiltand between7%and25%fortheclayfraction.Themineral compositionalsohasagreatvariability.Calciteranges betweenmorethan90%tolessthan30%,dolomitefrom 3%to6%,quartzfrommorethan40%tolessthan3%; feldsparshaveamaximumof10%,butcanbealsoabsent insomesamplesand,finally,theclaymineralsrangefrom 1%toslightlymorethan10%. RelictorOlderDeposits RedSilts. Ancientdepositscomposedofredsiltsareupto 15cmthickinsomesampledchambers.Thecolorisdark red(2.5YR4/6)whenmoistandyellowishred(5YR5/6) whendry.Beddingisverypoorlydefinedbutlocallythere isamillimeter-scaleblacklamination,resultingfrom organic-matterconcentration.Onthesurfaceofthese sedimentsdecimeter-scalemuddesiccationcracks(upto4 cmdeep)areclearlyvisible(Fig.12).Themeanorganic content(LOI)isaround2%.Texturallythisfacies representsamixingofsilts(52%)andclays(47%)witha verylowpresenceofsandmaterial(lessthan1%);itshows averygoodsorting.Themeangrain-sizecorrespondsto veryfinesiltandthemedian(D50)liesbetweensiltand clayfractions.Mineralogicalcontentisdominatedby quartz( 43%)andclays(mainlyilliteandkaolinitewith somemontmorillonite)withvaluesabove33%.Feldspar mineralsrepresentslightlylessthan13%,calcitenear9% anddolomite2%. Thissedimentaryfaciesiscomposedofclayandsilt particlesthataccumulateinpondedareasofpassages receivingepisodicslow-movingstormwaterinputfromthe collapseentrances.Thepresenceofabundantmudcracks splittingthesedimentationintopolygonalblocksonthe topoftheseredsiltymaterialsindicatesadryingperiod occurredafterthedepositionofsuchmudbyslow-moving water.Theexistenceofmudcracksonthesurfaceofthese sediments(todaylocatedinunderwaterpassagesextending wellbelowthesealevel)clearlyreflectsthefactthatthey correspondtoanearlyinfillingofthecavelinkedtosome Pleistocenesea-levelfall,probablycorrespondingtothe lastglacialevent. Figure10.IncongruentcorrosionoftheMiocenecalcarenitesrelatedtothemixingzonegiveswaytoagranular disintegrationproducingabundantcarbonatesediment;a) sandygrainsizerainofsediment;b)coreGL10fromGaleria delsDegotissosinCovadesaGledashowingthecoarsest sedimentsinthecavesystems,beingmainlycomposedof heterometriccarbonaterockparticles. Figure11.Carbonate,silicicormixedsedimentspresently extensivelycoveringthefloorofthecaves(Galeriadeles Co `niques,CovadesaGleda). Figure12.Sedimentsverysimilartopresent-dayreddish mud,butpresentingdesiccationcracksarepresentinsome places( 4m,GaleriadelLlacRas,Pirata-Pont-Piqueta cavesystem).Theirdepositionmustcorrespondtosomesealevelfallevent,probablytothelastglaciation. P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 96 N JournalofCaveandKarstStudies, April2009

PAGE 12

CoarseDetritalandGravityFallenDeposits. Thesematerialsareanalogoustothecoarse-graineddeposits previouslydescribed(mainlygravityemplaced)butwitha genesisrelatedtopreviousepisodesofbreakdowninthe cave.Theyaretheresultofautogenicbreakdownprocesses favoredbytherepetitivelossofhydraulicbuoyancyduring Pleistoceneglacialperiodsandthesubsequentsea-level falls.Usuallytheyarecoveredwithathinveilofmore recentsediments,whichcanbefoundalsoasmatrix infiltratesbetweenthebrecciaclasts. D ISCUSSION SedimentationinsidethecoastalkarstcavesofMallorcacomprisesbothautochthonousandallochthonous componentsprovidedbyarelativelywidediversityof geomorphicmechanismsactinginthedifferentaquatic environmentsexistingalongthecavesystems.Theprocessesinvolvedvaryfromrockcollapsetofine-grained waterlainclasticsediments,includingdiversemechanisms suchasunderwaterweatheringofrocksurfaces,precipitationofcalcitecrystals,andinfiltrationofsoilmaterials (Fig.13).Investigationscarriedoutontwoofthemost importantcoastalcavesoftheislandhaveallowedthe characterizationofseveraldifferentiatedsedimentfacies, whichclearlycorrespondtospecificlocationswithinthe wholecavesystem.Theproximitytothecaveentrancesor blockedopeningstothesurface,aswellasthegeochemical behaviorofthephreaticwaters(dissolution,precipitation, etc.),iscrucialinordertoexplainthecavesediments distributiontogetherwiththeparticulartopographyofthe underwatergalleriesandchambers. I NTERPRETATIONOFTHE C AVE S EDIMENTARY R ECORD Breakdownorcollapsehasbeenoneofthemost invokedprocessesoflaterevolutioninthedevelopmentof thecoastalcavesofMallorca(Gra `ciaetal.2006,2007a, 2007b).Therepeatedfloodingandemptyingoftheoriginal phreaticpassages,duringthePleistoceneglacio-eustatic oscillations,causedthepassagestobealteredinshapeand dimensionsbybreakdownprocesses,especiallywhenthe cavesweredrainedduringsea-levelfalls,andfavoringthe failureofroofsandwallsparticularlyalongthebedding planesandjoints(Gillieson,1996;Gine sandGine s,2007). Themaximumsea-levelfallduringthelastglacialperiod wasaround135mbelowpresent-daydatum(Lambeckand Chappell,2001;LambeckandPurcell,2005);thisfigure, togetherwiththerathermodestdepthattainedbythecaves (maximum25mbelowthecurrentsealevel)duetothe presenceofimperviousfaciesatthebaseoftheMiocene sequence,suggeststhatalltheknownchambersand passageshadbeenemptiedseveraltimesduringthe Pleistocenecoldperiods,favoringinthatwaytheactuation ofcollapseprocesses. Mostoftheentrancestothecavescorrespondto collapsedroofsthattodayactascollectorsofsomesurface runoff,andrepresentpreferentialsinkingpointsthat quicklydrainprecipitationtowardsthewatertable. Dependingonthestrengthofthestormeventsandthe resultingrunoff,manytypesofsedimentarymaterialcan betransporteddownslopebyepisodicwaterflows. Furthermore,theprocessesrelatedtothebreakdownfacies arealsopresentinthecaveopenings,whereintensive weatheringoftherockwallsalsotakesplace.Allthese materials,aswellasparticleserodedfromthesoil,are incorporatedinthelooseentrancefacies,andafteramore orlessprolongedsheetflowtransportmaybedepositedat somecavepoolsneartheentrancesformingadelta-like architecture.Thus,loosematerialscanaccumulateinthe entrancechamberssimplybygravityorbybeingcarriedin flowingwater.Theresultisamixingofgravels,sands,and siltymaterialdepositedinthecavesbytherunofffromthe landsurfacetogetherwithweatheredelementsfromthe cavewallsorceilingsaswellassoilparticles.TheMigjorn region,wherethecavesarefound,isanautogenickarst mantledwithonlyadiscontinuousthinsoil.Thecatchment areaisrelativelyminor(somesquarekilometers)sothe volumeofexternaldetritalsedimentisrathersmall. Episodicwaterflowsthatenterthecavesthroughthe existingcollapseopeningsafterintenseperiodsofrainfall, whichcharacterizetheMediterraneanweatherparticularly inautumn(Guijarro,1986),transportheterometricmaterialsuntilreachingthewatertableatthepondsinsidethe caves.Thecoarsestelementsaccumulateattheentrance Figure13.Sedimentaryfaciesandinvolvedprocessesinthe coastalkarstcavesofMallorcaisland. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 97

PAGE 13

poolsformingpseudodeltas,ashavebeenmentioned above,whilethemuddyfinestmaterialisintroducedinside thesubmergedgalleriesofthecavewherethesedimentary accumulationtakesplacemostlybydecantationaccording tothehydrauliccharacteristicsofthewaterbodyandthe grainsizepropertiesofthetransportedparticles.Clayand siltaccumulatethusindrownedinnerpassageswherethere isaslow-movingfloodwatertransport.Thedistanceinside thecavethatsedimentcanreachwilldependonthe geometryandmorphologyofthecavesystemandthe initialimpulseofthewaterfloodingafterheavystorms,as wellastheinteractionwiththeslowwatermovementinside theaquiferinducedbythetidal(and/orbarometric)sealeveloscillations.Inthismanner,differentfaciesare depositedvaryingmainlyingrain-sizeandorganicmatter content.Asaresult,theinnerfinesedimentationisclayey andwithveryloworganic-mattercontent(reddishmud facies),whereassedimentsemplacednearestthecollapse openingsaresiltyandusuallyshowveryclearblack laminationsoforganicmatter,andarerelatedtothe successivestormeventsthatsupplymaterialstothe entrancepools(brownorganicsilts).Carbon-14datings oforganiccomponents(220and330yr)confirmthe subactualcharacterofthesedeposits. Mineralogyofthefine-grainedsediments(bothbrown organicsiltsandreddishmud)ismainlysiliceous,including quartz,feldsparandclayminerals.Thisfactpointstoan allochthonousoriginrelatedtorains,comingfromthesouth orsouth-east,thatsupplysilicicdustofSaharansource; theseatmosphericsituationsarerelativelycommoninthe westernMediterraneanwhenacenteroflowpressure remainssituatedovertheIberianpeninsula(Fioletal., 2005).Althoughsomeoftheinsolubleresiduefromthe dissolutionofbedrockcancontributetothedetrital sedimentationinthecaves,theanalysisdoneonthe surroundingrocksshowsthattheirnon-carbonateimpuritiesarelessthan1%.Neverthelessthereisagreatamountof carbonategranular(sandsized)materialthathasbeen producedbytheweatheringanddisaggregationofthe bedrock,producingdetachedparticlesthatareeventually accumulatedsimplybyfallingintothesubmergedpassages andchambers.Thesecarbonategrainsarereleasedbythe differentialphysico-chemicalweatheringoflimestoneor dolomitefromthebedrock,aswellasbythedifferent solubilityofbioclasticgrainsaccordingtotheirmineralogy andinrelationtotheirbiologicalcomposition.Inthatway, theproductionofreleasedcarbonategrainsisfavoredowing totheincorporationoftheseparticulatematerialsintothe sedimentaryrecordofthecavesystem.Althoughthese processesaremosttypicalofcavesthatcontainlittleorno runningwater,beingexposedtolong-termweatheringinair (Palmer,2007),inlittoralcaves,asisourcase,weatheringin thisphreaticenvironmentcanbeenhancedbythegeochemicalactivityofthecoastalmixingzone. Depositsconstitutedbycalciteraftsandsarelinkedto thosecavepoolswithfreewatersurface,thatallowsfor CO 2 degassingandcarbonateprecipitation.Whenthese calciteraftsaresunkbydrippingwater,orforwhatever reason,theyaccumulateascoarsesandsofloosecrystal aggregatesofcalcitethatcanreachnotablethickness. Whensubmergedanddepositedinthebottomofthepools, thecalcitecrystalsmaypresentdissolutiontracesrelatedto thegeochemicalprocessesoccurringinthehaloclines.The socalledmixedfaciesarethemostabundantsedimentsin thestudiedcaves,beingamixtureofallthepreviously describedmaterialsindifferentproportions.Grainsizeis normallybimodal(sandandsiltfractionsarepredominant) andmineralogyisverydiverse,includingbothallochthonoussiliceousgrainstogetherwithcarbonateparticles releasedbydissolutionand/orcalciteraftcrystals.Their distributionandimportanceareveryvariableanddepend mainlyontherelationshipbetweenmainentrancestothe caves,presenceofpoolswithfree-airsurfaces,anddepthof thehaloclinesinthesubmergedgalleries. Inspiteofthepreviouslydiscussedspecificityofthe coastalkarstenvironment,mostoftheobservedsediment typescorrespondinbroadtermstothebackswampfacies ofBoschandWhite(2004)becausetheyareconstitutedby avarietyofinfiltratedsoilmaterials,weatheringresidues, andcarbonateprecipitateshavingexperiencedverylimited lateraltransport.Ontheotherhand,thereddishmud depositspresentintheinnerpartsofthecavesystemsshow someanalogiestotheslackwaterfacies(BoschandWhite, 2004),butframedinalow-energeticcontextlinkedtothe coastalkarsthydrologicalbehaviorquitedifferentfrom conventionalstreamcaves. Relictorancientsedimentsarealsopresentinthe studiedcaves,beingverysimilartopresent-dayreddish muddeposits,butshowingdesiccationcracksthatsituate theirdepositioninsomesea-levelfallevent(perhapsthe LastGlaciation).Mineralogypredominantlyincludes allochthonoussiliceousgrains,suppliedbydustrainsof Africanorigin.Anadequateinterpretationoftheseancient sedimentationphaseswillstronglybenefitfromthe investigationofthepresent-daydepositsfromthecoastal cavesoftheisland. C ONCLUSIONS Severalmainconclusionscanbedrawninreferencetothe sedimentaryprocessesoccurringinthisspecificcave environmentofthecoastalkarstsystem.ClasticsedimentationinthecoastalcavesofMallorcaislandisfully conditionedbythehydrogeologicalspecificityofthelittoral karstificationaffectinghighpermeabilityrocks(calcarenites UpperMioceneinage).Noallogenicstreamsarepresentin theMigjornkarstregion,andthecoastalaquifersare characterizedbyaratherdiffuseflowtothesea.Nevertheless,animportantfactorinthedistributionofsedimentis episodicinfluxofmeteoricwaterduringstormevents. Severalfaciescanbefoundinthecavesystems,which canbegroupedintogravityandwaterflowcoarse-grained P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 98 N JournalofCaveandKarstStudies, April2009

PAGE 14

deposits,waterlainclasticfine-graineddeposits,carbonate depositsandmixedfacies.Relictdepositsarealsopresent, beingrelatedtothemorphologicalevolutionofthecave duringthePleistocene. Thediversityofrepresentedfaciesisrelatedtothe topographicalandgeomorphologicalpositionofsampled pointswithinthecavesystem.Breakdownprocessesof undergroundpassages,triggeredbyPleistocenesea-level oscillations,giveplacetocollapseentrancesandchoked openingsfavoringthepresenceofentrancefacies,andthe penetrationofsedimentscarriedbyrunoffwaters.Brown organicsiltfaciesareparticularlywidespreadnearpresentdayentrancestothecaves.Theabundanceofsiliceous sedimentsinthefine-graineddepositsmustberelatedto dustrainsofSaharanorigin,whicharefrequentinthe WesternMediterranean.Particlesofcarbonatematerialare importantconstituentsofsomedeposits,intheformof rockgrainsreleasedbydifferentialdissolutionweathering orsunkencalciterafts.Itisverycommontohavethe presenceofmixedfaciesthatincludeallochthonous siliciclasticsiltsandmudtogetherwithautochthonous carbonateparticles.Coastalkarstcaveshostaveryspecific sedimentassemblagethatrequiresincreasingattention. A CKNOWLEDGEMENTS Thisworkwaspartiallysupportedbytheresearch fundofMinisteriodeEducacio nyCiencia–FEDER, CGL2006-11242-C03-01/BTE.WethanktoF.Hierro (SEM),J.Cifre(X-rayanalysis),andM.Guart(grain-size analysis)forhelpinginlabanalysis.Theauthorsare gratefultoB.Clamor,M.FebrerandP.Gamund fortheir helpduringthefieldsamplingandA.Gine sfortheir constructivecommentsandhelpinwritingthemanuscript. R EFERENCES Bosch,R.F.,andWhite,W.B.,2004,Lithofaciesandtransportofclastic sedimentsinkarsticaquifers, in Sasowsky,I.D.,andMylroie,J.,eds., Studiesofcavesediments:Physicalandchemicalrecordsof paleoclimate:NewYork,KluwerAcademic/PlenumPublishers, p.1–22. Fiol,L.,Forno s,J.J.,Gelabert,B.,andGuijarro,J.A.,2005,Dustrainsin Mallorca(WesternMediterranean):Theiroccurrenceandroleinsome recentgeologicalprocesses:Catena,v.63,p.64–84. Ford,D.C.,andWilliams,P.W.,2007,Karsthydrogeologyand geomorphology:Chichester,Wiley,561p. Ford,T.D.,2001,Sedimentsincaves:Somerset,U.K.,BCRACave StudiesSeries,no.9,32p. Forno s,J.J.,andGra `cia,F.,2007,Datacio delssedimentsrecentsque rebleixenlescavitatsdesaGledaidelsistemaPirata-Pont-Piqueta: Primeresdades:Endins,v.31,p.97–100. Gillieson,D.,1996,Caves.Processes,Development,Management: Camridge,BlackwellPublishers,324p. Gine s,A.,andGine s,J.,2007,Eogenetickarst,glacioeustaticcavepools andanchialineenvironmentsonMallorcaIsland:Adiscussionof coastalspeleogenesis:InternationalJournalofSpeleology,v.36, no.2,p.57–67. Gine s,J.,Forno s,J.J.,Trias,M.,Gine s,A.,andSantandreu,G.,2007, Elsfeno `mensendoca `rsticsdelazonadecan’Olesa:lacovade s’O ` nixialtrescavitatsve ¨ nes(Manacor,Mallorca):Endins,v.31, p.5–30. Goudie,A.S.,andMiddleton,N.J.,2001,Saharanduststorms:Nature andconsequences:Earth-ScienceReviews,v.56,no.1–4,p.179–204. Gra `cia,F.,Clamor,B.,Forno s,J.J.,Jaume,D.,andFebrer,M.,2006,El sistemaPirata-Pont-Piqueta(Manacor,Mallorca):Geomorfologia, espeleoge `nesi,hidrologia,sedimentologiaifauna:Endins,v.29, p.25–64. Gra `cia,F.,Clamor,B.,Jaume,D.,Forno s,J.J.,Uriz,M.J.,Mart n,D., Gil,J.,Gracia,P.,Febrer,M.,andPons,G.,2005,LacovadesColl (Felanitx,Mallorca):Espeleoge `nesi,geomorfologia,hidrologia,sedimentologia,faunaiconservacio :Endins,v.27,p.141–186. Gra `cia,F.,Forno s,J.J.,andClamor,B.,2007a,Cavitatscostaneresdeles Balearsgeneradesalazonademescla,ambimportantscontinuacions subaqua `tiques, in Pons,G.X.,andVicens,D.,eds.,Geomorfolog a LitoraliQuaternari.HomenatgeaJoanCuerdaBarcelo .Mon.Soc. Hist.Nat.Balears,v.14,p.299–352. Gra `cia,F.,Forno s,J.J.,Clamor,B.,Febrer,M.,andGamund ,P.,2007b, LacovadesaGledaI.SectorCla `ssic,SectordePonentiSectorCinccents.(Manacor,Mallorca):Geomorfologia,espeleoge `nesi,sedimentologiaihidrologia:Endins,v.31,p.43–96. Gra `cia,F.,Jaume,D.,Ramis,D.,Forno s,J.J.,Bover,P.,Clamor,B., Gual,M.A.,andVadell,M.,2003,LescovesdeCalaAnguila (Manacor,Mallorca).II:LaCovaGenovesaoCovad’enBesso . Espeleoge `nesi,geomorfologia,hidrologia,sedimentologia,fauna, paleontologia,arqueologiaiconservacio :Endins,v.25,p.43–86. Guijarro,A.,1986,Contribucio nalabioclimatolog adelasBaleares [Ph.D.thesis]:PalmadeMallorca,UniversitatdelesIllesBalears, 232p. Lambeck,K.,andChappell,J.,2001,SealevelchangethroughtheLast GlacialCycle:Science,v.292,no.5517,p.679–686. Lambeck,K.,andPurcell,A.,2005,Sea-levelchangeintheMediterranean SeasincetheLGM-.Modelpredictionsfortectonicallystableareas: QuaternaryScienceReviews,v.24,p.1969–1988. Mylroie,J.E.,andCarew,J.L.,1990,Theflankmarginmodelfor dissolutioncavedevelopmentincarbonateplatforms:EarthSurface ProcessesandLandforms,v.15,p.413–424. Mylroie,J.R.,andMylroie,J.E.,2007,Developmentofthecarbonate islandkarstmodel:JournalofCaveandKarstStudies,v.69,no.1, p.59–75. Palmer,A.N.,2007,CaveGeology:Dayton,CaveBooks,454p. Pomar,L.,1991,Reefgeometries,erosionsurfacesandhigh-frequency sea-levelchanges,UpperMioceneReefComplex,Mallorca,Spain: Sedimentology,v.38,p.243–269. Sasowsky,I.D.,andMylroie,J.,eds.,2004,Studiesofcavesediments. PhysicalandChemicalRecordsofPaleoclimate:NewYork,Kluwer Academic/PlenumPublishers,329p. Tuccimei,P.,Gine s,J.,Delitala,M.C.,Gine s,A.,Gra `cia,F.,Forno s,J.J., andTaddeucci,A.,2006,Lastinterglacialsealevelchangesin Mallorcaisland(westernMediterranean).HighprecisionU-series datafromphreaticovergrowthsonspeleothems:Zeitschriftfu ¨r GeomorphologieN.F.,v.50,no.1,p.1–21. White,W.B.,2007,Cavesedimentsandpaleoclimate:JournalofCaveand KarstStudies,v.69,no.1,p.76–93. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 99

PAGE 15

A PPENDIX I Table1.Samplecolor,indryandwetconditions,andLOIofsedimentsprese ntatPirata-Pont-PiquetaSystem. CoreSampleNumberDepth(cm) Color LOI(%) moistdry PP-00 010–2.55YR4/6–– 022.5–47.5YR6/8–– PP-01 010–32.5YR5/87.5YR6/66.27 023–97.5YR5/87.5YR7/65.08 039–157.5YR5/87.5YR7/66.60 0415–202.5YR3/67.5YR5/83.11 0520–262.5YR3/67.5YR5/82.37 PP-02 010–37.5YR5/87.5YR7/67.63 023–3.55YR4/67.5YR6/62.86 033.5–77.5YR5/87.5YR7/63.89 047–13.57.5YR6/87.5YR8/43.34 0513.5–217.5YR6/610YR7/62.52 0621–237.5YR7/67.5YR8/42.77 0723–257.5YR6/6– PP-03 010–32.5YR3/37.5YR5/64.34 PP-04 010–45YR4/67.5YR6/61.76 024–95YR4/67.5YR6/61.90 PP05 010–72.5YR4/67.5YR5/64.07 027–132.5YR4/67.5YR5/63.39 0313–162.5YR4/67.5YR6/63.07 0416–212.5YR4/67.5YR6/63.41 PP-06 010–32.5YR4/65YR5/82.24 023–52.5YR4/85YR5/82.44 035–7.52.5YR4/65YR5/63.49 047.5–112.5YR4/85YR5/63.07 0511–162.5YR4/85YR5/62.07 PP-07 010–32.5YR3/67.5YR5/63.97 023–72.5YR3/67.5YR5/63.31 037–9.52.5YR3/47.5YR6/62.80 049.5–12.55YR3/37.5YR6/64.60 0512.5–16.55YR3/47.5YR6/64.69 0616.5–19.55YR3/47.5YR6/64.66 0719.5–20.55YR3/47.5YR6/65.17 0820.5–22.55YR3/37.5YR5/44.43 0922.5–265YR2.5/27.5YR5/43.91 1026–345YR2.5/17.5YR6/44.09 PP-08 010–3.52.5YR4/45YR5/65.26 023.5–62.5YR4/67.5YR5/65.15 036–12.52.5YR4/65YR5/84.65 P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 100 N JournalofCaveandKarstStudies, April2009

PAGE 16

CoreSampleNumberDepth(cm) Color LOI(%) moistdry 0412.5–17.55YR3/47.5YR5/64.55 0517.5–25.55YR3/47.5YR5/64.74 0625.5–27.55YR3/47.5YR5/64.20 0727.5–325YR3/47.5YR5/65.22 0832–375YR3/37.5YR5/64.77 PP-08bis 0937–457.5YR3/27.5YR5/67.06 1045–5310YR3/37.5YR5/47.59 1153–6210YR3/37.5YR6/46.81 PP-09 010–102.5YR4/85YR5/64.45 PP-11 01surface7.5YR4/65YR4/612.50 PP-12 010–107.5YR7/87.5YR7/61.09 0210–207.5YR7/87.5YR7/41.36 0320–307.5YR7/87.5YR7/61.18 0430–407.5YR7/87.5YR8/41.53 Table1.Continued. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 101

PAGE 17

Table2.Samplecolor,indryandwetconditions,andLOIofsedimentsprese ntatCovadesaGledapassages. CoreSampleNumber Depth Color LOI(%) (cm)moistdry GL-01 010–5.52.5YR4/87.5YR6/49.43 025.5–135YR4/67.5YR6/49.39 0313–175YR4/37.5YR6/69.80 0417–217.5YR3/310YR5/49.59 0521–26.57.5YR3/47.5YR6/311.03 0626.5–3010R4/87.5YR5/69.34 0730–362.5R4/47.5YR6/49.85 0836–395YR3/310YR5/49.68 0939–43.55YR3/37.5YR5/49.86 1043.5–465YR3/27.5YR5/410.91 1146–505YR3/310YR4/410.99 1250–51.82.5YR4/6–– 1351.8–52.52.5YR5/87.5YR5/610.03 1452.5–56.52.5YR4/67.5YR5/49.55 1556.5–607.5YR3/47.5YR5/69.35 GL-02 010–42.5YR4/67.5YR5/69.66 024–107.5YR4/410YR6/69.67 0310–14.57.5YR4/410YR6/68.78 0414.5–177.5YR4/310YR5/49.09 0517–20.57.5YR2.5/210YR5/49.49 0620.5–232.5Y3/310YR5/311.17 0723–24.52.5Y2.5/12.5Y5/38.72 0824.5–282.5YR4/32.5Y6/47.39 0928–29.55YR4/65YR6/84.54 10F29.5–3210YR8/610YR8/411.36 11A32–34?7.5YR5/610YR7/411.54 11F–––– GL-03 010–310YR3/210YR4/411.72 023.5–55Y4/110YR5/115.38 035–5.55Y7/110YR6/1– 045.5–65Y6/110YR6/118.57 056–75Y5/12.5YR5/120.62 067–10.55Y4/12.5YR6/114.81 0710.5–145Y3/12.5YR5/117.59 0814–175Y2.5/12.5YR5/121.50 0917–215Y4/12.5YR6/114.99 1021–245Y4/12.5YR6/115.88 1124–275Y4/12.5YR6/11.29 1227–315Y3/12.5YR5/113.93 GL-04 010–27.5YR3/47.5YR5/614.47 022–67.5YR4/67.5YR6/615.52 036–7.57.5YR5/67.5YR6/6– 047.5–9.57.5YR6/67.5YR6/64.19 059.5–1310YR7/610YR7/63.40 0613–17.510YR7/610YR7/62.15 0717.5–2010YR6/610YR5/69.28 0820–22.510YR5/610YR6/65.15 P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 102 N JournalofCaveandKarstStudies, April2009

PAGE 18

CoreSampleNumber Depth Color LOI(%) (cm)moistdry GL-05 010–37.5YR4/65YR6/6– 023–4.55YR4/65YR6/615.32 034.5–6.57.5YR5/67.5YR5/66.92 046.5–9.52.5YR4/65YR5/87.90 059.5–122.5YR4/65YR5/86.12 GL-06 010–3.57.5YR6/610YR6/419.15 023.5–6.55Y2.5/110YR4/120.63 036.5–8.510YR5/410YR7/419.53 048.5–105Y3/210YR5/221.43 GL-07 01surface7.5YR4/47.5YR4/416.20 GL-08 010–3.55YR5/67.5YR5/610.18 023.5–95YR5/65YR6/66.65 039–10.55YR4/67.5YR5/67.75 03a10.5–115YR5/47.5YR5/45.13 0411–172.5YR4/67.5YR5/68.27 0517–202.5YR4/67.5YR6/67.57 0620–235YR4/67.5YR6/67.38 0723–275YR4/67.5YR6/67.63 0827–292.5YR4/67.5YR6/67.81 0929–312.5YR4/67.5YR5/67.47 1031–345YR4/67.5YR5/48.49 1134–365YR4/67.5YR6/68.06 1236–395YR4/67.5YR6/67.17 1339–42.52.5YR4/67.5YR6/67.07 GL-09 010–57.5YR7/47.5YR8/220.77 025–92.5YR6/65YR8/322.63 039–1210R6/65YR7/423.81 0412–16.510R6/65YR6/423.40 0516.5–2010R6/45YR8/323.29 GL-10 010–55YR8/35YR7/312.64 01a–––– 025–105YR8/35YR8/33.07 0310–15.55YR6/35YR7/39.72 03a–––– 0415.5–215YR5/45YR7/312.13 0521–275YR4/45YR6/312.17 0627–312.5YR4/65YR6/312.74 0731–352.5YR4/42.5YR6/47.58 07aN9N9– 0835–365Y5/310R6/410.75 0936–402.5YR4/35YR7/35.17 GL-11 010–62.5YR6/85YR7/44.02 026–122.5YR6/85YR7/44.84 Table2.Continued. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 103

PAGE 19

Table.3.Percentagevaluesofthedifferenttexturalparametersofsedim entsfromPirata-Pont-Piquetasystem. Core Sample NumberGravel(%) Sand Silt(%)Clay(%) VCS(%)CS(%)MS(%)FS(%)VFS(%) PP-00 0185.005.00? 0260.0025.00? PP-01 010.000.405.5017.1018.3018.9032.687.12 020.000.687.2017.8013.0014.8035.1011.40 030.000.946.1014.4010.205.0039.6023.30 040.000.000.000.060.401.5081.4016.60 050.000.000.000.242.607.9078.3011.00 PP-02 010.000.000.000.002.1113.9069.3014.70 020.000.203.0010.4011.7013.5050.8010.40 030.003.9818.4016.406.609.3035.2010.10 040.000.000.118.1019.5021.7043.527.08 050.002.9011.8022.7013.7012.1032.874.34 060.000.000.3514.1027.0018.9035.773.93 PP-03 010.000.000.000.000.000.0080.2119.80 PP-04 010.000.000.000.192.706.0083.857.25 020.000.000.000.122.607.4081.178.73 PP-05 010.000.000.381.603.202.9074.3017.60 020.000.000.000.465.403.3072.8018.00 030.000.000.000.064.702.5073.4019.30 040.000.000.000.010.502.0078.2019.30 PP-06 010.000.000.000.011.283.2066.5029.00 020.000.000.000.000.000.0452.8047.20 030.000.000.000.000.000.0050.5149.50 040.000.000.000.000.000.0050.4149.60 050.000.000.000.000.000.0039.8760.10 PP-07 010.000.000.000.000.000.0484.1615.80 020.000.000.000.000.000.6787.2012.10 030.000.000.270.701.104.3081.9011.70 040.000.000.000.000.000.1485.0014.90 050.000.000.070.751.402.8079.4015.60 060.000.000.000.932.303.3076.9016.60 070.000.000.000.282.003.0080.3014.40 080.000.000.000.010.611.6081.6016.20 090.000.000.080.162.303.8078.4015.20 100.000.000.000.061.802.6077.6017.90 PP-08 010.000.000.160.903.207.2075.6012.90 020.000.000.000.052.755.2073.6018.40 030.000.000.000.000.000.0080.7519.20 040.000.000.000.010.612.3079.3017.80 050.000.000.000.421.803.7076.0018.10 060.000.000.000.041.632.7077.1018.50 070.000.000.000.000.000.0080.3019.70 P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 104 N JournalofCaveandKarstStudies, April2009

PAGE 20

Core Sample NumberGravel(%) Sand Silt(%)Clay(%) VCS(%)CS(%)MS(%)FS(%)VFS(%) 080.000.000.000.432.802.5077.1017.20 PP-08bis 090.000.000.000.112.053.5078.0016.30 100.000.000.012.237.605.5070.9014.00 110.000.000.000.000.000.0081.2718.70 PP-09 010.000.000.000.000.000.0061.1838.80 PP-11 010.000.000.000.448.9017.8067.445.46 PP-12 010.000.0011.8038.5016.9011.6018.572.63 020.000.003.6015.6017.1020.5036.736.47 030.000.776.8012.907.8014.3050.407.31 040.000.555.088.904.8019.6056.714.39 Table3.Continued. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 105

PAGE 21

Table.4.Percentagevaluesofthedifferenttexturalparametersofsedim entsfromCovadesaGleda. Core Sample NumberGravel(%) Sand Silt(%)Clay(%) VCS(%)CS(%)MS(%)FS(%)VFS(%) GL-01 010.000.000.000.000.000.0153.2946.70 020.000.000.000.001.205.8053.5039.50 030.000.000.000.000.000.0050.3049.70 040.000.000.000.001.001.9046.8050.30 050.000.000.000.001.002.2055.2041.60 060.000.000.000.001.106.3053.8038.80 070.000.000.000.000.606.6056.8036.00 080.000.000.000.000.201.8049.5048.50 090.000.000.000.000.001.5054.6043.90 100.000.000.000.002.006.8064.6026.60 110.000.000.000.000.002.0052.8045.20 12–––––––– 130.000.000.000.803.907.5053.6034.20 140.000.000.000.000.302.1055.1042.50 150.000.000.000.000.101.8047.5050.60 GL-02 010.000.000.000.000.305.9051.8042.00 020.000.000.000.000.203.1057.0039.70 030.000.000.000.000.003.3051.3045.40 040.000.000.000.001.102.0050.6046.30 050.000.000.000.000.703.4055.4040.50 060.000.000.000.002.2011.8059.7026.30 070.000.000.000.002.009.2064.3024.50 080.000.000.000.000.305.6056.2037.90 090.000.000.000.000.101.6052.1046.20 10F0.000.000.000.000.201.0067.2031.60 11A–––––––– 11F0.000.000.000.000.001.0069.9029.10 GL-03 010.000.000.000.002.5017.3070.909.30 020.000.000.000.002.7019.0069.209.10 030.000.000.000.006.0025.4060.508.10 040.001.704.509.9014.8026.0039.373.73 050.000.000.000.006.4028.2057.607.80 060.000.000.000.108.0030.3053.198.41 070.000.001.109.1020.0022.7040.256.85 080.003.6014.8022.9015.2014.0025.593.91 090.000.000.000.1010.6031.1050.297.91 100.000.000.103.6015.3026.6046.837.57 110.000.000.407.6019.8023.4042.526.28 120.000.000.408.8019.8023.5041.615.89 GL-04 010.000.000.304.2014.5027.1046.727.18 020.000.000.000.005.9024.5061.138.47 030.000.000.000.107.4024.0061.277.23 040.003.609.7016.7012.9016.4036.224.48 050.001.109.6016.6017.5018.5034.682.02 060.000.003.2014.8024.8023.1031.872.23 070.000.201.306.3011.0025.6050.754.85 080.001.605.007.0011.9025.4044.444.66 P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 106 N JournalofCaveandKarstStudies, April2009

PAGE 22

Core Sample NumberGravel(%) Sand Silt(%)Clay(%) VCS(%)CS(%)MS(%)FS(%)VFS(%) GL-05 01–––––––– 020.000.000.000.004.9017.5067.0010.60 0319.1321.4421.4716.889.941.483.480.32 040.000.000.000.004.5019.4067.958.15 050.000.000.000.002.6015.6073.158.65 GL-06 010.000.904.8010.1015.0022.0041.315.89 020.3011.0023.0018.1010.4010.9023.362.94 030.000.001.3010.1018.8024.1039.706.00 040.003.5013.0018.1014.8014.9031.414.29 GL-07 010.000.000.001.805.107.4049.3036.40 GL-08 010.000.000.000.002.6019.8069.248.36 020.000.000.000.005.5026.8060.307.40 030.000.000.000.002.1019.2069.459.25 03a0.000.000.000.001.9022.9067.807.40 040.000.000.000.005.1025.9061.367.64 050.000.000.000.002.9023.3065.817.99 060.000.000.000.004.1023.9063.638.37 070.000.000.000.004.4025.1062.108.40 080.000.000.000.001.6017.9070.609.90 090.000.000.000.001.6017.5070.5010.40 100.000.000.103.109.6022.1056.188.92 110.000.000.000.005.3021.0064.758.95 120.000.000.000.001.3016.8071.979.93 130.000.000.000.002.9021.5066.419.19 GL-09 010.000.505.109.608.2014.1051.5011.00 020.001.908.9011.107.6014.2047.328.98 030.000.000.605.206.7021.3056.629.58 040.000.000.503.707.9023.3055.738.87 050.000.001.108.607.2018.1055.839.17 GL-10 01–––––––– 01a–––––––– 02–––––––– 03–––––––– 03a–––––––– 04–––––––– 050.000.000.205.1011.8026.4051.065.44 06–––––––– 07–––––––– 07a–––––––– 080.000.000.000.006.6028.5058.566.34 09–––––––– GL-11 010.0014.1637.3018.607.803.4010.488.26 020.006.2028.3026.4111.805.5012.609.19 Table4.Continued. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 107

PAGE 23

Table.5.TexturalstatisticalparametersofsedimentsfromPirata-Pont -Piquetasystem. CoreSampleNumberMean( m m)Median( m m)m/MMode( m m)S.D.( m m)Skewness PP-00 01 02 PP-01 01161.8094.771.71170.80188.201.86 02168.0074.322.26361.80216.001.88 03134.5010.7212.55291.90223.302.28 0412.447.781.6110.5218.675.25 0527.9616.141.7330.7336.053.04 PP-02 0128.6413.862.0717.9835.011.71 02102.9029.643.47137.80159.602.55 03272.9091.222.99618.40343.801.55 0492.3560.431.53170.8094.051.29 05248.70139.401.78361.80295.001.97 06124.6095.251.31211.60111.801.26 PP-03 017.695.511.408.496.841.37 PP-04 0129.6620.901.4227.6134.583.28 0227.9217.121.6327.6134.292.95 PP-05 0124.306.283.876.8562.774.95 0221.665.993.626.8546.133.29 0317.715.543.206.8539.293.60 0410.235.491.867.6317.705.27 PP-06 0111.404.292.655.5324.464.53 024.102.181.881.115.794.33 032.902.021.431.532.411.52 042.762.021.371.372.141.18 052.161.601.351.231.621.34 PP-07 0111.167.731.4410.5210.731.59 0215.5310.521.4811.7114.581.28 0323.7111.102.1410.5247.696.76 0412.137.841.559.4512.141.58 0518.857.602.389.4540.206.55 0620.077.582.659.4541.584.66 0717.397.932.199.4532.634.78 0811.486.401.798.4919.205.98 0917.897.152.508.4937.985.92 1013.605.882.317.6326.534.71 PP-08 0128.0610.192.759.4551.584.61 0218.976.982.728.4933.753.39 039.096.021.518.498.961.64 0412.766.392.008.4920.335.00 0517.266.452.687.6334.494.80 0614.146.012.357.6327.154.66 078.135.491.487.638.091.91 0817.856.862.608.4935.774.51 P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 108 N JournalofCaveandKarstStudies, April2009

PAGE 24

CoreSampleNumberMean( m m)Median( m m)m/MMode( m m)S.D.( m m)Skewness PP-08bis 0917.356.972.498.4931.024.11 1034.557.944.357.6364.492.73 117.085.131.386.166.111.42 PP-09 014.002.831.414.973.421.35 PP-11 0148.2129.121.6638.0852.591.68 PP-12 01267.60250.601.07361.80206.300.64 02138.7077.981.7872.46158.201.75 03142.9044.823.1958.48215.002.25 04116.0044.232.6265.10189.802.88 Table5.Continued. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 109

PAGE 25

Table.6.TexturalstatisticalparametersofsedimentsfromCovadesaGle da. CoreSampleNumberMean( m m)Median( m m)m/MMode( m m)S.D.( m m)Skewness GL-01 017.564.361.744.448.982.46 0215.145.652.684.4422.572.45 036.354.031.584.446.812.20 048.413.872.184.0514.294.12 0514.605.422.694.8826.763.71 0617.235.762.994.0526.372.49 0717.306.642.614.8825.082.38 088.944.162.154.0515.434.38 099.374.701.994.4413.493.35 1022.2110.482.1212.4030.102.51 119.994.562.194.4414.392.68 12–––––– 1326.006.903.774.4446.653.16 1410.904.932.214.4417.413.74 158.593.942.184.0515.664.63 GL-02 0114.965.132.924.0523.432.55 0212.295.512.235.3618.743.24 0311.384.562.504.0517.512.89 0410.904.402.484.0522.224.76 0513.355.312.524.4421.583.32 0626.9211.132.429.3734.261.83 0724.7811.202.2110.2932.402.18 0816.775.972.814.4426.272.70 099.884.492.204.0514.963.48 10F10.567.021.5010.2913.394.26 11A–––––– 11F11.487.321.579.3712.732.42 GL-03 0139.1631.711.2441.6833.011.21 0240.0331.931.2550.2233.931.13 0349.0039.671.2466.4440.840.97 04149.1075.471.9880.07216.303.26 0551.4142.981.2072.9441.320.80 0654.5946.261.1887.9044.120.75 07104.6070.341.49116.30113.001.79 08274.90166.201.65429.20294.901.63 0958.7649.651.1896.4946.900.66 1075.7754.881.38105.9078.892.19 1196.3565.651.47127.6099.331.68 12100.8068.111.48127.60103.301.71 GL-04 0178.9556.491.4096.4984.132.34 0248.5539.281.2460.5240.361.00 0350.8239.661.2855.1342.911.09 04208.3093.592.23429.20268.502.16 05195.80104.900.87127.60227.201.95 06145.90105.101.39140.10139.401.51 0790.4754.911.6566.44121.103.75 08135.8064.572.1087.90214.803.25 GL-05 01–––––– P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 110 N JournalofCaveandKarstStudies, April2009

PAGE 26

CoreSampleNumberMean( m m)Median( m m)m/MMode( m m)S.D.( m m)Skewness 0241.383.781.3541.6838.441.31 0379.6846.851.7037.9795.482.45 0443.3033.471.2941.6837.451.26 0537.9429.581.2834.5832.741.37 GL-06 01136.7069.291.97105.90190.602.84 02438.50297.301.48567.70463.001.30 03108.7072.351.50116.30116.201.80 04246.80121.702.03471.10294.901.81 GL-07 0131.828.403.7931.5055.562.93 GL-08 0141.3033.591.2345.7533.551.07 0250.2041.911.2060.5239.280.87 0339.8032.341.2345.7532.801.07 03a43.6738.741.1350.2232.150.81 0449.0440.781.2060.5238.730.89 0544.3537.051.2055.1334.840.91 0646.1538.001.2155.1337.100.95 0747.1738.971.2160.5237.690.89 0838.1931.351.2245.7531.521.06 0937.5130.281.2441.6831.501.09 1062.8240.281.5655.1373.952.73 1145.4235.731.2745.7539.161.19 1237.2831.281.1945.7530.241.05 1342.6435.221.2150.2234.680.99 GL-09 01115.1037.223.0934.58186.602.59 02163.3049.273.3150.22250.702.17 0368.6340.131.7155.1393.582.88 0465.2642.551.5360.5280.392.96 0581.1039.792.0445.75113.002.34 GL-10 01–––––– 01a–––––– 02–––––– 03–––––– 03a–––––– 04–––––– 0577.2052.681.4787.9086.212.40 06–––––– 07–––––– 07a–––––– 0852.2844.071.1972.9441.430.84 09–––––– GL-11 01557.61528.941.05684.16435.230.61 02410.23360.781.14517.18348.290.73 Table6.Continued. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 111

PAGE 27

Table.7.Semi-quantitativepercentagesofthemineralogicalcompositi onofsamplesfromPirata-Pont-Piquetasystem. CoreSampleNumbermontmorillonitaclh/montillitekaolinitequartzfel dsparsdolomitecalcite PP-00 010.001.774.292.4640.816.64i44.04 02 PP-01 013.980.003.040.9114.132.173.9871.77 020.000.001.880.0013.660.004.7175.29 030.003.193.653.6828.993.5011.5745.41 042.450.008.555.5970.626.020.006.78 050.000.005.583.6768.4313.313.875.44 PP-02 010.001.743.371.0910.500.002.4783.84 020.000.002.102.1619.895.781.3968.78 030.000.006.732.2513.601.262.3673.82 040.000.006.292.3413.331.602.6873.76 050.000.001.530.005.101.100.5991.69 060.000.001.430.003.143.490.7991.14 PP-03 010.000.0011.984.0766.8812.710.004.36 PP-04 010.003.305.992.7164.4411.605.876.10 020.000.004.983.1673.5113.240.005.10 PP-05 010.000.0010.763.8667.2510.721.765.66 020.000.0014.396.0263.1211.270.005.19 030.000.0017.915.3659.4410.711.964.61 044.240.0013.765.5156.8212.760.006.91 PP-06 010.005.2814.567.7444.1610.103.4314.74 020.000.0026.9010.0137.3513.455.536.77 030.004.8222.439.7939.9914.400.008.59 040.009.2722.6710.3040.6612.340.004.75 050.003.0818.2511.0346.3813.290.007.96 PP-07 010.000.007.975.3369.7011.010.005.98 020.000.004.341.9575.3015.910.002.50 030.000.007.172.4178.738.510.003.18 040.005.158.832.8365.8912.930.004.36 050.002.368.443.2270.6010.341.313.63 060.003.828.614.1467.8610.240.005.33 070.005.219.634.2666.279.710.004.92 P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 112 N JournalofCaveandKarstStudies, April2009

PAGE 28

CoreSampleNumbermontmorillonitaclh/montillitekaolinitequartzfel dsparsdolomitecalcite 081.970.007.232.3268.8214.730.005.67 090.000.0010.274.8667.8210.540.006.45 100.000.009.324.3170.4711.420.004.38 PP-08 010.000.007.092.2676.119.860.004.69 020.000.0013.935.5964.8512.040.003.59 030.000.006.392.9477.989.540.003.15 040.000.005.213.2372.1411.812.545.07 050.000.008.784.6368.8113.750.004.03 060.000.0013.644.3867.1611.280.003.53 070.000.0016.864.7266.5511.880.000.00 080.003.2310.004.9268.0610.962.820.00 PP-08bis 090.000.4710.664.9872.4911.400.000.00 100.000.0012.005.1273.179.720.000.00 110.000.0015.264.5565.3711.150.003.68 PP-09 010.001.3721.249.0155.198.140.005.06 PP-10 rock0.000.000.000.001.400.000.0098.60 rockimpurities0.000.000.650.0099.340.000.000.00 PP-11 010.000.006.761.8276.6910.261.223.24 PP-12 01i0.000.910.0012.270.001.5585.27 020.000.000.001.688.090.003.7886.45 030.000.001.080.002.850.002.6793.41 040.000.000.000.002.280.000.0097.72 Table7.Continued. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 113

PAGE 29

Table.8.Semi-quantitativepercentagesofthemineralogicalcompositi onofsamplesfromCovadesaGleda. Core Sample Numbersmectiteillitegypsumkaoliniteanhydritearagonitequartzfeld sparscalcitedolomite GL-01 010.0016.130.006.420.000.0057.843.969.995.66 020.0014.760.005.7111.980.0049.007.4311.120.00 032.358.720.002.890.000.0063.657.4914.900.00 040.0011.470.006.259.480.0053.4511.048.300.00 050.009.430.005.796.720.0053.287.7710.886.13 060.0012.030.007.810.000.0058.378.867.025.92 070.009.140.006.080.000.0065.348.856.324.27 080.0012.480.009.290.010.0060.108.299.820.00 090.0013.910.007.370.000.0060.766.397.094.49 106.478.590.005.410.000.0056.745.658.228.91 110.0017.210.008.936.180.0045.537.3912.332.43 120.0012.030.005.980.000.0054.505.9814.127.40 140.0011.270.006.250.000.0059.833.8812.586.19 1512.4512.600.006.584.050.0046.134.777.985.45 GL-02 010.0017.460.007.030.000.0069.670.000.005.85 020.0012.730.005.270.000.0060.278.146.507.08 030.0015.180.006.070.000.0064.407.866.490.00 040.0012.690.009.050.000.0060.889.917.470.00 050.0014.142.935.770.000.0068.410.008.750.00 060.006.980.004.320.000.0040.409.4632.166.68 070.007.910.003.670.000.0055.107.4412.9912.89 080.007.900.005.480.000.0056.724.8115.759.34 090.009.410.005.750.000.0060.757.189.707.21 10F0.005.560.001.900.000.009.944.207.9470.46 11A0.000.000.000.020.000.000.000.0092.477.51 11F0.000.000.000.000.000.0015.580.0030.2454.18 GL-03 010.009.504.785.630.000.0056.865.2817.950.00 020.003.550.000.010.010.0034.084.0147.8410.50 030.000.000.000.000.000.0016.902.1144.8736.12 040.000.010.000.000.000.0034.832.5241.9920.65 050.007.820.002.650.002.9022.323.9744.1216.22 060.007.790.002.460.000.0024.173.4858.213.89 070.007.030.001.650.000.0026.883.9657.493.00 080.006.260.003.153.640.0025.769.3549.901.94 090.000.010.000.010.000.0024.453.9168.533.09 100.000.000.000.000.000.0024.852.4769.123.55 110.003.910.001.750.000.0034.472.8453.433.61 120.005.100.000.000.000.0031.260.0060.892.74 GL-04 010.009.870.006.874.650.0049.836.064.5518.17 020.007.730.004.460.010.0030.584.303.5649.36 030.008.910.000.024.420.0031.420.0017.3037.94 040.001.810.001.450.000.007.400.0082.356.98 050.001.580.000.000.000.003.250.0095.170.00 060.000.000.000.000.000.004.320.0095.680.00 070.0017.190.008.010.000.0035.595.5433.670.00 080.004.860.002.600.000.008.320.0084.220.00 GL-05 010.001.100.000.010.000.005.330.0091.312.26 P RESENT-DAYSEDIMENTARYFACIESINTHECOASTALKARSTCAVESOFMALLORCAISLA ND(WESTERNMEDITERRANEAN) 114 N JournalofCaveandKarstStudies, April2009

PAGE 30

Core Sample Numbersmectiteillitegypsumkaoliniteanhydritearagonitequartzfeld sparscalcitedolomite 020.005.330.003.610.000.0041.496.969.0733.54 030.001.320.000.000.000.007.880.0089.481.32 040.006.970.003.193.230.0072.515.662.006.45 050.009.460.005.114.310.0068.625.457.040.00 GL-06 010.005.160.003.070.000.0061.902.243.0624.57 020.007.690.002.340.000.0072.855.626.105.40 030.006.390.002.510.000.0034.685.3117.4133.71 040.003.740.003.1914.680.0051.655.6711.209.87 GL-07 011.113.510.002.240.000.0074.247.3111.590.00 GL-08 010.006.900.004.190.000.0064.645.9718.290.00 020.0016.040.0025.070.000.006.7618.3227.106.70 030.007.330.003.720.000.0058.158.8819.002.92 03a0.005.400.002.370.000.0063.082.6326.520.00 040.006.980.004.160.000.0067.057.0114.800.00 050.009.200.004.780.000.0063.778.0514.200.00 060.0011.440.004.460.000.0055.309.1119.690.00 070.009.620.004.230.000.0056.697.4422.020.00 080.007.100.003.550.000.0049.264.6629.585.86 090.009.790.005.880.000.0063.807.7812.740.01 100.008.640.004.470.000.0061.276.2519.370.01 110.007.920.004.750.000.0057.664.4925.170.01 120.017.170.004.200.000.0061.754.4422.440.00 130.008.180.003.360.000.0063.524.7318.271.95 GL-09 010.000.000.000.010.000.000.000.0047.9052.10 020.000.010.000.000.000.005.980.000.0094.01 030.005.910.002.540.000.0016.072.810.0072.66 040.005.410.003.070.000.0015.030.006.0870.41 050.000.010.000.000.000.0012.840.001.6985.45 GL-10 010.004.260.001.870.000.003.370.0028.3662.14 01a0.000.010.000.000.000.000.000.004.5795.42 020.000.010.000.000.000.005.030.0059.4235.55 030.000.010.000.000.000.004.090.0060.3835.52 03a0.000.000.000.010.000.002.370.0037.4960.13 040.000.010.000.010.000.004.610.0041.1954.17 050.005.770.002.670.000.008.142.2817.7763.37 060.008.870.003.280.000.0010.740.0028.6248.49 070.003.580.002.220.000.009.240.0013.9870.99 07a0.000.000.000.000.000.000.000.000.00100.00 080.000.010.003.840.000.0010.370.0062.6723.10 090.003.150.000.000.000.004.310.0049.7242.82 GL-11 010.000.000.000.000.000.000.470.0085.6113.92 020.000.000.000.000.000.001.090.0085.5213.39 Table8.Continued. J.J.F ORNO S ,J.G INE S AND F.G RA ` CIA JournalofCaveandKarstStudies, April2009 N 115



PAGE 1

April 2009 Volume 71, Number 1 ISSN 1090-6924 A Publication of the National Speleological Society JOUR NA L OF C AV E AN D K A R S T STUD I E S IN THIS ISSUE: Journal of Cave and Karst Studies Volume 71 Number 1 April 2009Journal of Cave and Karst StudiesVolume 71 Number 1 April 2009 Editorial 1 Journal of Cave and Karst Studies Use of FSC-Certied and Recycled Paper Malcolm S. FieldArticle 2 Oreonetides Beattyi, A New Troglobitic Spider (Araneae: Linyphiidae) from Eastern North America, and Re-description of Oreonetides Flavus Pierre Paquin, Nadine Duprr, Donald J. Buckle, and Julian J. LewisArticle 16 Monk Seal (Monachus Monachus) Bones in Bel Torrente Cave (Central-East Sardinia) and Their Paleogeographical Signicance Jo De Waele, George A. Brook, and Anke OertelArticle 24 Seasonal Distribution and Circadian Activity in the Troglophile Long-Footed Robber Frog, Eleutherodactylus Longipes (Anura: Brachycephalidae) at Los Riscos Cave, Quertaro, Mexico: Field and Laboratory Studies Adriana Espino del Castillo, Gabriela Castao-Meneses, Mayra J. Dvila-Montes, Manuel Miranda-Anaya, Juan B. Morales-Malacara, and Ricardo Paredes-LenArticle 32 Caves as Sea Level and Uplift Indicators, Kangaroo Island, South Australia John E. Mylroie and Joan R. MylroieArticle 48 Formation of Seasonal Ice Bodies and Associated Cryogenic Carbonates in Caverne de LOurs, Qubec, Canada: Kinetic Isotope Effects and Pseudo-Biogenic Crystal Structures Denis Lacelle, Bernard Lauriol, and Ian D. ClarkArticle 63 Wavelet Analysis of Late Holocene Stalagmite Records from Ortigosa Caves in Northern Spain A. Muoz, A.K. Sen, C. Sancho, and D. GentyArticle 73 Limitations of Hendy Test Criteria in Judging the Paleoclimatic Suitability of Speleothems and the Need for Replication Jeffrey A. Dorale and Zaihua LiuArticle 81 Scuttle Flies (Diptera: Phoridae) from Caves in Meghalaya, India R. Henry L. DisneyArticle 86 Present-Day Sedimentary Facies in the Coastal Karst Caves of Mallorca Island (Western Mediterranean) Joan J. Forns, Joaqun Gins, and Francesc GrciaArticle 100 The Legend of Carbon Dioxide Heaviness Giovanni Badino

PAGE 2

G U I DE TO A UT H OR S The Journal of Cave and Karst Studies is a multidisciplinary journal devoted to cave and karst research. The Journal is seeking original, unpublished manuscripts concerning the scientic study of caves or other karst features. Authors do not need to be members of the National Speleological Society, but preference is given to manuscripts of importance to North American speleology. LANGUAGES: The Journal of Cave and Karst Studies uses American-style English as its standard language and spelling style, with the exception of allowing a second abstract in another language when room allows. In the case of proper names, the Jour nal tries to accommodate other spellings and punctuation styles. In cases where the Editor-in-Chief nds it appropriate to use nonEnglish words outside of proper names (generally where no equivalent English word exists), the Journal italicizes them. However, the common abbreviations i.e., e.g., et al., and etc. should appear in roman text. Authors are encouraged to write for our combined professional and amateur readerships. CONTENT: Each paper will contain a title with the authors names and addresses, an abstract, and the text of the paper, including a summary or conclusions section. Acknowledgments and references follow the text. ABSTRACTS: An abstract stating the essential points and results must accompany all articles. An abstract is a summary, not a promise of what topics are covered in the paper. STYLE: The Journal consults The Chicago Manual of Style on most general style issues. REFERENCES: In the text, references to previously published work should be followed by the relevant authors name and date (and page number, when appropriate) in parentheses. All cited references are alphabetical at the end of the manuscript with senior authors last name rst, followed by date of publication, title, publisher, volume, and page numbers. Geological Society of America for mat should be used (see http://www.geosociety.org/pubs/geoguid5. htm). Please do not abbreviate periodical titles. Web references are acceptable when deemed appropriate. The references should follow the style of: Author (or publisher), year, Webpage title: Publisher (if a specic author is available), full URL (e.g., http://www. usgs.gov/citguide.html) and date when the web site was accessed in brackets; for example [accessed July 16, 2002]. If there are specic authors given, use their name and list the responsible organization as publisher. Because of the ephemeral nature of websites, please provide the specic date. Citations within the text should read: (Author, Year). SUBMISSION: Effective July 2007, all manuscripts are to be submitted via AllenTrack, a web-based system for online submission. The web address is http://jcks.allentrack2.net. Instructions are provided at that address. At your rst visit, you will be prompted to establish a login and password, after which you will enter information about your manuscript (e.g., authors and addresses, manuscript title, abstract, etc.). You will then enter your manuscript, tables, and gure les separately or all together as part of the manuscript. Manuscript les can be uploaded as DOC, WPD, RTF, TXT, or LaTeX. A DOC template with additional manuscript specications may be downloaded. (Note: LaTeX les should not use any unusual style les; a LaTeX template and BiBTeX le for the Journal may be downloaded or obtained from the Editor-inChief.) Table les can be uploaded as DOC, WPD, RTF, TXT, or LaTeX les, and gure les can be uploaded as TIFF, EPS, AI, or CDR les. Alternatively, authors may submit manuscripts as PDF or HTML les, but if the manuscript is accepted for publication, the manuscript will need to be submitted as one of the accepted le types listed above. Manuscripts must be typed, double spaced, and single-sided. Manuscripts should be no longer than 10,000 words plus tables and gures, but exceptions are permitted on a case-bycase basis. Authors of accepted papers exceeding this limit may have to pay a current page charge for the extra pages unless decided otherwise by the Editor-in-Chief. Extensive supporting data will be placed on the Journals website with a paper copy placed in the NSS archives and library. The data that are used within a paper must be made available. Authors may be required to provide supporting data in a fundamental format, such as ASCII for text data or comma-delimited ASCII for tabular data. DISCUSSIONS: Critical discussions of papers previously published in the Journal are welcome. Authors will be given an opportunity to reply. Discussions and replies must be limited to a maximum of 1000 words and discussions will be subject to review before publication. Discussions must be within 6 months after the original article appears. MEASUREMENTS: All measurements will be in Systeme Internationale (metric) except when quoting historical references. Other units will be allowed where necessary if placed in parentheses and following the SI units. FIGURES: Figures and lettering must be neat and legible. Figure captions should be on a separate sheet of paper and not within the gure. Figures should be numbered in sequence and referred to in the text by inserting (Fig. x). Most gures will be reduced, hence the lettering should be large. Photographs must be sharp and high contrast. Color will generally only be printed at authors expense. TABLES: See http://www.caves.org/pub/journal/PDF/Tables. pdf to get guidelines for table layout. COPYRIGHT AND AUTHORS RESPONSIBILITIES: It is the authors responsibility to clear any copyright or acknowledgement matters concerning text, tables, or gures used. Authors should also ensure adequate attention to sensitive or legal issues such as land owner and land manager concerns or policies. PROCESS: All submitted manuscripts are sent out to at least two experts in the eld. Reviewed manuscripts are then returned to the author for consideration of the referees remarks and revision, where appropriate. Revised manuscripts are returned to the appropriate Associate Editor who then recommends acceptance or rejection. The Editor-in-Chief makes nal decisions regarding publication. Upon acceptance, the senior author will be sent one set of PDF proofs for review. Examine the current issue for more information about the format used. ELECTRONIC FILES: The Journal is printed at high resolution. Illustrations must be a minimum of 300 dpi for acceptance.The Journal of Cave and Karst Studies (ISSN 1090-6924, CPM Number #40065056) is a multi-disciplinary, refereed journal published three times a year by the National Speleological Society, 2813 Cave Avenue, Huntsville, Alabama 35810-4431 USA; Phone (256) 852-1300; Fax (256) 851-9241, email: nss@caves.org; World Wide Web: http://www.caves.org/pub/journal/. Check the Journal website for subscripion rates. Back issues and cumulative indices are available from the NSS ofce. POSTMASTER: send address changes to the Journal of Cave and Karst Studies, 2813 Cave Avenue, Huntsville, Alabama 35810-4431 USA. The Journal of Cave and Karst Studies is covered by the following ISI Thomson Services Science Citation Index Expanded, ISI Alerting Services, and Current Contents/Physical, Chemical, and Earth Sciences. Copyright 2009 by the National Speleological Society, Inc. Front cover: Snake Lagoon on Kangaroo Island, South Australia, see Mylroie and Mylroie in this volume. Photo by J. Mylroie.Published By The National Speleological SocietyEditor-in-Chief Malcolm S. FieldNational Center of Environmental Assessment (8623P) Ofce of Research and Development U.S. Environmental Protection Agency 1200 Pennsylvania Avenue NW Washington, DC 20460-0001 703-347-8601 Voice 703-347-8692 Fax eld.malcolm@epa.govProduction EditorScott A. EngelCH2M HILL 700 Main Street, Suite 400 Baton Rouge, LA 70802 225-381-8454 scott.engel@ch2m.comJournal ProofreaderDonald G. Davis441 S. Kearney St Denver, CO 80224 303-355-5283 dgdavis@nyx.netJOURNAL AD VISOR Y BO ARD Dave Culver Gareth Davies Harvey DuChene Annette Summers Engel John Mylroie Megan Porter Elizabeth White William White Carol Wicks BO ARD OF EDITORSAnthropology George Crothers University of Kentucky211 Lafferty Hall Lexington, KY 40506-0024 Conservation-Life Sciences Julian J. Lewis & Salisa L. LewisLewis & Associates, LLC. Cave, Karst & Groundwater Biological Consulting Earth Sciences Gregory S. SpringerDepartment of Geological Sciences 316 Clippinger Laboratories Exploration Paul BurgerCave Resources Ofce Microbiology Kathleen H. LavoieDepartment of Biology State University of New York Plattsburgh, NY 12901 Paleontology Greg McDonaldPark Museum Management Program National Park Service 1201 Oakridge Dr. Suite 150 Fort Collins, CO 80525 Social Sciences Joseph C. DouglasHistory Department Volunteer State Community College Book Reviews Arthur N. Palmer & Margaret V PalmerDepartment of Earth Sciences State University of New York Oneonta, NY 13820-4015