INFORMATION TO USERS
This manuscripthas been reproducedfrom the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertationcopies are in typewriterface, while others may be from any type of computer printer.
The qualityof this reproductionis dependentupon the qualityof the copy submitted. Brokenor indistinctprint, colored or poor quality illustrations and photographs,print bleedthrough,substandard margins. and improper alignment can adversely affect reproduction.
In the unlikelyevent that the authordid not sendUMI a completemanuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials e.g., maps, drawings, charts! are reproduced by sectioningthe original,beginning at the upperleft-hand corner and continuing from left to right in equal sectionswith smalloverlaps.
ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 800-521-0600
ECOLOGY AiID BIOLOGY OF THE ROI.'GH-TOOTHED DOLPHI>
S TELO BRED.JIVE.%'SIS!
A DISSERTATION Sl'Bil ITTE D TO TH F. GRADL ATE Dli'ISIO> OF TH F l:~IYERSITW OF HA~IAI'I l> PARTIAL FI;I.FILL~IE>T OF THF REQl'IREilE>TS FOR THE DEGRFE OF
DOCTOR OF PHILOSOPHY
BIOIIF.DICAI. SCIF.. IiCES PHYS IOI.OGY!
DECEilBF.R 2002
By christi L. ll'est
Dissertation Committee:
G. Causey %'hitto», Chairperson Claude Payri, Co-chairperson! David Lally Alemandre Gannier John Claybaugh Jay Sweeney James Carpenter James ~lead referee! Yu-Chong Lin l'incent Ridouv referee! UMI Number: 3070732
UMI Microform 3070732 Copyright2003 by ProQuestInformation and LearningCompany. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 ECOLOGIE ET BIOLOGIE DU DAUPHIN A BEC ETROIT
STENO BREDA NENES'IS
DISSERTATION CONJOINTEMENT PRESENTEE A L'UN[VFRSITE DE HAWAI'I ET L'UNIVERSITE DE I-A POLYNESIE FRANCAISE EN VUE DE L'OBTENTION DU
DOCTO RAT
F.N
SCIENCF BIOMEDICALE ET FCOI OGIE MARINE
DECEMBRE 2002
paf Kristi L. West
Membresdu Jury:
Causey G. Whittow, Co-Directeur Claude Payri, Co-Directrice David Lally, examinateur Alexandre Gannier, tuteur John Claybaugh,examinateur Jay Sweeney,examinatew James Carpenter, examinateur James Mead rapporteur! Yu-Chong Lin, examinateur Vincent Ridoux rapporteur! UNIVMQvt 0%LA University of Hawai'i at Manoa ~mksts rauocruss
University of Hawaii at Manoa Universithde la Polynhsiefranqaise
Associate Dean, Graduate Division La prbsidente The graduateChair of Departtnent of Physio ogy
Consideringthe generalagreemcnt signed in 1994by the presidentofboth Universityde la PoIynbsic franchise ex University~isc du Pacifique!and The Universityof Hawaii at Manoa Consideringthe spo:ific letter of agreementsignod m I 997between the two Universities concerning the joint Ph.D dissertation for Ms Kristi WEST
The two institutionsagree upoa the foilowiag committeel'or tbe Pb.D. viva voce czatlllnatlon:
Article I The viva voce examination for thc PhD dissertation of Ms Kristi WEST will bc held Scptcmber30, 200' st the Vnivetsity of Hawat at Maooa ln the Pbysiology Coo fcrence room room T408 ln the Biomedical ScieaccsBuildiag!.
ice dissertation title is: Ecol Biolo f the rou h-ooth d I hn rcno !~rrrlantnsi
Article 2- Thc viva voce committee includes:
David Lally, Professor,UH, reader Yu-Cbong Lia, Professor, UH. reader J nhaClaybaugh, Professor,UH, reader James Carpenter, Professor, UH, reader Causey Whittow, Professor,co-advisor Claude Payri, professor oo-advisor Alexaodre Gaaaicr, PhD, reader Jay Sweeney,Dr VeterinarianDolphin }vest, reader ViaceatR!BOUX, ProfessorUniversity of La RochclleFrance, referee JamesMEAD, MarineMammal Curator, Smithsonian Institution Washington DC, referee
Article3 The universitiesinvolved arc responsible for the application of thisde:ision.
For Universityof ldawaii For Universityof FrenchPolynesia
The Associate Dean, Gmluntc Division
Peter V. GARROO III DUAL PH.D. PROGRAM BETWEEN THE UMVERSITY OF HAWAI'I AND LVNIVERSITE DE LA POLYNESIE FRANQAISE
This dissertationis part of a dual Ph.D. program under a joint agreementbetween the
University of Hawaii at Manoa and LVniversite de la Polynksie Franqaise.The dissertationcommittee is composedof an equal number of membersto representboth
Universities. Although the format of the dissertationdocument is not traditional,each of the threechapters follows similar formatting for journal submission. An overall abstract.
introduction and discussion are also included in the dissertation document that follows more traditional formatting. 9'e certify that we haveread this thesisand that. in our opinion.it is satisfactoryin scopeand quality as a dissertationfor thedegree of Doctorof
Philosophyin BiomedicalScience Physiology!.
DISSERTATION COMMI I I I I: ~Ne certiti tnat ice haie reai thi' tl esis and tiiat. in our opinion. it sat,stactor in scopeand quality zs a dissertationtor th" degreeot Doctor
DIS>ER I A Tl Ct'iliXIITTEE Greater knowledge of the rough-toothed dolphin..S'Ieno bredanensis, is needed to effectively contribute to conservationand managementefforts for this species. 1he primarypurpose of this researchwas to describeecological and biological parameters for
X bredanensis that will be useful in future assessmentsof population stress. Several approacheswere usedto studyX bfedanensis,including investigations of free-ranging populations.dead specimens,and captive individuals. Free-rangingrough-toothed dolphinsdistributed near small oceanicisland environmentswere found to be more commonly sightedin-shore than off-shore. fn the Windwardislands of FrenchPolynesia, this speciespreferred water depths of 1000to 2000m anda distanceof 1.8to 5.5 km from the barrier reef. Group sizesof rough-tootheddolphins sighted in FrenchPolynesia rangebetween 1 and35 individualswith a meansize of 12.1. Fndocrinologydata for 5 bredanensiswas establishedin captive healthy and strandedindividuals. Rangesand means were provided for progesterone,testosterone, cortisol and thyroid hormones.
Changesin thyroid hormone concentrationswere reflective of health status and testosteroneappeared to be suppressedin ill individuals.Reproduction in X bredanensis wasinvestigated by determiningthe size and age range that this species attains sexual and physicalmaturity. Femalerough-toothed dolphins attain sexual maturity by 9 to 10years of ageand malesbetween 5 and 10 yearsat a similarlength of approximately216 cm.
Physicalmaturity is generallyreached at an olderage and larger size for bothmales and females. Ecologically healthy and unhealthypopulations of S. bredanensiswere describedin this investigationand these finding will beuseful in assessingfuture threats to this species. Une meilleure connaissancedu dauphin a bec etroit. S en' bredanenvis.semble cruciale pow
ameliorer1'impact des efTorts de conservationet de protectionde cetteespece. Le but principalde cette etude est de decrire les parametres ecologiques et biologiques du dauphin Steno
bredanensis, afin d'avoir dans le futur une meilleure evaluation du niveau de stress de cette
population.Plusieurs approches ont et' utiliseesdans cette etudedu dauphin..S.bredanensis,
incluant des observationsen mer, desechantillons prelevessur animauxdecedes et des animaux
vivant en captivite.Les populationssauvages etudiees dans un environnementinsulaire ont ete
plus &equemmentobservees pres des cotes qu'au large.Dans les Iles du Vent en Polynesie
franqaise, cette espece semble se situer preferentiellementdans des eaux presentant une
profondeurde 1000a 2000metres et dansune zone distante de I.S i 5,5 kilometresde la barriere
de corail. La taille des groupesde dauphinsa becetroit observesen Polynesic fran>;aiseva de 1 a
35 individus avec une moyennede 12,1par groupe. Desmesures de taux d' hormonespour Steno
bredanensisont ete etablies a partir de prdlevementsfaits sur dauphins en bonne santeen captivite et sur dauphinsechoues. Les valeurs et leurs moyennessont donneespour la
progesterone,la testosterone,le cortisolet les hormonesthyroidiennes. Les variationsdes taux d'honmnesthyroYdiennes sont correlees avec 1'etat de sante de I'animal; la testosteroneapparait
ge&ralement bassechez les animaux malades.La physiologic reproductricedes .Cteno
bredanensisa eteetude: en detertnittantla taille et 1'age moyena laquellecette ence atteintsa
maturitesexuelle et morphologique.Les femelles atteignent la maturitesexuelle entre 9 et 10ans,
lesmiles entre5 et 10ans, leur longueurcorporeHe ayant alors ete evaluee a approximativement
216 cm. La maturitephysique est logiquementatteinte plus tardivement, la taille desanimaux
etant alors superieure,aussi bien chez les malesque chez les femelles. La comparaisonde I'ecologic des populationsde Stenobredanensis en bonne ou mauvaisesante, decrite dans cette etude.sera tres utile pour evaluerles menacesfutures qui pesentsur cetteespece. TABLE OF CONTENTS
Abstract. .Vl Resume. List of Tables...... List of Figures Xt Introduction. .1 Chapter I: Distribution of the rough-tootheddolphin Steno bredanen»i»!around the Windward Islands French Polynesia! Abstract. 73 Introduction. 74 Methods. ~6 Results. 30 Discussion 33 References 37 Chapter2: Reproductionin the rough-tootheddolphin. Stenohredanen»i»: Attainment ot sexualand physical maturity. Abstract. 50 Introduction. 51 Methods. 5'7 Results. 58 Discussion 66 References. 71 Chapter3: flormone concentrationsduring rehabilitationand in healthy rough-tootheddolphins, Stenobreclanensis .84 Abstract. .85 Introduction. .86 Methods. ..88 Results 93 Discussion 96 References.. 104 Synthesis. Re ferences. 145 LIST OF TABLES
1.1. SurveyEffort and Sighting Conditions 40
1.2 Mean Sighting Ratesfor Different Strata.
2.1 TheAge, Body Length and Body Weightof S. bredanensisfor PhysicallyMature and Immature Males and Females. 7'7
2.2 TheAge, Body Lengthand Body Weightof X bredunensisfor SexuallyMature and Immature Males and Females. 73
2.3 UterineTissue Measurements Among ReproductiveStates.
2.4 Mammary TissueComposition and StructureMeasurements Among Reproductive States. 75
Appendix I: Male reproductivedata for X bredanensis. .80
Appendix 2: Femalereproductive data for S. bredanensis. .81. 82
3.1 CortisoL Total T~, FreeT~, Total T3, FreeT3, Progesteroneand Testosteronein S bredunensis. 108 LIST OF FIGURES
1.1. Map of Study Location in the Windward Islands.French Polynesia.. 4'7
1.2 Study RegionsSub-divided According to Area.
1.3 SurveyEffort Correctedby Season:September - November
1.4 SurveyEffort Correctedby Season:December - February.
1.5 Survey Effort Corrected by Season: March - May
1.6 Grcup sizesof.S. hredanensis...
2.1 Age VersusTotal Body Lengthin X hredanensis. 76
2.2 Weight VersusTotal Body Lengthin X bredanensis. .77
2.3 TestisLength and Weight VersusTotal Body Lengthin.S'. bredanensis..
2.4 Ovary Length and Weight VersusTotal Body Lengthin.S. bredanensis...... 79
3.1 Cortisol ConcentrationsProfiled Against Time in Rehabilitation. .109
3.2 Thyroid HormoneConcentrations Profiled Against Time in Rehabilitation for Dolphin F- l. .110
3.3 ThyroidHormone Concentrations Profiled Against Time in Rehabilitation for Dolphin F-2
3.4 Thyroid HormoneConcentrations Profiled Against Time in Rehabilitation for Dolphin F-3. 112 3.5 Thyroid HormoneConcentrations Profiled Against Time in Rehabilitation for Dolphin F-4.
3.6 Thyroid HormoneConcentrations Profiled Against Time for Dolphin P...
3.7 Thyroid HormoneConcentrations Profiled Against Time for Dolphin M...
3.8 Thyroid HormoneConcentrations Profiled Against Time for Dolphin T...
3.9 Testosteroneand ProgesteroneConcentrations Profiled Against Time in Rehabilitationfor 4 StrandedDolphins.
3.10 TestosteroneConcentration Profiled Against Time for Dolphin M...
3.11 ProgesteroneConcentration Profiled Against Time for Dolphin P andT... INTRODUCTlON
Background
Marine ecosystemsare being impactedby naturaland anthropogenicsources on a world-wide scale. Thesechanges are likely to causestresses to cetaceanpopulations and have the potential to harm individuals and lead to reductionsin population sizes. The definition of stressdeveloped in the early 1900's describesthe biological responsein three stages:I! alarm and adaptation,2! hormonalevents. and 3! resistance.exlmustion and death.which are independentot the natureof the damagingagent Neyan. 1998in St.
Aubin and Dierauf. 2001!. The adaptationresponse apparent during this first stage is
likely to retlcct populationlevel changes. For example,shiAs in habitat preferencemay signify a first sign ot impactas the sourceof knownpotential stressors all involvethe
immediatesurroundings of theseanimals. As the biological stressresponse proceeds to the secondand third stage,detectable physiological changes are likely to becomeevident on an individual level.
In a few of the relatively well-known speciesof cetacean,such as the bottlenose dolphin, beluga whale, and spinnerdolphin. immunologicaland molecularbiology markingtechniques are currentlybeing developed to studystress due to anthropogenic
factors Aldridgeet al., 2001;De Guiseet al., 1998;Romano et al., 1992;Romano et al.,
1994; Southern,2000; St. Aubin and Dierauf. 2001!. In thesespecies research is now
focusingon defining precisebiological perturbations on an individuallevel. Cetacean
work in this area has included the developmentof specific monoclonalantibodies for a
few species Aldridge et al., 2001;De Guiseet al., 1998!and molecular studies initiated with the bottlenosedolphin are now being extended to beluga whales Romano et aL,
1992; 1999!. Recentefforts to identify stressactivated proteinsby molecularanalyses of tissuessuggest promise for studying the effects of coastalpollution on individual beluga whalesand tuna fishery stresseson individual spinner dolphins Southern,2000!.
The effects of environmentalimpacts on wild populationsof the cetaceanspecies
Stenobredanensis are unknown. It is undoubtablethat changingmarine ecosystems have someimpact on dolphinspecies such as S. bredanensis.However, basic information and quantitativedata for assessingimpacts on eithera populationor individuallevel for this speciesis lacking. Defined distributional patterns for S. Aredanensiswoukl provide a startingpoint for investigatingpopulation level changes.Means to sensethe adaptation ot populationsand eventuallyphysiological responds to stresson an individuallevel wouldbe a significantcontribution to the futureconservation of thisdolphin species.
lsepactsAffecting Sonre Small DolphinSpecks
Very little is knownabout the rough-tootheddolphin when compared with other tropical and sub-tropicalsmall dolphinswith similar world-widedistributions. In the tropical Pacific ocean, speciesof the genus Stenella have been relatively well documented, and dec ides of research focusing on this genus provides a very comprehensiveoverview comparedto the Steno genus. Most likely the only other speciesof dolphin where more informationis availableis the bottlenosedolphin, T. truncatus, which is the best known of all cetaceans Reynolds et al., 2000; Wells and
Scott, 1999!. Direct catch of small dolphins is one of severalthreats to populationsthroughout the world. andhigh numbersot' take for humanfood consumptionor otherproducts can result in reducedpopulation sizes. Direct catchof rough-tootheddolphins have been reported.but far moreinformation on theextent of this impactis availablefor othersmall dolphin species. The tropical and subtropicalspinner dolphin, Slenellalongirostris, appearsto inhabitsimilar waters to X bredanensis.The distribution of X longirostrisis primarily pelagic but this speciesalso occurs near islands throughoutits range
Klinowska.1991!. Althoughthe primarythreat to this speciesis incidentalby-catch in the tuna fishery,some direct catchof spinnerdolphins does occur, and reportedareas includethe SolomonIslands, Japan and the l.esserAntilles Klinowska. 1991!. The pantropicalspotted dolphin..Slenella aaenuata is anothersmall dolphin species tound in the samewater temperature range as .'i. bredunenvis,and in both coastaland off-shore habitats Klinowska, 1991!. Similar to spinnerdolphins, pantropicalspotted dolphins are consideredto be the most threatenedby incidentalcatch but this dolphin is taken in direct fisheriesin Japan,the Caribbean, Sir l~ andthe SolomonIslands Klinowska, 1991!.
The stripeddolphin, Ãlenella coeruleoalba is an exampleof anotherspecies tound in a similar distributionalrange to X bredanensis,but strjied dolphinsare also found in temperatewaters and are morecommon in the MediterraneanSea Leatherwoodand
Reeves,1983!. A directJapanese fishery for this speciesdates back centuries, and in the
1960'sannual catches were estimatedto be as high as 20,000animals in this area. Direct fisheriesare also presentin PapuaNew Guineaand the SolomonIslands, and striped dolphinswere harpooned in the EasternAtlantic from fishingboats to providecrew with meat Leatherwoodand Reeves.1983!. T. truncatus,have a world-widedistribution and are found in all tropical and temperateenvironments, absent only &om cold waters
Leatherwoodand Reeves1983; Klinowska, 1991:Reynolds et al., 2000; Wells and
Scott, 1999!. The ~~e of S. hredanensisis covered within that of T. truncatus. and thereis concernregarding the effects of environmentalimpacts on inshorepopulations of
T. truncatus as the nearshorehabitat is especially suseptibleto disturbances Klinowska.
1991!. Direct catch of bottlenosedolphins also posesa significant threat to this sanies and has resulted in populationdeclines severe enough to close a Turkish fishery in 1983
Klinowska, 1991!. In the past,direct fisheries for bottlenosedolphin were common in small localuM areasand in severalcountries including Russia,Bulgaria, the USSR.and
Japan.Today. Japan and L'wopean countries remain the primaryareas where bottlenose dolphins are taken to providemeat for humanconsumption Klinowska, 1991!. A relatively small numberot T. truncatusare also taken for public display purposesin some partsof the world as this speciestends to do well undercaptive conditions Klinowska,
1991!. Ikowever,no significantnumber of wild caughtindividuals have Ixwn brought into captivitysince 1989 in theUnited States. and animals are not takenfor publicdisplay from all regions Andrews,1999!. Thereare a few selectedareas where significant numbersof bottlenosedolphins are still takenfor publicdisplay, for examplehundrcxh have beenremoved from Cubanand Mexican waters Krames,personal comm.!.
Direct fisheriesfor thespecies S. bredanensishave been identified lrom the West
Indiesand WestAfrica, and specimenshave beeninvestigated from drive fisheriesin
Japanand in the SolomonIslands Klinowska. 1991; Miyazaki, 1980!. Numberstaken to supply food for human consumptionare small from all areas, but the limited distributional information available for this speciesindicates that populations to draw &om are not especiallylarge in any regionidentified to date Klinowska,1991!. lt is unknown what atTectthese direct fisherieshave on the current populationstatus of rough- tootheddolphins and in order to assessthese potential effects. increasedspecies knowledgeis required Young et al., 1993!. Baselinedistributional data for S. hredanensiswould provide a meansto determineaberrant distribution and eventual populationdecline resulting from potentialthreats to this speciessuch as excessive direct catch.
incidentaldeath through fishery interactions poses a significantthreat to the future of manydolphin populations.Perhaps the most well knownexample of this problem comestrom largenumbers of delphinidby-catch in the f'.asternTropical Pacific '.TP! tunafishery. Rough-tootheddolphins are sometimes associated with tunain the ETPand fishermen accounts indicate that this association occurs elsewhere as well .'arwardine et al., 1998;l~therwood and Reeves,1983!. 1lowever,the total numberof S. hredanensis individuals documentedas incidental deathsin the 1'.TPis not nearlyas extensiveas for speciesin the SIenellagenus Klinowska, 1991;f~herwood andReeves, 1983; Wade and Gerrodctte, 1993!. Tuna fisheries in the 1:TP annually reported hundreds of thousandsof spinnerand spotted dolphins as by-catchduing the 1980's Klirmwska,
1991!. As a resultof intenseexploitation pressure on spinnerdolphins, populations in this area are well studiedand informationon distributionalpatterns over a largeopen oceanregion is available.1 nthe deep FTP waters,S. longirostrisis o Aensighted in pods numberingin the hundredsor thousatids,much greatergroup si7~ than for S bredanensiswhen found in this region Perrinet al., 1987;Wade and Gerodette, 1993!.
Theexploitation of S. atIenuatain thetuna purse seine f sheryhas led to detailedstudy of thisspecies in theETP. Podsizes of thepantropical spotted dolphin range from only a few individuals to thousands,and the smaller pod sizes are similar to that of 8, bredanensis found in the FTP Wade and Gerodette. 1993!. X aaenuala is also incidentiallylost in the hook-likefishery off of Japanwhere this speciesis believedto be in competitionwith fishermenand is blamedfor interference Perrin and flohn, 1994!.
The stripeddolphin experiences some incidental mortaility in thc F.'I'Ptuna fisheryas well Leatherwoodand Reeves,1983!. T. truncatusis lost incidentially in fishery operationswhich includegill nets,purse seines, and beachnetting tor sharksin some regions I.eatherwood and Rcevcs, 1983; Klinowska. 1991!. The bottlenose dolphin also hasa reputationof beingdamaging to tishcryprotits in someI'.uropcan areas and this specieshas beenknown to be killed off of Japanduc to apparentcompetition with tishermen Lcathcrwoodand Reeves,1983; Klinowska. 199I!.
In the 1970'sby-catch information from thc I'.TP indicatedthat althoughS. hredanensiswas taken, numbers were not high when comparedto the Steno lugenus
Perrinand flenderson,1984!. At leastsome rough-toothed dolphins were killed in the purse-seineindustry Miywaki and Perrin.1994!, and assessmentot the deadspecimen collection at the SmithsonianInstitution in 2000 indicatesa significant contributionof specimensfrom incidental catch in the Pacificduring the 1980'sand 1990's personal observation!.Tuna fishery interactions likely affect rough-toothed dolphin popoulations in otherregions in additionto the I'.TP. In theearly 1990's,although the total incidental takeof S. bredanensisfor the Indo-Pacificwas not reported, the effect on theIndo-Pacilic populationof rough-tootheddolphins is believedto be insigniticant Young et al., 1993!.
No estimateof incidentalcatch is availablefor theAtlantic, and the potentialeffect on the X bredanensis population in the Western North Atlantic and Gulf of Mexico is consideredunknown Young et al., 1993!. In the Indianocean, a small numberof rough- toothed dolphins are also lost in gillnets in Sir Lanka Klinowska. 1991!. On a world- wide scale,loss of this speciesin pelagicdriAnets is unknown Kiinowska, 1991!.
Environmental change can result in distributional shiIIs of small cetacean populations. For the speciesS. bredanensis,very limited distributional information is available,and not enoughis currentlyknown to accesspotential changes in distribution due to impact effects. Oceanographicconditions during Fl Nino years indicatechanging distributional patterns of spotted dolphins as well as other species of small cet'~ans
Fielder and Reilly, 1994;Reilly and Fielder, 1994!. Since the 1982/83 El Nino event, a northwardrange extension of T. trunca ushas also beenapparent off ot the coastof
California Klinowska, 1991; Wells, 1990!. The effec ot either natural or unnatural environmental variations on X hredanensis populations have not been previously investigated.
Ilabitat degredationis recognizedas yet anotherexample of a potential threat to small cetacean populations. Vhe near-shore habitat is especially vulnerable to anthropogeniceffects such as humanencroachment, pollution and other disturbances.
For spinnerdolphin populationsusing near-shorewaters as a resting area, human disturhux:eswithin the preferredhabitat are believedto haveresulted in distributional patternshiAs as the populationadapts to thedisturbance Klinowska, 1991; Perrin, 1989!.
Changesin thedistribution of bottlenosedolphin populations off of boththe Eastand
West coastsof the USA havebeen attributed to pollution, pollution linked decline in food availability, and increased boat traffic as likely causes Klinowska, 1991!. For the speciesS. bredanensisit would be useful to establishexpected distributional patternsin near-shoreenvironments for future assessmentof thesetypes of threats to populations which are relatedto quality of habitat.
Impacton ReprodnctiveParameters
Extreme exploitation cf dolphins can result in severe population stress when numbers experiencegreat declines. A density dependentresponse may occur on a population leveL where reproductiveparaineters of individuals shiR in an attempt at compensation. this phenomenonprovides another example of the biological stress response,and in this casethe reproductivephysiology is altered on an individual basisas part of the population adaptationalphase. Age ot sexual maturity attainment is an especially important reproductive parameter as it is believed to vary in individuals accordingto the degreeof populationexploitation Pcrrin and Reilly, 1984;Smith, 1983!.
Other reproductivemeasures may also vary accordingly. Pressuresdue to low population numberswould be expectedto result in younger ageand smaller size of sexualmaturity attainment in individuals, a higher rate of pregnancy.and shorter lactations Perrin and
Reilly, 1984!.
Exploitation effects on reproductiveparameters in populationsof small delphinids have been investigated in Stenella populations from the ETP and Western Tropical
Pacific WTP! where heavy exploitation has occurred. Tuna fisheries in the eastern tropical pacific ETP! annually reported hundredsof thou~s of spinner and spotted dolphins as by-catchduring the 1980's Klinowska, 1991!. Over a twenty year period, the spinnerdolphin population in the ETP region was estimatedto have beenreduced to only 20'/oof pre-exploitationabundances Klinowska. 1991!. This extremeexploitation is evident in the reproductive parametersof ETP spinner dolphins, believed to be responsiveto the loss of huge numbers from this population Perrin and Henderson,
1984!. Attainmentof sexualmaturity in the E I P spinnerdolphins were comparedwith an unexploitedpopulation in the Gulf of Mexicowhere both males and females reach sexualmaturity at a longerand heaviersize Perrin and Henderson.1984!. Female spinnerdolphins in theeastern tropical pacific reach sexual maturity at anaverage of 5.5 growth layergroups GLG's! and at an averageof 7-10 GLG s in the Gulf of Mexico
Klinowska, 1991; Perrin and Henderson,1984!. Striped dolphins off of the coast of
Japanhave beentaken in high numbersas direct catch for centuries,and tens of thousandshave been estimated as kill in this region on an annualbasis Leatherwoodand
Reeves,1983!. A shorter calving interval, a decreasein both the length of lactationand in the resting period were apparentin the exploited Japanesestriped dolphins Kasuya.
1984!. Over 3000 individual striped dolphins were investigatedin one of thesestudies, and over 4000 spinner dolphins in another Miyazaki, 1984; Perrin and Henderson,
1984!.
When thousandsof individuals are available for reproductivestudies and the level of exploitation is known, it is possibleto comparereproductive parameters among populationsof the samedolphin species. For the speciesS. bredanensis,reproductive information is lacking and it is not evenknown for any region at what ageand size range theseanimals attain sexualmaturity. A baseline to assessreproductive pirameter shifts resulting from exploitation pressures into the future is needed for the species X bredanensis. Is~ diets oe HonaowalEvents
Adversely impacted cetaceanpopulations are likely to reflect biological and physiologicalchanges at the individual level. Hormonalevents representstage two of the biological stress response,directly following the alarm and adaptation responseof populations. Hornmnal change in relation to stressful events can be investigatedby obtaining serum samplesfrom individual dolphins to determine the circulating hormone concentrations Hadley, 1994; St. Aubin, 2001!. In order to be able to detect hormonal changes signifying that a species is experiencing significant stress, baseline endocrinologydata from healthyanimals is first requiredfor comparativeanalyses.
Cortisol is a primary glucocorticoid secretedby the adrenal gland, and is oflen considered as a stress hormone. Glucocorticoids affect the metabolism of gluclose, protein, and fat and the speedof theseprocesses increase when the body is under strum. ln the bottlenosedolphin, cortisol concentrationshave been found to rise to more than three times baselinewhen the animal undergoescapture and handling stress I hompson and Geraci. 1986!. Thyroid hormoneconcentrations are likely to be affected by various factors in cetaceans,and the stress of captureand release in bottlenosedolphins and belugawhales has resulted in decreasedthyroid hormonalconcentrations St. Aubin and
Geraci, 1988; St. Aubin and Geraci, 1992; Schroeder and Keller, 1989!. In beluga whales, studies of the stress responseto capture and handling identified an interplay betweenchanges in both cortisol and thyroid hormonesand this has been suggestedto define part of the biological mechanismsassociated with stressfor this species St. Aubin and Geraci, 1992!. Hormonalconcentrations are also effected by envirommntal surroundingsand the topic of endocrinedisruption has recently received a great deal of attention. Pollution levels are believed to be reflected in thyroid hormone values in marine mammals St.
Aubin, 2001!. ln the gray sealthyroid hormonesact as biomarkersand in the common seal a thyroid deficiency has been found to be associated with environmental contamination, specifically polychlorinated biphenyls PCB s! Brouwer et al., 1989;
Hall et aL, l998!. Evaluationof thyroid glands from dead harbor sealsin the North Sea are thought to be morphologicallydefective due to pollution in this area Schumacheret aL, l993!. Abnormal histological findings from beluga whale thyroid glands also correlate with high organochlorinecontamination in the St. l.awrenceestuary !c Guise et aL, 1994!. To be able to accuratelyassess the effects of environmentalstressors such as contaminateson hormonalevents, baseline concentrations from healthy populationson a species specific basis are first needed. No published values for any hormone are available for the rough-tootheddolphin, 3'.hredanensiv.
Limitr4iefonmtfioefor stcaobnrduaeasis
Although it becomesclear that unprecedentedchange on marine ecosystemswill ultimately affect 6ee-rangingcetacian populations, information is lacking to assessthe effects of environmental impactson some species. In the relatively unknown rough- tootheddolphin, no immurmlogicalor molecularresearch has beenundertaken, and there remainsa needto obtain baselineinformation to aid in determining the potential impact of environmentalchanges on wild populationsof this species. There is so little known about S. bredanensisthat it is indeterminatewhether this speciesis even endangeredor threatenedor experiencingsignifcant population stresses Klinowska, 1991!. For X bredanensisthere are no distributional or preferred habitat usagestudies conductedto date in near-shorewaters. 'I here is almost no knowldegeon reproductiveparameters in this species,nor areany hormonalvalues available for 8'.bredanensis. 1 hisinformation is necessaryto providea baseto aid in assessinglarge-scale population changes and individual perturbationsthat may be causedby environmentalimpacts. 1'herough- toothed dolphin is a diAicult to accessspecies, and opportunisticstudies of both controlled confmed populationsand natural wild populationsare neededto obtain this type of information.
Historyof Stenobredanensis
'I he speciesX bredanensiswas first identifiedfrom skulls in the Parismuseum andoriginally namedLe Dauphina Ixwmince which translates to thedolphin with a thin beak. In the early 1800'sthis specieswas classified as Delphinusfrontatus along with severalother types of smalldolphins Cuvier, 1817!. In 1823,an importantaddition was publishedindicating that this previouslyidentified animal was in fact a new species
Cuvier, 1823!. Lesson 828! lists this specieswith a newname, Delphinus bredanensis and this is the first indication of a name change &om frontatus in historical literature.
This specieswas includedin the broaderDelphinus genus until Gray 846! divided
Delphinusand established the Steno genus. Accordingto Gray 846! the Stenogenus included: The Steno,Steno rostratus, I he Malay Dolphin, Stenomalayanus, the Fronted
Dolphin, Stenofrontatus, the Compressed-BeakedDolphin, Steno eompressus,the
Attenuated-BeakedDolphin, Stenoattenuatus, and the CubanSteno, Steno fuscus. 1 his
12 genuswas originallynamed after a Danishnaturalist and anatomist,Dr. NikoiasSteno
Gotch, 1995!. In 1934Steno rostratus was being used for this species,incorporating both the original nameand the new Stenogenus Richards, 1934!. In 1955 Muller and
Kellog introductedthe scientific nameSteno bredanensis and Fraserand Purvcs 960! classified the family Stenidae which included the Steno, Sousa and Sotalia genera
Today the rough-tootheddolphin, X bredanensis,is classified in the Subfamily Stenidae of the Family Delphinidae.S. bredanensisis the only speciesin the Stenogenus. The common namefor S. bredanensis,the rough-tootheddolphin. is associatedwith wrinkled enamel found on the teeth of this species Harrison and Bryden, 1988!. S. bredanensis skulls are part of museumcollections in many locations including Marseille, I'lorence,
London, WashingtonD.C. and Honolulu.
Status of Nild Populations
'I'hestatus of the speciesS. bredanensisis describedas insufficientlyknown, and there is no estimateof worldwidepopulation numbers Klinowska, 1991; Leatherwood and Reeves, 1983!. Only one true abundance estimate is available from extensive cetaceansurveys in the easterntropical Pacific. In this region,approximately 150,000 rough-tootheddolphins are present,with confidencelimits ranging between 100,000 and
250,000 individuals Wade and Gerrodette,1993!. A summaryof the populationstatus of S. bredanensisin otherareas, including the NorthPacific, Western North Atlantic, and
Gulf of Mexicostates that populations are unknown Young et al., 1993!.
S. bredanensisis describedas a widely distributedspecies throughout warm watersof the worldsoceans. Although isolated sightings of rough-tootheddolphins have beenreported from severalregions, distribution of this speciesis centeredin tropical and sub-tropicalareas. Live, free-ranginganimals are believedto generally inhabit seaswith a water temperaturegreater than 25'C Klinowska, 1991; Leatherwood et al.. 1982;
Leatherwood and Reeves, 1983; Miyazaki and Perrin, 1994!. However, in the Canary
Islandsrough-toothed dolphins are quite commonly sighted where summertemperatures only range between 22 and 24'C and reach as low as 17-19'C in the winter Ritter,
2002!. Other than the presenceof this species around the Canary Islands, a ~e extension has typically been described when sighted in cooler waters Miya7Mi and
Perrin. 1994!. For example,a few strandedrough-toothed dolphins have beenreported off the Washington and Oregon coasts in the North Pacific, and a range extension documented in the South Atlantic off Brazil Ferrero et al., 1994; Ott and Danielwitz
1996!. '1'he rough-toothed dolphin is found throughout the eastern tropical Pacific, although at low densities,as indicated from cetaceansurvey work in this area Au and
Perryman, 1985; Hewitt, 1985; Wade and Gerrodette, 1993!. S. bredunensis is also presentaround at least someisland groups in the Pacific, such as llawaii Mazuccaet al.,
1999; Miyazaki and Perrin, 1994; Mobley et al., 2000!. Iwatherwood et al. 982! may have been the first authors to recognize and draw attention to the presenceof S. bredanensisspecifically in Polynesia. Study by Gannierand Gannier 998! has further indicated that Frerx:h Polynesia representsan area where S. bredanensisis found in relatively high concentrationcompared to other cetaceanspecies. Surveysof cetaceans in this region have ranked S bredanensisas the second most common speciesto be sighted in these waters Gannier, 2000!.
14 Slmwdiags of Qrao brefuacase
1lxre are certainly records of both massand individual rough-tootheddolphin strandingsIrom severalareas Chantrapornsylet al., 1996;Ferrero et al., 1994;Gaspar et al., 2000; Mazuccaet al., 1999; Miyazaki, 1989; Parsons,1998; Smith and Whitehead,
1999;Stolen et aL, 1999!. However,this cetus> speciesis usually describedas one that only occasionally strands,or where there are relatively few records of S. bredanensis strandingscompared to that of other cetaceanspecies Ellis, 1982; Leatherwoodand
Reeves,1983!. It has beensuggested that this may be due to the pelagic natureof rough- tootheddolphins, as off-shore speciesare less likely to be found washedashore Baker,
1983; Bruyus, 1971!. Mass strandingsof S. bredunensishave occurred, primarily in tropical or sub-tropicalregions which is in agreementwith the world-wide distributional information that is available for this species. Severalevents have beendocumented from the Florida area, dating back to 1965, and specimensinvestigated from three separate eventsduring the 1990's Stolen et al., 1999!. In 1976,two massstrandings of rough- tootheddolphins occurred,one off of Hawaii and the other in Virginia. Rough-toothed dolphin skulls also indicate a smaller-scaleevent that either occurred in the Indonesian archipelagoor Irom Senegalat somepoint in the past Rudolph et aL, 1997!. From more temperateregions, generally only a single individual or very few animals have been documentedto strand. For example, Irom the MediterrammtScL skulls are included in museumcollections in France,England, and Italy note by Schevill cited in Watkins et aL, 1987!. Along the Washingtonand Oregoncoast, Balcomb 980! reporteda single individual found in Washington, and Ferrero et al. 994! described three separate animals found in this area. Other regionswhere individual strandingshave beenknown to occurinclude Brazil, the Galapagos, Hong Kong, and Thailand Chantrapornsylet aL,
1996; Hetzel and Lodi, 1998; Palacios,1995; Parsons, 1998; Smith and Whitehead,
1999!. FromFrench Polynesia, there are accounts of rough-tootheddolphin strandings from the Windward Islands Gasparet J~.,2000!.
Biologyof Stc'aobmfaacase
The physicalappearance of S. bredanensisis distinct from all other cetacean species.The rough-tootheddolphin has a long and slenderbeak, a uniquehead shape where the beak blends with the forehead and o Benthe white coloration is apparent on the tip of the beak Ellis, 1982;Leatherwood et aI., 1976!. I hisspecies tends to havelarge eyes,a dark gray coloration,and a high numberof scars,marks and lighter colored splotches.External markings appear to increasewith age Leatherwoodet aL, 1982!.
Adult S. bredanensisrange from approximately210 to 240 cm in length,and weigh around100 kg Leatherwoodet aL, 1976!. The Sinu hasa leanbody shapecompared with other cetaceansof a similar size. Previousstudies focusing on the physiologyof S. hredanensis include documentation of internal body temperature, where the core temperatureof the rough-tootheddolphin was found to be lowerthan that of thespinner dolphin Whittow et al., 1978!.
In termsof thereproductive biology of S. bredanensis,very few publishedreports are available. This presentsa ckar lackof necessaryinformation for effectivespecies conservationefforts, as detmitionof reproductiveparatneters is requiredfor population modeling. The needfor knowledgeon reproductivebiology has beenemphasized as paramountfor conservationmeasures aimed at the rough-tootheddolphin Klinowska, 1991!. Only one study has addressedreproduction in S. bredanensisto date. Miyazaki
980! examinedreproductive organs and aged adult specimenscaught in a Japanese drive fishery. His flings indicate that at 14 GLG's and a length of 225 cm males are sexually mature,and femalesare mature at an estimatedage of 17 yearsbased on GLG's with a length of 225 cm. llowever, no immaturespecimens were aged,and even basic information such as what age or size range the rough-tootheddolphin attains sexual maturity is not yet known for this species. Miyazaki's work 980! suggestedthat X bredanensismay undergoseveral simultaneous ovulations by examinationof corpora as describedby Perrin and Donovan 984! for cetaceans.flowever, no investigationshave yet been made into reproductive seasonality.reproductive rate, or any other specific
Florida,and theseauthors were able to generatepreliminary growth curvesfor both males and females,indicating that malesand femaleshave attainedmaximum growth at 238 cm for males and 232 cm for females, both around 15 years of age. Unfortunately, no reproductiveinformation is availableto accompanythese aged individuals. Behaviorof CaptiveSteno bredanensis Most observationsand study of rough-toothed dolphin behavior cortes Gem animals held under captive conditions. During the 1960'sand 1970's,22 animalswere capturedin Hawaiianwaters. Theseanimals were on exhibitand trained at sea,some of which left during training sessionsor were purposely released Norns, 1965!. ln the 1970'sand 1980 s, Japancaught rough-tootheddolphins to be housedin oceanaria.and of these, 10 animals were measured for body length and weight during this time Miyazaki. 1980!. Rough-tootheddolphins have also been taken and maintained in captivity in Europe in the past. but not recently Miyazaki and Perrin. 1994!. In the 1990's, 5'. bredanensiswas maintained in captive conditions in the lagoonal waters of I'rench Polynesiawhere biological sampleswere collected from these individuals on a regular basis. In the late 1990's strandedanimals were rehabilitatedin the United States at Mote Marine Laboratory and many serum samplescollected. Currently, only one rough-tootheddolphin is known to be held in captivity in the United States. Although the number of rough-toothed dolphins that have been held under captive conditions throughout history is very small compared to that of bottlenosedolphins, the unique behavior and advancedtraining skill of .S.bredunensis has beenwell documented. The fIrst training experienceswith this speciesdate back to the 1960's in IIawaii, and there are several accounts of surprising, spontaneousand creative behaviors, as well as contrasts provided betweenthe rough-tootheddolphin and the more docile bottlenose dolphin Norris, 1974; Pryor, 1975!. 1 heability of this speciesto create its own novel behavior without traditional shaping methods led to its description as an intelligent, investigative animal that can becomeagitated at trainer mistakesand has the ability to manipulateits captiveenvironment Norris, 1974;Pryor, 1975!. Therehave been a few captive situations where S. bredanensis has been housed with the bottlenose dolphin I'ursiops Iruncatus. At SeaLife Park,the only known hybrid betweenS bredanensisand T. Iruncatus was born healthy. and this animal lived until almost 5 yearsof age Dohl et al., 1974;Pryor, 1975!. Recentlyin French Polynesia,a strandedX hredanensiscalf was documented to induce lactation in an adult female T. Truncalus that demonstrated adoptivebehavior to this youngcalf Gasparet aL, 2000!. Notablebehavioral characteristicsin S. hredanensisdo not appearto be limitedto thecaptive environment. ln thewild, care giving behavior among rough-toothed dolphins was observed in Brazil, where an adult continuedto providecare for a young dead animal Lodi, 1992!. intelligenceand the ability to manipulatetheir environment also appears to bethe case for free-rangingrough-toothed dolphins in Hawaii,where several attempts by National Marine FisheriesService to solve the problem of bait stealing by this specieshave been unsuccessful Schlais, l 984!. Statesseej of PN~ Exploitation and environmentalstressors pose potential threats to the future of A. bredanensispopulations. This study on the rough-tootheddolphin was conductedto establishbasic parametersassociated with the biology and ecology of the species. The goal in accumulating this data was to provide species knowledge to aid in future population assessmentsand ultimately contribute to conservation of rough-toothed dolphins throughoutthe world. Severalapproaches are neededto define populationand individual level characteristicswhich representthe current statusof S. bredanensis.Once modalitiesare employedand baselinedata becora:savailable on the ecology and biology of X bredunensis,a startingpoint will be provided for comparingthe populationstatus of this speciesinto the future. Regionsof preferred habitat needto be identified for 5'. bredanenvis,and surveys of free-rangingrough-toothed dolphin distributional patterns around oceanicislands were undertakenas part of this study. There is no evidencethat rough-tootheddolphins in f'rench Polynesiaare currently under exploitation stressand this island archipelago was used to describe the distribution and group size of an ecologically healthypopulation of this species. Life-history parameterswerc investigated by examination of dead specimens to determine the age and size range that h. bredaneavisattains sexual and physical maturity. SIeno populations experiencing extremeexploitation stresshave yet to be identified, and any nieasureof sexualmaturity attainment is needed for determining impact effects. This information can then be utilized to assesspopulation response to large scalepressures facing this speciesin times of rapid change to marine ecosystems. Opportunistic biological sampling was also conducted Rom seven individual rough-toothed dolphins in captive conditions to 20 investigateme;isures of stressand reproductivestate in live animals.Some of these animals had stranded and were in ill health, while others served as controls for comparisonof endocrinologydata according to stresslevels of individuals. The overallgoal of this study on the speciesS. bredanensiswas to addressthe followingquestions: What are the biologicaland ecologicalcharacteristics of current free-rangingSteno populations believed to representecologically healthy animals? How do physiologicalpirameters compare betweenhealthy rough-toothed dolphins and individualsexperiencing stress? I his researchwill providebasic tools to assessthe health statusof individualsand populations into the future. lt will also provide scientific data to aid in determiningif and when suspectedenvironmental threats are harming wild X hrcdanensis populations,allowing for proactive responsiveby wildlife managersto ameliorate the environmental stressors. CHAP1 ER I Distributionof the rough-tootheddolphin Stenobredunensis! around the Windward Islands French Polynesia! A. Gannier'. K.L. West. PROJECTED JOURNAI. SUBMISSION: Pacific Science 1~631/02 Laboratoired'Ecologic Marine, U.P.F., BP6750 Faaa Aeroport. TAHITI, French Polynesia Laboratoired'Ecologic Marine, U.P.F., BP6750 Faaa Aeroport, TAHITI, French Polynesia,University of Hawaii,Department of Physiology,John A. BurnsSchool of Medicine, 1960East-West Road, BiomedT-608, Honolulu, Hl Correspondenceto: [email protected] 22 ABSTRACT Therough-toothed dolphin Steno bredanensis! hasbeen described asa primarily pelagiccetacean species, found in nest oceans ofthe world. Littk information isavailabk ondistribution patterns or habitatpreference forthis species from any region. This study reportson thedistribution of S. bredanensisaround the Windward lshnds of French Polynesia.Data was obtained from vessel surveys between 1996 and 2000 where rough- tootheddolphins were sighted 38 tinm. Ciroupsizes of rough-tootheddolphins ranged between1 and35 individualswith an average size of 12.1individuals. When corrected for effort.results indicated that in FrenchPolynesia. S. hredunensis is found over a widearea, but is morecommonly distributed in-shore than oAshore. Rough-toothed dolphins were mostcommonly sighted 1.8 to 5.5 kmfrom the barrier reef. in waterdepths between 1000 and 2000 m. Our resultsalso demonstrate the year-roundpresence of thisspecies around Tahiti and Moorea X bredunensishas been reported in manyoceanic archipelagos, and our findingsmay provide insight into prefemM habitat and smallscale oceanographic conditionsassociated with regions where this cetacean species is relatively abundant. 23 INTRODUCTION Informationon rough-tootheddolphin steno hredanensis!distributions are limited, and publishedreports which includethis speciesare quite widespread. X hredanensisis rarelythe focusof field studies,but is oAenjust mentionedfrom a few sightings.Present in almostall tropicaland subtropicalregions, sighting data suggests that 3'. hredanensisis not a particularlycommon species in any area Au and Perrynian, 1985; flewitt, 1985; Mullin et al., 1994; Wade and Gerrodette, 1993;!. Previousreports on the distributionot rough-tootheddolphins are primarilyfrom the astern tropicalPacilic, covering very largeregions of' openocean waters. These studieshave extended over 50 degreesof latitude,and in thisopen ocean environment. a largescale distribution map for X hrectunensishas been generated from approximately 40 sightings Wade and Gerrodette.1993!. Otherlarge scale surveys in the easterntropical Pacificsupport the presenceof 5 hredanensisin relativelylow abundancescompared to other delphinidspecies in this area Au and Perryman,1985; flewitt, 1985!. Cetacean surveyscovning large areasot the westerntropical indian Ocean have also mapped ,'i.hredanensis sightings, where the percentageof cetaceanabundance represented by this speciesappear similar to thatof theeastern tropical Pacilic and to the Gulf of Mexico wherethis species is alsopresent BaHance and Pitman,1998!. An aerialcetacean survey in the Gulf of Mexicodocumented only a singleX hredunensissighting, but in a water depthvery similar to theinshore areas of FrenchPolynesian islands -1000 m! Gannier, 2000; Mullin et al., 1994!. X bredunensismay also lrequent coastal areas. Basedon strandedanimals and a few docum:ntedsightings, it appearsprobable that rough-tootheddolphins inhabit a 74 shallow,lagoon environrmnt in Brazil Andreet al., 1997;Ott andDanikwicz, 1996!. Opportunisticsightings, as wellas specimensfrom drive fisheries and strandings are availabkto indicateits presence,at leastoccasionally, in someareas such as the Mediterranean,off of theOregon and Washington coasts VSA!, andin Japan Ferrero, 1994;Miyazaki, 1980; Watkins et aL, 1987!. Cetaceansurveys near island environments oAen indicate the presence of rough- tootheddolphins. On a similarsade to thatof FrenchPolynesia, S. bredanensishas been sightedduring aerial surveys in Hawaii,but is oneof theleast common cetacean species found in this archipelago,and is rarely sightedwithin 5 miks of shore Mobley et al., 2000!. Dedicatedor opportunisticsurveys have indicatedfrequent sightings ot S. bredanensisoff the CanaryIslands Martin et aL, 1998!. Rough-tootheddolphins have alsobeen sighted in theIndonesian Archipelago, although not frequently Rudolph et al., 1997;Rudolph, 1998!. I'rom other archipelagos such as the Philippines, galapagos, and the SolomonIslands, rough-toothed dolphins have not beensighted during cetacean surveys,although there is reportof 5'.bredanensis asa speciesinvolved in SolomonIsland drive fisheriesand a strandedspecimen has been recovered from the CJalapagos Dolar et aL, 1997;Smith and Whitehead,1999; Rudolph et aL, 1997!. Strandedspecie:ns have also beenrecorded off of Hong Kong and in Thailand Chantrapornsylet al., 1996; Parsons,1998!. This suggeststhat rough-tootheddolphins are presentaround several sub-tropical/tropical archipelagos, but generallyin low abundances. AlthoughS. bredanensisdistributions have been assessed over the wide eastern tropicalPacific, which may allow for insightinto preferredhabitat associated with large scaleoceanographic conditions, smaller scale studies on thedistribution of this s~ies 25 haveyet to be undertaken.In FrenchPolynesia, cetacean surveys have indicated that S. bredanensisis the secondmost commnnly observed s~ies aAer5tenella Ionpirostris Gannier,2000!. This provides an opportunity to determinethe habitat preference of this >acies aroundsmall oceanic islands. Based on multiplesightings of rough-toothed dolphins,and on samplingeffort spread over 4 years of which2 yearsthe second author participated!,this study reports on seasonal and temporal variation in distributionaround the islands of Tahiti and Moorea. METE IODS Area of study The studyarea focused on Tahitiand Moorea, which are the mainWindward Islands Society Archipelago!, located in thecentral tropical Pacific at a latitudeof 17S anda longitudeof 150W FigureI!. Bothinshore and off-shore areas v eresurveyed. Dueto the 12-18' slopeof thesevolcanic islands, the water rapidly reaches depths in excessof 2000m. within the inshorearea. Although large scale primary production does not occurin thesouthern central tropical Pacific Longhurst, 1999!, passes may provide nutrientsin thephotic zone, due to theoutflow of lagoonand river water. Edge effects maycause local eddy-induced vertical mixing, as shown around several tropical oceanic islands Aristegui et al., 1997!.The offshore area lies beyond 10 km of thebarrier reef, withwaters generally deeper than 3000 m. Its hydrobiologyis dominated by oligotrophy and a very deep 00-300 m! and stablethermocline. However, the seasurface temperaturefeatures a significantseasonal change, with lowervalues of 25-26'Cin August-Septemberandhigher temperatures of 29-30'C in February-April.These normal 26 seasonaltemperatures may shiA duringan El Nino SouthernOscillation event, which refersto thewarm phase of a naturallyoccurring sea surface temperature oscillation in the tropicalPacific. The studyarea was sub-divided into 4 regionsto investigatedistribution of rough- tootheddolphins Figure 2!. Thefirst region is the channel located between the islands of Tahitiand Moorea, the secondcovered the rest of Mooreaup to 10 kilometersRom shore which is essentiallya leeward area. The third regioncomprised all areasoff Tahiti. excludingthe channeL mostly with a windwardexposure. and extended to 10km offshore. Thelast region included offshore waters, beyond 10 km of shoreand extended to thesmall islandof Maiao,west of Moorea Figwe 2!. A seriesof 11 surveyswere conducted between March 1996and May 2000 using a 12m sailboat.Sampling took place when wind speed was less than or equalto Beaufort 4, primarilyusing diesel propulsion atan average speed of 9.5km.h '. Bothinshore and offshorestrata were sampled, and zig-zagtracks were conducted around the islands wheneverpossible. A GPSsystem and an auto-steering device were used for positioning andnavigation. During the surveys, two or threeobservers stood on thedeck, about 2 metersabove sea level and searchedthe sideand frontal sectors.Position was recordedon averageevery 3 km,as well as wind, sea state and relative cloud cover. When cetaceans weresighted, a GPSposition was immediately recorded and radial distance and bearing wereestimated. Schools of rough-tootheddolphins were systematically approached to estimateschool size and composition, and for recordingof podactivity. 27 Date processing Data obtained from samplingwas enteredinto a database Dbase IV!. and made compatiblewith Oedipe Xlfremer GeographicInformation System Masse and Cadiou. 1994!. GIS was used for mapping, data post-stratification and effort calculations. Accuratebottom depthand distanceto the barrierreef for eachsighting were obtained Irom nauticalcharts issued by SHOM, the FrenchNavy HydrographicService. Numerous parameters may cause dolphin detectability to vary Hiby and Hammond. 1989!. However, for a given survey protocol i.e. sameplatform and same numberof observers!and a givenspecies, the assumptionof constantdetectability holds if sightingconditions are constant.Sighting conditions may vary with the wind speed,the swell height and the luminosity. Becausewind in excessof Beaufort 3 may haveadverse effects on the detection of small cetaceans Buckland et al.. 1993; Hihy and Hammond, 1989!, only survey effort in conditions betweenBeaufort 0 to 3 were used to estimate sightingrates. As ow survey effort was not homogeneousacross the different strata regions, periods,areas! distribution had to beeffort-corrected. Hence a meansighting rate y~!was computed,relating to the numberof rough-tootheddolphins seen per kilometer.For estimatingy~ we retainedthe samesampling unit as that usedfor data recordingand computeda sightingrate yi for eachsample, y,= n,. S,i I, where ni is the number of rough-toothed dolphin schools detected during the sample in fact n = 0 or n = I !, Si is theschool size and li is thesample length. Our data set includedmany 0 values due to the short samplelength -5 km! used,and 2S the arithmetic mean was not adequate for estimating sighting rates. A Pennington estimator Penningtonand Berrien, 1984! was used insteadto computeaverage sighting rates. This estimator is basedon the log-normaldistribution of the non-zerovalues of a serie of data, referred as delta distribution by Aitchison and Brown 957!. This estimator producesunbiased estimates of meanand variancewhen the dataset includesmany zero values,unlike the arithmetic estimator. For a distribution of n sampleswith m non-zero values m>l!, with y the samplemean of the non-zerolog, valuesand s' a sample varianceof the log, values,Aitchison and Brown 957! give the estimateof the mean.y~: y,= m/n. exp y ! . G s /2! Pennington 983! givesthe estimatorof the variance.var yp!: var y,! m/n . exp .y ! . [m/n . G s /2! - m-1/n-I !.G m-2/n-I !.s!] where G x! is defined by: G x! = I + m-I/n!.x+ E.. [ m-I! ' '. x'] /[m' m+1! m+3!... m+2j-3!.j] These estimateswere obtained by processingdata with Passtecsoftware Ibanez and Etienne,1994!. T-tests were then usedto comparethe meany~ estimates. The seasonalvariation of the rough-tootheddolphins sighting rate was investigated by processingthe inshoredata a secondtime: the datawas split into threetemporal strata, by consideringa cold waterseason for datacollected &om Septemberto November,an intermediateseason Irom December to February and a warm water seasonfor data obtained &om March to May. Seasonalsighting rates were estimatedwith the same methodologyas above.As our samplingeffort was greaterin RegionI and 2, which representedthe channelregion and Moorearespectively, only thesetwo areaswere 29 consideredfor analysesof seasonalchanges. RESULTS Effort Our total surveyeffort over the entirestudy area was 6458 km Table I!. An effective effort of 3441 km was conductedin the channelarea, 1706 km around Moorea. 758 km aroundTahiti, and 553 km in the off-shoreregion. For eachregion. effort was furtherexpressed according to Beaufortconditions experienced during the survey Table I!. In Regionl. 2 and3 the channelarea, Moorea and Tahiti respectively!, etTort in excellentsighting conditions Beaufort 0-1! representedI2.5-15' ot the totaleffective effort.effort in verygood sighting conditions Beaufort 2! represented3 -40' . andeffort in goodsighting conditions Beaufort 3! represented45-53/o. Ilence. average sighttng conditionswere similarexcept for the offshorearea. where Beaufort3 conditionswere moreprominent. comprising 64.5'/0 of theeffective effort. Three different time periods duringthe year were investigated for seasonal effects in RegionsI and 2: an effective effort of 2157 km of effortbetween September and November Figwe 3a!. 559 km was obtainedbetween December and February Figure 3b!, and2475 km duringMarch to May Figure3c!. Sigktingsand GroupSize Therough-toothed dolphin was observed 38 timesover the study period. Smaller schoolsizes of S.bredanensisoccur more commonlythan larger group sizes Figwe 4!, averageschool size was 12.1 SE=10.2!individuals with a rangebetween I and 35 30 individuals. Out of our 38 total sightings,29'/o of rough-tootheddolphin groups had at leastone calf presentwhere a calf wasdefined as an animalthat appearsto be either newly born or still maternallydependent. Rough-toothed dolphin school size was found to be homogeneousacross all four sub-regions.In the channelregion 5 sightings!the average group sizewas 13.5 individuals SE=12.2!, ofF Moorea 1 sightings!the averagegroup size was 12.8 SE=10.9!. off Tahiti, S. bredanensiswas sightedon 9 occasionswith an averagegroup sizeof 8.3 individuals SE=3.0! and in the off-shoreregion. the specieswas observed3 timeswith an averagegroup size of 14.0 SE=12.5!. In the Windward Islands, the rough-tootheddolphin was sometimesfound in mixed schools,once associatedwith bottlenosedolphins, and twice with melon-headedand Fraser'sdolphins. 5 hredanensis was also observedinteracting with humpbackwhales on two occasions. Although both the spinner dolphin and pilot whale are commonly found in FrenchPolynesia Gannier, 2000!, rough-tootheddolphins were never seen associatedwith either of these species during our surveys. Distribution S. bredanensis was sometimes observed less than 100 meters f'rom reef barrier. and occasionallymore than 30 kilometersofFshore, indicating a widedistribution range. When consideringthe entire study area, the total distributions correctedfor effort indicate that S. bredanensisare found at a rate of 0.124 individuals/lun.Inshore regions sub-regionsI to 3! indicateda higher numberof S. bredanensisthan the offshorearea sub-region 4!, with 0.238 and 0.079 ind./km respectively Table 2!. However, samplesize in the offshore region was smalland this differencewas not significant T-test. p>0.10!. 31 Effort-correcteddistribution of rough-tootheddolphins was considered according to distanceIrom the barrierreef, bottomdepth. and different sub-regions sampled Table 2!. Overall,S. bredanensiswas moreIrequently sighted 1.8-5.5 km off-shore,with 0.42 ind./km n=19; SE = 0.17!, than in the 0-1.8 km area, with 0.07 ind./km n=8; SE = 0.03!. This differencewas statisticallysignificant T-test, p<0.001!. Likewise,rough-toothed dolphinswere more Irequentlysighted in depthsbetween 1000 and 2000 m. with 0.19 ind./km n=13; SE =0.08!, than in the 0-1000 m stratum, with 0.17 ind./km n=18; SE = 0.06!, althoughthis difference was not significant T-test, ~0.10!. Effort-correcteddistributions were very similar off Tahitiand Moorea sub-regions 3 and 2!, with respectivevalues of 0.33 ind./km SE = 0.13!. and 0.27 ind./km SE = 0.13!, and somewhatlower in the channelregion sub-regionI! with a figureof 0.19 ind./km Table 2!. Pairwisestatistical comparison indicated that only Tahiti and the channelregion were significantly different T-test, p<0.05!. In summary,rough-toothed dolphins were foundover anextensive range. but most commonlyobserved in inshorewaters 1.8 to 5.5 km off the barrierreef, in a depthrange of 1000to 2000 m, andseemed to beevenly distributed around the Windward Islands. Seasonal vuriatios Differencesin distributionwere apparentbetween the colder water season Septemberto November!,medium water temperature range Decemberto February!and the warmerwater season March to May!. Duringthe cold waterseason September to November!,a sightingrate of only 0.10 ind./km n=8; SE= 0.05! was obtained,much lower than for the two otherperiods. Between December to Februarythe sightingrate 32 was 0.21 ind./km n=5; SE = 0.14!, closeto the figureobtained for the warmseason Marchuntil May!, with a sightingrate of 0.15 n=l 1; SE = 0.08!. However,the cold seasonestimate was probably influenced by an uneveneffort distribution in thatperiod September-November!,when sampling favored inshore waters as surveyswere focused on humpbackwhales Figure 3!. DlSCVSSlON Our distributionresults in the WindwardIslands indicated that rough-toothed dolphinsexhibit a specifichabitat preference in FrenchPolynesian waters. S. hredaneasis ismost &equently found in waterdepths between 1000 and 2000 m, approximately1.8 to 5.5 km &omshore. These results do not demonstratea purely pelagic habitat usage in this region.in contrastto manygeneral reports describing S. hredanensis asonly a deep-water species Leatherwood et al. 1982;Leatherwood and Reeves, 1983.- Miyazaki and Perrin. 1994!. In Hawaii S. hredanensisis apparentlyfound fartheroffshore Mobley, 2000! sometimesin water more than20 milesoffshore, but wheredepths are similarto those foundcloser inshore in FrenchPolynesia Schlais, 1984!. I-lowever,water depth alone wasnot foundto be a significantfactor affecting sighting rates in our study Table 2!, despitea probableinfluence on prey availability. It is not known which specific species of cephalopodsand fish S. bredanensisin French Polynesia feed upon, however during our studyrough-toothed dolphins were observed feeding on epipelagic fishes including flying fishes!during the day. In otherregions of theworld, the inferred diet of thisspecies has beendescribed &om stomachcontents, which have usually been classified as off-shore or pelagicspecies of fish,octopus, and squid Leatherwood et al., 1982; Leatherwoodand 33 Reeves,1983!. Inshoreprey species have been suggested &orn stomach contents of strandedindividuals in Hawaii. althoughan offshoredistribution of S. bredanensisis apparentinthis region Shallenberger, 1991!. By-catches reported by Monteiro-Neto etal. 000! &om coastalgillnet fisheries off' northeasternBrazil also suggest that rough- tootheddolphins enter inshore waters at leastperiodically. In the WindwardIslands of FrenchPolynesia. our overallaverage sighting rate of S. bredanensiswas 0.124 individualslkm.Although comparisons are difficultas survey platformsand observation protocols are specific to eachstudy. sighting rates in other oceansdo indicatelower &equenciesof S. bredanensiswith estimatesof 0.0047 individuals/kmand 0.00258 individuals/kmin the easterntropical Pacific and western tropicalIndian Ocean, respectively Ballance and Pitman, 1998; Wade and Gerrodette, 1993!.Rough-toothed dolphin importance in delphinidpopulations may betterbe comparedin term of sighting&equency. S. bredanensisranks second in the Society Islandswith 34/o of on-effortdelphinid sightings Gannier, 2000!. In the Marquesas.the northernmostarchipelago of FrenchPolynesia, the speciesranked sixth with a sighting &equencyof 4.0'/o Gannier, 1999!. In theSolomans, rough-toothed dolphin ranked Mh amongdelphinids with a sighting&equency of 5.0'/o Shimada and Pastene, 1995!. In the SuluSea, in the Philippines,the speciesalso ranked Mh, with a sighting&equency of 2.9'/o Dolar et aL, 1997!. S. bredanensiswas rankedas the ninth most commoncetacean speciesto be sightedin boththe eastern tropical Pacific and the western tropical Indian Ocean Ballanceand Pitman,1998; Wadeand Gerodette,1993!. Hence&om results availablein the literature,French Polynesia, and in particularthe WindwardIslands, appearto be a favoredarea for S.bredanensis.Certainly, this species is presentaround 34 manysmall oceanic islands in warm temperateto tropical regions. Our seasonaldistribution results indicated surprising differences: the lowest seasonalsighting rate was obtainedduring the cold water season,and the highestsighting rate during medium water temperaturevalues Decemberto Febru'ey!. This does not agreewith relativeabundance estimates of Gannier 000! whichindicated that delphinids are generally3 timesmore likely to be sightedduring cold water periodsas opposedto the warm seasonin the Society Islands. However, the melon-headedwhale and Fraser's dolphinaccount for two of thesefour delphinidspecies, and these species were sighted in very large groups during the cold water season,influencing the result presentedfor all delphinidspecies in Gannier 000!. It is also possiblethat ow resultson S. bredanensis seasonaldistributions are at leastpartially effected by an unequalsampling effort regarding depth and distancefrom shore betweenthe seasons. During the cold water period, our efforts were more concentrated on shallower areas within I mile of the barrier reef becausesurveys were sometimesfocused on humpbackwhales during that portion of the year as comparedto the other seasonaltime periods Figure 3 a, b, and c!. This may influencethe lower sighting rate of S. bredanensisduring the cold season,as ow results show that this speciesis more commonly found at a distanceof l.8 to 5.5 km &om the barrier reef, and in deeper waters. Ackaowledgemeals We are especiallygrateful to Claude Payri and the Laboratory of Marine Ecology at the Universitede la PolynesieFrancaise for 6nancialsupport and making this project possible. The Delegationa la Rechercheof the Territoirede PolynesieFrancaise also 35 kindly pro~idedfunding towardsour research.We would like to thank Dolphin Questfor supportof the secondauthor's Ph.D. thesis,of whichthis work is a part. 36 REFERENCES Aitchison J. and J.A.C. Brown. 1957.The Lognormaldistribution. CambridgeUniversity Press,Cambridge. U.K. Andre. P., F. Carvalho,and A. Ximenez. 1997.Observation on the rough-tootheddolphin Steno hredanensis off Santa Catarina Island, Southern Brazilian coast. Biotemas, 10 !:71-79. Aristegui J.P., A. Tett. G. Hernandez-Guerra,M.F. Batserretxea,K. Montero, P. Wild, S. Sangra,M. Hernandez-Leon,J.A. Canton, M. Garcia-Braun,S. Pacheco,and E.D. Barton. 1997.The influenceof island-generatededdies on chlorophyll distribution: a studyof mesoscalevariation aroundGran Canaria.Deep-Sea Research I. 44: 71-97. Au D.W.K., and W.L. Perryman.1985. Dolphin habitats in the easterntropical Pacific. Fishery Bulletin, 83: 623-643. Ballance L. I .. and R.L. Pitman. 1998.Cetaceans of the WesternTropical Indian Ocean: distribution, relative abundance,and comparisonswith cetaceancommunities of two other tropical ecosystems.Marine Mammal Science,14 !: 429-459. Buckland S.T.. D.R. Anderson, K.P. Burnham, and J.L. Laake. 1993. Distance sampling: Estimatingabundance of biological populations.Chapman and Hall Ed., London. UK. Chantrapornsyl,S., K. Adulyanukosol,and K. Kittiwathanawong.1996. Records of Cetaceansin Thailand. PhuketMarine Biological Center ResearchBulletin, 61:39- 63. Dolar M.L., W.F. Perrin. A.A.S.P. Yaptinchay, S.A.B.H. Jaaman,M.D. Santos,M.N. Alava, and M.S.B. Suliansa. 1997. Preliminary investigation of marine mammal distribution, abundance and interactions with humans in southern Sulu Sea. Asian Marine Biology, 14: 61-81. Ferrero, R.C., J. Hodder, and J. Cesarone.1994. Recent strandingsof rough-toothed dolphins, Steno bredanensis,on the Oregon and Washington coasts. Marine Mammal Science, 10 !: 114-116. Gannier,A. 2000. Distribution of Cetaceansoff the Society Islands French Polynesia! asobtained &om dedicatedsurvey. Aquatic Mammals,26 !: 111-126. Gannier,A. 1999. Distribution of cetaceansin the MarquesasIslands FrenchPolynesia! asobtained &om a dedicatedsurvey. Thirteenth Biennial Conferenceof the Biology of Marine Mammals, Abstracts. 37 Hewitt, R.P. 1985. Reactionof dolphins to a survey vessel: effects on censusdata. Fishery Bulletin, 83 !: 187-193. Hiby A. and P.S. Hammond. 1989. Survey techniques for estimating abundanceof cetaceans.Reports of the International Whaling Commission.Special issue 11: 47- 80. Ibanez F. and M. Etienne. 1994. Passtec, programme d'analyse des series spatio- temporelles en ecologic cotiere. User s manual. Observatoire Oceanologique Villefranche sur Mer!. I..eatherwoodS.. R. R. Reeves,W.F. Perrin,and W.E. Evans. 1982.Whales, dolphins and porpoisesof the easternNorth Pacificand adjacentarctic waters.NOAA Technical Report NMFS Circular 444: 1-245. Leatherwood S.. and R. R. Reeves. 1983. The Sierra Club Handbookon Cetaceans.Sierra Club Books, San Francisco, CA. Longhurst,A.R. 1999.Ecological Geography of theSea. Academic Press. San Diego. CA. Martin V., V. lani. and F. Schweikert. 1998. Cetaceanssighted in the Canary Islands during the Caremexexpedition January-April 1997!. EuropeanResearch on Cetaceans. 12: 77. MasseJ. andY. Cadiou.1994. Oedipe User's Guide.IFREMER Nantes!.38 pgs. MiyazakiN. 1980.Preliminary note on agedetermination and growth of therough- tootheddolphin, Sreno bredanensis, off the Pacificcoast of Japan.Reports of the InternationalWhaling Commission,3: 171-179. Miyazaki.N. andW.F. Perrin. 1994. Rough-toothed dolphin Steno hredanensis Lesson, 1828!.In Handbookof MarineMammals. Volume 5: The first bookof dolphins. eds! Ridgway,S.H. and R.J. Harrison,Academic Press, London. UK. Mobley, J.R., S.S. Spitz, K.A. Forney, R. Grotefendt,and P.H. Forestell.2000. Preliminaryresults of 1993-98aerial surveys. Southwest Fisheries Science Center AdministrativeReport LJ-OO-14C. 26 pgs. AvailableIrom SWFC,P.O. Box 271, La Jolla, CA 92038. Monteiro-Neto C., T.T. Alves-Junior, F J.Capibaribe Avila, A.A. Campos,A.F. Costa, C.P.Negrao Silva, and M.A.A. Furtado-Neto.2000. Impact of fisherieson thetucuxi Somaliajluviarilis! andrough-toothed dolphin Sienobredanensis! populations off Cearastate, northeastern Brazil. AquaticMammals, 26 !: 49-56. 38 Mullin. K.D., W. Hoggard.C.L. Roden,R.R. Lohoenfener.and C.M. Rogers.1994. Cetaceanson the upper continentalslope in the north-centralGulf of Mexico. FisheryBulletin, 92: 773-786. Ott. P.II.. and D. Danielwicz. 1996. Southwardrange extension of S'tenohredanensis in the Southwest Atlantic and new records of,Stenellu coeruleoalha for Brazilian waters.Aquatic Mammals.22 !: 185-189. Parsons.F..C.M. 1998. Strandingsof small cetaceansin Ilong Kong territorial waters. Journal of the Marine Biological Associationof the United Kingdom. 78:1039- 1042. Pennigton,M. 1983.EAicient estimators of abundance,for tish andplankton surveys. Biometrics, 39: 281-286. Pennington.M. andP. Berrien.1984. Measuring the precision of estimatesof totalegg produtionbased on planktonsurveys. Journal of PlanktonResearch. 6 !:869-879. RudolphP., C. Smeenkand S. I eatherwood.1997. Preliminary checklist of Cetaceain the IndonesianArchipelago and adjacentwaters. Zoologisches Verhandelingen. 312:1-50. Rudolph.P. 1998.Observation on cetaceansin notheasternSulawesi, Indonesia. FuropeanResearch on Cetaceans,12: 77. Shallenberger,E.D. 1981. The statusof I lawaiiancetaceans. Marine Mammal CommissionReport. MMC-77/23 NTIS PB82-109398.79 pags. Schlais.J.F. 1984.Thieving dolphins: a growing problemin Hawaii's fisheries.Sea Frontiers, 30 !:293-298. ShimadaH. and L.A. Pastene.1995. Report of a sightingsurvey oA the SolomonIslands with commentson Bryde'sWhale Distribution. Reports of the InternationalWhaling Commission, 45: 413-417. Smith D.S., and H. Whitehead. 1999. Distribution of dolphins in Galapagoswaters. Marine Mammal Science, 15 !: 550-555. WadeP.R. and T. Gerrodette. 1993.Estimates of cetaceanabundance and distribution in theEastern Tropical Pacific. Reports of theInternational Whaling Commission, 43: 477-493. Watkins.W.A.. P. Tyack,and K.E. Moore.1987. Steno bredanensis in the Mediterranean Sea. Marine Mammal Science, 3 !: 78-82. 39 Table1. Surveyeffort andsighting conditions where percentage of total in eachcolumn is given in brackets. 40 Table2. Meansighting rates for differentstrata in dolphin/100kmof effort!. 41 42 43 44 45 Figure 1. Studyarea located in the Windward Islands,French Polynesia. Figure2. Surveyarea was subdivided into four regionsfor comparisonof rough-toothed dolphin distribution. Regionsinclude the channel,Moorea, Tahiti, and offshore areas. Figure 3a. b. and c. Effective surveyeffort was correctedaccording to season.Effective effort is representedbetween September and November a!; Decemberand February b!; and March to May c!. Figure4. Group size for eachsighting of rough-tootheddolphins. 48 CHAPTFR 3 Reproduction in the rough-toothed dolphin, Steno bredanensis: Attainment of sexual and physical maturity K.L. West', J.M. Mead, and C.W. Potter PROJECTEDJOURNAL SUBMISSION: CanadianJournal of Zoology 02/31/03 Keywords:Rough-toothed dolphin, reproduction,sexual maturity. physicalmaturity. life-history, age, growth ' Universityof Hawaii Departmentof Physiology John A. Burns School of Medicine 1960 East-West Road, Biomed T-608, Honolulu, HI 96822 ' MarineMammal Program National Museum of Natural History Smithsonian Institution, Washington DC, USA 20560-0108 Correspondenceto: kristiwest.earthlink.net 49 ABSTRACT Theage and size range at sexualmaturity is requiredfor populationmodeling and eftective managementof cetaceans.In this study we report on the attainmentof sexual maturity in the rough-tootheddolphin, Steno bredanensis.Twenty-seven females and 22 males collected &om the Pacific and Atlantic oceans were examined at the Smithsonian Museumof Natural History. We estimatedthe ageof thesedolphins by counting growth layers in a preparedthin, longitudinal tooth section. We determinedphysical maturity according to fusion of vertebral epiphysesand reproductivestatus by histological and gross examination of reproductive organs. The histological examinations of uterine epithelial and mammary tissue indicate that female X hredanensisbegin the gradual process of sexual maturation between 3 and 6 years of age. Femalesreach sexual maturity by 9 to 10 years,at a lengthof 212 to 217 cm and a weight between101 and 108 kg. Our youngestphysically maturefemale specimenwas estimatedat 12 yearsof age, with a length of 210 cm and a weight of only 85 kg. Male rough-tootheddolphins attain sexual maturity between5 and 10 yearsof age,at a length of 216 cm and a weight of 92 to 102kg. Physicalmaturity in maleS. bredanensisis attainedbetween 5 and14 yearsof age, at a length of 227 to 231 cm, and weight of 119to 130 kg. Sexualdimorphism was evident in S. bredanensisand our findings provide the first report on sexual maturity attainmentand characterizationof diAerentreproductive states for this species. 50 INTRODUCTION Informationpertaining to the life-historyof .Stenohredanensis is limited. Significantinvestigations intoreproductive parameters inthis species have only focused on populationsfound in Japanesewaters Miyazaki. 1980; unpublished data cited in Miyazakiand Perrin, 1994!. Growth has been described as rapid during the first five yearsof life andthe attainmentof physicalmaturity in Japanesedrive specimens estimatedat 16 yearsof ageand at a lengthof 231cm Miyazakiand Perrin. 1994!. Miyazaki 980! examineda small sampleof individualsfrom two X hrcdanensis schoolswhere age and sexualmaturity wereassessed. A total of 39 specimenswere aged,but testeswere weighed for only ninemales. and only two of thesespecimens examinedhistologically for spermpresence Miyazaki, 1980!. The youngestmale examinedwas estimated at 14years of age,and smallest testis weight recorded at 175g. with no malespecimens classified as immature.Miyazaki's 980! examinationof femalespecimens did not includethe age or sizerange where sexual maturity is attained althoughfemales greater than 17 yearsof age were classifiedas mature. Ilowever. Miyazakihas estimated the attainmentof sexualmaturity for femalesfound in Japanese watersto occurat about10 years of ageand between lengths of 210-220cm as citedas unpublisheddata in Miyazakiand Perrin,1994!. In additionto the work in Japan.a strandedmale recovered in Oregonestimated at 7 yearsof agewith a lengthof 209 cm wasconsidered irranature, based on a testisweight of 23.3g Ferreroet al., 1994!.In the Gulf of Mexico, Florida, 34 X bredanensisspecimens obtained since 1990 &om strandingshave also been aged using tooth growth layer groups GI.G's; Stolen et al., 1999!,but no informationon the reproductivestatus of theseaged animals is yet available. Asymptotic length basedon a combinationof both sexeswas estimatedat 239 cm for Florida populationswhich is slightly greater than the combinedestimate of 231 cm for Japanesewaters Miyazaki and Perrin, 1994;Stolen et al., 1999!. It is not possibleto obtain extensive life-history information for rough-toothed dolphins Irom the large samplesizes that have been availablefor other speciessuch as the spinner. spotted and striped dolphin Chivers and Myrick, 1993; Klinowska. 1991; Miyazaki, 1984; Myrick et al., 1986; Perrin and Henderson,1984!.,S'. bredanensis specimenscollected since 1976were examined at the SmithsonianMuseum of Natural History. We examinedteeth, reproductivetissue, and necropsydata Irom approximately 50 individuals. Primaryobjectives of the presentstudy includedaging of individuals and gross and histological examination of both male and female reproductive organs to determinethe ageand size rangethat S. hredanensisattains sexual and physical maturity. We also aimed to describe histological parametersin tissuesrepresentative of different reproductivestates for this species. METHODS CollectedSpreisreas S. bredanensiscollection specimensat the SmithsonianMuseum of Natural History were studied,including 27 femalesand 22 males.Approximately two-thirds of thesespecimens were collected Irom one of two massstrandings events which occurred in 1976,one in Hawaii and the other in Virginia Beach. The remainingthird of the total specimenswere collectedby the National Marine FisheriesService as incidentalby-catch in the Northwest Pacific. From all specimens,sex, total body length,a grossassessment 52 of sexual maturity, and information regarding weights and dimensionsof reproductive organswere recorded at the timeof necropsy.From most collected specimens, weight of animal. and measurements of mid-dorsal, mid-lateral, and mid-ventral blubber thickness wererecorded. Vertebralepiphyses fusion was assessedin most individualsas either open,closed but visible,or fusedand invisible.In mostcases, reproductive organs were initially preservedin 10'Ioformalin and latertransferred to 7P/oethanol for longer-term storage. EstimatedAges Ages of all .S'.hredanensis specimens were estimatedby preparationof a thin, longitudinaltooth sectionfor countingof growth-layergroups GLG's; Hoh> 1985: Perrin and Merrick, 1980!. Growth-layer groups have been previously validated in bottlenosedolphins by countingof layersupon the deathof animalsof knownage Hohn, personalcommunication!. As depositionof growth layersmay be seasonallydependent and not as pronouncedin tropical regions,a captive rough-tootheddolphin from a tropical environmentwas usedto validatethe use of this methodfor the speciesS. bredanensis.In this knownaged animal, yearly growth layers were clearly visible, and only mild seasonalchange did not appearto affect clarity of depositedlayers. In all samples,growth-layer groups in the dentinewere counted. and each GLG wasassumed to signify one yearof growth.Counts were made independently by two readersto the nearest0.5 of a yearand thenaveraged. Key specimenswere determined as thosethat had either recently attained sexual maturity or were thought to be nearthis stage. For these animals, GLG's were counted a second time by both readersto ensurethe best 53 estimatepossible. Photomicrographs of each tooth section were also taken to allowfor further examinationin caseswhere ageswere questionable. In older individuals it was sometimesdiAicult to distinguishbetween GLG's and theseanimals were aged by the minimumcountable layers and denoted with a + sign i.e. 5++!. GrossEraatinatiow of ReproductiveOrgans Sexualmaturity in malespecimens was initially assessedaccording to testicular size. Rightand left testicularweights were taken and dimensions including length, width and depthmeasured for both testes.Testicular tissue was also preparedfor further histologicalexamination to confirm reproductive status of eachindividual. Femalespecimens were first examinedto determineif theanimal was lactating and/orpregnant, and fetussize recordedwhen animals were found to be pregnant. Presenceor absenceof corporawere used to categorizefemales as sexuallymature or immature, as it is assumedthat if one or more corpora are present the animal has previouslyundergone at leastone ovulation and is thereforeconsidered mature Perrin andReilly, 1984!. The number of corporalutea and corpora albicantia were recorded and the diametermeasured when possible. For eachovary, the weight and dimensions includinglength, width, and depthwere also recorded.Ovaries were qualitatively describedaccording to shapeand degree of foldingand wrinkling. Flat diameters of the leAand right uterine horns were measured. Uterine condition was initially assessed on a grossbasis as one of thefollowing: immature, mature, mature and dialated, mature and inactive,pregnant, or vascular. Uterineepithelial tissue was preparedfor further histologicalexamination Irom S. bredanensisindividuals of variousreproductive states. 54 Mammarygland length, width anddepth were measuredand mammary tissue described accordingto shadesof white,pink, tan, or browncoloration. Mammary tissue from individualsof variousreproductive states was also prepared for histologicalexamination. HistologicalExamination of Rcpn4nctiveOrgans Testicular, uterine and mammary tissues were prepared by embedding and stainingin hernatoxylinand eosin. A Modified Masson'sTrichrome stain was also preparedfor eachof the tissuesand an ocular micrometerused to take measurements from each of the slides. Male histological sections were tested for the presenceor absenceof spermatozoausing Berg's stain Berg, 1963!. Seminiferoustubule diameter was measuredeight times to obtain an averagefor each specimenhistologically examined.Uterine tissue was measuredfour times and averagedto obtain endometrium thicknessand myometriumexternal muscle layer, myometriumvascular layer, and myometriuminternal muscle layer thickness. Mammary tissue was measured four times to obtainan averagelobule diameter, alveoli diameter,fat cell diameterand m'immary ductlength. Mammary slides were also quanitatively assessed for percentcomposition of mammarytissue, ducts, connective tissue, fat, veins,arterioles, blood vessels,and parasites. Data Analyses An initial analysis was performed to compare the morphologicaland physiologicaltraits between individuals examined &om the Pacific and Atlantic oceans. Therewere no significant differences in males&om the Pacific and Atlantic ocean for all 55 11 traits examined:Seminiferous Tubule Diameter, Age, Body Length, Body Weight, RightGonad Weight. Left GonadWeight, Right Gonad Length, Right Gonad Width. RightGonad Depth, LeA Gonad Length, LeA Gonad Width, LeA Gonad Depth all P's! 0.05!. For females,there was no significantdifference for 9 of 11traits examined: Age. RightOvary Weight, Left Ovary Weight, Right Ovary Length, Right Ovary Width, Right OvaryDepth, Left Ovary Length, Left Ovary Width, Left Ovary Depth. Flat Diameter of Right UterineHorn, Flat Diameterof Left UterineHorn, MammaryGland Length, MammaryGland Width. MammaryGland Depth all P's ! 0.05!. There were 2 significantdifferences among females from the Pacificand AtlanticOceans: Body Weight t = 2.73,df = 14, P = 0.016:Atlantic mean = 100.8kg, Pacificmean = 68.2kg! and BodyLength t = 2.63,df = 22 P = 0.015:Atlantic mean= 217.4cm. Pacific194.1 cm!. However,female specimens were not equally representedbetween these two oceans,only 5 of the27 specimenswere from the Atlantic. Themajority of theyounger and smaller individualssampled were 6om the Pacific which likely reflectsthese significant differencesversus true populationdifferences between the two oceans Appendix2!. Overall,these results indicate that males and females Rom the Atlantic and Pacific Oceansare very similar for most characteristicsmeasured in tlat study. Consideringthe relativelysmall samplesizes available for S. bredanensis,it was reasonableto assumethat thesetwo locations can be combined into one populationfor furtheranalyses to obtainbasic life-history information on this species. Age-lengthcurves were generated and the attainment of physicalmaturity in this specieswas characterized according to age,total bodylength and weight. Gompertz 56 growthcurves were fitted to age-lengthdata. The general Gompertz equation applied to .S'.hredunensis is: S = A exp -bexp!-kt!!! The Gompertzequation applied follows that previouslyused for stripeddolphins Calzadaet al., 1997!where S representsthe sizeof the animalexpressed as total body length.A is the asymptoticvalue, b is the integrationconstant, k is the growthrate constant,and t is theage. Secondorder regression analyses were performed for female lengthversus age. male length versus age. and a combinationof sexesfor lengthversus age. Weight-lengthcurves were generated for bothmale and female5'. hredanensis specimensand second order regression analyses performed for males,females and a combinationof the sexesweight versus length. Theage, total body length.and weight rangethat male and female X hredanensisattain sexual maturity was determined. Gonad weightsand dimensionswere analyzedaccording to differentreproductive states characterizedin this study. Linearregression analyses were performed to examinethe relationshipsamong morphologicaltraits and reproductiveorgan traits. Log transformationswere performedon somemorphological traits to obtaina linear relationshipbetween some of the morphologicaland reproductivecharacteristics examinedin bothmales and females. Meansand ranges from histologicalmeasurements were used to describe characteristicsof testicular, uterine, and mammary tissue for differentreproductive states. All dataobtained &om individuals examined at thetime of necropsyand gross assessment of reproductive condition is presentedwith histological findings Appendices1 and2!. 57 Sexualdimorphism of weightand lengthand gonadaltraits for immatureand mature individualswere analyzedusing t-tests not assumingequal variancesfor the different traits in the age/sexclasses. A t-test ~m also performedto determineif significantdifferences in seminiferoustubule diameters between immature and mature maleswas evident from the histology results. Regres-ion and correlational analyses were performedto examinethe relationshipsamong age and morphological traits. including body size, weight, testis weight and length, and ovary weight and length. Log transformationswere performedon somemorphological traits to obtain a log-linear relationshipfor statisticalanalysis. Aged individualsdenoted with a ++ sign were excluded &om data analyses.but are listed in Appendices1 and2. RESULTS Mules: Age-Length Our malespecimens were determined to bebetween 1 and15 years of agebut it is possiblethat theserepresent underestimates as some individuals could not beconfidently agedafter a certainnumber of GLG's Appendices1 and 2!. Fitting of a Gompertz growth curve to the male length-agedata yielded the following: S = 231 exp -0.5671exp -0.3858t!!! The asymptoticvalue for this equationwas established as the largestphysically immaturespecimen total body length,this data point representsthe upperlimit of the rangebetween smallest mature and largest immature. The developmentof this formula for maleS. bredanensisrequired a bodylength value at time 0, or birth length. As this is not known for this species,an estimate of 100cm wasused. The largestrecorded fetus of 58 this speciesis 93 cm, and the smallestanimal agedat kss that 1 yearis 126 cm, and based on these available data. 100 cm was chosen as a reasonable value for equation development. The relationshipbetween male age and male total body length was stronglypositive and had high significance F = 10.7.df = 2.11.P <0.003. R = 66.0'/o!. Male asymptoticlength appears to be reachedat approximately230 cm and 13 GLG's Figure 1!. Our data on vertebral epiphysesfusion in 15 individuals indicates attainment of physicalmaturity over a wideage range fur males. Our smallestmature male was aged as 8+ years,and largestimmature male aged at 14 Table 1!. However,attainment of physicalmaturity spanned a narrowerrange when assessed according to total bodylength and weightas opposedto age. Malesreach physical maturity at a total body lengthof 227 to 231 cm and a total body weight between 130and 137 kg Table 1!. This range in total bodylength at physicalmaturity attainment is consistentwith thegraphic illustration of asymptoticlength Figure 1!. Mules: H'eight-Length The relationship between total body weight and total body length in S. bredanensiswas examined&om 14 males Figure 2!. In our male specimens,there is a lack of samplesIrom very young animals, where no weightdata was recordedat the time of necropsy. However, we still found a highly significant and positive relationship betweenmale total body length and weight F = 58.7,df = 2,12, P < 0.001, R = 90.7'/o!. 59 Mules: Scrsul Maturity Sexualmaturity was assessedby grossexamination of reproductiveorgans and statusconfirmed by histologicalsections of reproductivetissue whenever possible. Gross examinationof testeswas usedto scoremale individuals as sexually matureor immature and our male sampleset was equallyrepresentative of immatureand matureanimals. MaleX bredanensisattain sexualmaturity at a total body lengthof 216cm, and there wasno overlapin our samplewhen comparing length to reproductivestatus Table 2!. Overlapwas apparentwhen comparing estimated age and weightof malesagainst reproductivestatus. Males of thisspecies reach sexual maturity between approximately 5 and10 years of age,our youngest mature specimen aged at 5 years Table 2!. Totalbody weightof our lightestmature specimen was 92 kg and 102kg for our heaviestimmature animal Table 2!. Attainment of sexual maturity in males was investigated by comparing testis dimensions with total body length of individuals. From our 22 male specimens examined,right testis length and weight is provided for 20 and 18 specimens, respectively.Overall, there was a significantlypositive relationship between the logof right testisweight and log of right testislength F = 526.4,df = 1,17,P 0.001,R = 97.1/o!. Therewas also a significantlypositive relationship between male total body lengthand both the log of maletestis weight F = 66.9,df = 1, 17,P 0.001,R = 80.7'/o! andthe log of maletestis length F = 56.1,df = 1, 18, P 0.001,R = 75.7/o!. When comparingtotal body length with malegonad data, there is a distinctseparation between sexuallymature and immatureanimals Figure 3!. Male S bredanensisreach sexual 60 maturitywhen testis length increases to approximately 20 to 25 cmat a weightof approximately400g Figure3!. Allof ourindividuals investigated that had smaller testis lengthand weight were classified as sexually immature and those with greater values mature. More detaileddata on reproductivestatus and gonad parameters are providedin Appendixl. Histologicalexamination of malespecimens included Bergs stain of testicular materialand measurementsof seminiferoustubule diameter for 15 and 14 specimens. respectively.Males were scored as sexuallymature or immaturebased on gross examinationof reproductiveorgans, and histological results generally agreed with these findings. However,for two male specimens,Berg's stain was positivebut these individualswere still classifiedas sexuallyimmature. Measurementsof seminiferous tubulediameters in immature5'. bredanensisranged between 35.4 and 167 pm, and all measuredless than 72 pmexcept for one of thefive specimens. The average seminiferous tubulediameter was 163.2pm in maturemales and 77.1 pm in immaturemales t = 3.22. df = 12, P = 0.018!. Therange of tubulediameter in maturespecimens varied between 82.9 and209.4 pm, withno tubulediameter less then 82.9 pm for maturespecimens. Comparisonof Berg's stain and seminiferous tubule diameter results were consistent exceptin twoof ourexamined specimens. One specimen scored as immature on gross examination 04491! hada negativeBergs stain but average tubule diameter of 167pm. Anotherspecimen 04495! classified as mature on gross examination stained positive forsperm presence but only had an average tubule diameter of 82.9pm Females: Age-Lettgth Our 17 femalespecimens were agedbetween 1 and18 years. and similar to male ages,this may underestimate age of someindividuals Appendix 2!. Ourmale specimens were determined to be between 1 and 15 years of age but it is possible that these representunderestimates assome individuals could not be confidently aged after a certain numberof GLG's Appendices1 and 2!. Fitting of a Gompertzgrowth curve to the femalelength-age data yielded the following: S = 221 exp -0.5475exp -0.31923t!!! As with the male data. the female asymptotic value for this equation was establishedas the total body length of the largestphysically immature specimen. This datapoint representsthe upperlimit of the rangebetween the smallest mature and largest immature animal. The development of this formula for female S hredanensisalso requireda body lengthvalue at time 0, or birth length. As this is not known for this species,a reasonableestimate of 100 cm was usedfor femaleequation development as well. There was a highly significant and positive relationship between female body lengthand female age F = 18.9,df = 2,12,P <0.001, R = 74.4'/o!.Our samplesuggests that female S. bredanensis attain asymptotic length at approximately 217 cm, and between8 and 12 yearsof age. Data on fusion of vertebral epiphyseswas available from 13 female specimens. Like males, physical maturity attainment in females indicated a wide range when consideredaccording to age,the largest immaturefemale was estimated at 4 yearsof age, and the smallest matureaged at 13+ years of age. A narrower size rangewas apparentin 62 femalesas well. where physical maturity is reachedbetween 217 and 221 cm and between 101 and 121 kg Table 1!. Females:H'eight-Length For females. the weight-length curve indicates a statistically significant relationshipin this species body lengthand body weight regression: F = 625.0.df = 2.15. P < 0.001, R = 98.8'/o!. Femalesappear to reacha maximumweight of approximately 120kg at their asymptoticlength of 217 cm. Females: Sexual maturity Ow femalesample indicated that sexual maturity is attainedbetween a total body lengthof 212and 217 cm andat a bodyweight ranging between 101 and 108kg basedon our largestand longestimmature specimen and smallestand shortestmature animal Table 2!. Unfortunately,our sampleset lacked femaleindividuals between the agesof 4 and 9, and accordinglyour age results indicatedthe smallestmature specimen at 10years of ageand the oldest immature animal estimated at 9 years Table 2!. ln female S. bredanensisthere was a significantly positive relationship between femalebody length and right ovary weight F = 30.1, df = 1, 15, P < 0.001, R = 66.7/o! but not right ovarylength F = 2.2,df= 1, 13,P =0.16, R = 14.3'/o!.Furthermore, there wasno clearseparation between immature and mature females for eitherovary lengthor weight with total body length Figure 4!. However,our resultsare basedon ovary lengthsfor only 12 specimensand ovary weightsfor 17 individuals. From our sample size,it doesappear that ovaryweight is a betterindicator of sexualmaturity statusthan 63 ovary length. All of our femaleswith a right ovary weight greaterthan 4 g had already attainedsexual maturity. Available weights and dimensionsincluding length.width and depthof both right and leA ovariesare provided in Appendix 2. Uterine tissue slides were prepared for 6 female X bredunenvis specimens. representing several different reproductive states. Two immature females were investigated as well as two specimensclearly undergoing the process of sexual maturation. In the immatureand maturing specimens,endometrium thickness was less than that of our mature and pregnantand mature and lactating individuals investigated Table 3!. A considerablediA'erence in endometriumthickness was evident betweenthe immature and maturing animals,with an averageof 833 pm in the maturing animals. which was much closer to our mature and pregnant specimenat 853 pm. than the immatureanimals which only averaged471 pm. Endometriumthickness was greatestin our mature and lactating animal, measuring 1128 pm. There was little difference in myometrium external, vascularand internal layer thickness betweenthe immature and maturing individuals Table 3!. However,all myometrium measurementswere higher in our matureand pregnant animaL and even greater in our matureand lactatingindividual Table 3!. Exanunationof mammarytissue slides prepared for 8 individuals indicatedclear distinctionsamong different reproductive states. Immature mammary tissue was obvious, as was a slide representativeof a matureand lactating animal. However, there were no apparentdifferences in measuredmammary tissue parametersfor this specieswhen attemptingto distinguishbetween maturing and mature animals. Measurement results are therefore presentedas a maturing and mature group when categorizing mammary analyses Table 4!. In termsof overallmammary tissue composition in S. bredanensis, glandularmammary tissue was higher in our matureand lactating individual composing 77/0 of the tissue. 48'Io of the slide in mature/maturinganimals, but only 16'/0 in our immature individuals Table 4!. Connectivetissue was the primary componentof the mammarytissue in immatureanimals. comprising 73'/0 versus 33'to in mature/maturing animalsand only 18'loin the maturelactating specimen Table 4!. Fat percentagewas highestin themature/maturing animals at 14'. andthe degree of ducts.arterioles. veins andparasites was not much different among the various reproductive states investigated. Measurements were also taken of lobules, alveoli, fat cells, and ducts in the mammar> tissue. Lobulesize was much greater in our matureand lactating individual. estimated at 1020pm andless than 300 pm in boththe matwe/maturingindividuals and the immature specimens Table 4!. Sizeof alveoli,fat cells.and duct sizes were not different among the reproductivestates represented Table 4!. Comparisonbetween males and females Sexualdimorphism is evidentin X hredanensis Table I; FigureI!, physically maturemales were not significantlyheavier t = 0.47, df = 5, P = 0.659! but were significantlylonger t = 5.86,df = 4, P = 0.004!than mature females. There was no differencein theage of the physicallymature males and females examined in this study t = 0.47, df = 5, P = 0.0659!. Physicallyimmature males were significantly older t = 2.41, df= 12,P =0.03!, heavier t= 2.45,df= 14,P = 0.03!and longer t = 2.33,df= 10,P = 0.04! than immaturefemales. 65 Sexuallymature males were heavier t = 2.91, df=19, P = 0.009! and longer t = 3.00, df = 17. P = 0.009! than sexuallymature females. However. there was no differencein the age of the maturemales and females t = 0.72, df = 6. P = 0.50!. Sexuallyimmature males were significantlyheavier t = 2.91. df = 20. P = 0.009! than immaturefemales. There was no significantdifferences in the length t = 1.81,df = 20, P = 0.085!or age t=1.45. df = 17, P=0.164!of sexuallyimmature males and females. DISCUSSION The relationshipbetween age and growth in this studyof X bredanensissupports that the most rapid growth in this speciesoccurs during the first five years of life Miyazaki and Perrin. 1994!. This wasevident for bothmales and femaleswhere some very young individualsof both sexesare represented Figure 1!. The relationship betweenlength and weight was also investigatedfrom a total of 34 strandedor live capturedspecimens in this study,which provides a moreaccurate estimate for a length- weight formula for this speciesthan the previousdata on 15 live caught animals Miyazaki and Perrin,1994!. Our studypresents the firstavailable data for determination of the weightwhere physical maturity is attainedin this species,which is considerably higherin malesthan females Table I!. Sexualdimorphism is apparentin 5'.bredanensis as malesreach a largersize than females.Although a greatersize in maleswas originally suggested by Miyazaki 980!, an asymptoticlength of 231 cm was estimatedas the samefor both sexesin Japanese waters.This Japanesepopulation consisted primarily of olderanimals between 14 and32 yearsof age Miyazaki, 1980!. A study of 34 rough-tootheddolphins from Florida strandingevents represented a greaterrange of age classes,but again many of the animals were older, the oldest animal estimated at 48 years Stolen et al. 1999!. Our study population&om both the Pacificand Atlantic oceans was comprised mainly of relatively young individuals, none of our animals were over 20 yearsof age. We did have some specimens noted with a + sign indicating that growth layer groups were not distinguishable aAer a certain age. In both of the previously investigated Steno populations&om Japan and Florida, both males and females were combined to obtainan asymptoticlength for the species. Our femaleasymptotic length was lower. ranging between 217 and 221 cm whereas our male estimate of 227 to 231 cm is very similar to the species estimate of 229 cm provided for Florida waters and 231 cm in Japan Miyazaki. 1980;Stolen et al. 1999!. However, we estimatean age range between8 and 14years where male .S'. bredanensis attain physical maturity, and between 4 and 13years of age in femaleswhich is considerablylower than the overall 16 year old estimate previouslyprovided for this species Miyazaki and Perrin,1994!. Our 4 yearold female specimencould be anomalous,but regardless,our study indicatesthat at least some individuals attain physical maturity at a younger age, and our varied results are most likely a functionof theage classes which have been the focusof thesedifferent studies. Sexualmaturity in S. bredanensishas only beenpreviously assessed in specimens &om Japanand two strandedmales, one off of Oregonand the otheroff of the coastof Washington Ferrero et al., 1994;Miyazaki, 1980: Miyazaki and Perrin, 1994!. Fromthe Japanesesample set, a 14 year old malewith a total body lengthof 225 cm and testis weightof 175g wasclassified as sexuallymature based on spermpresence Miyazaki, 1980!. However,reproductive organs were not examinedin immaturemales Miyazaki 67 1980!. Our resultsprovide the first reporton the ageand total sizerange where males attainsexual maturity as well as provideaccompanying gonad weight and lengthdata Table2; Figure3!. Themale recovered Rom Washington was presumed to be immature basedon a totalbody length of 197cm andtestis weight of 23.3g, whichagrees with our findingswhere males under 216 cm aresexually immature. The male specimen from Oregonis questionablehowever, as this animalwas aged at 14years which according to our resultswould suggest maturity, but hada totalbody lengthof 209cm Ferreroet al., 1994!which is lessthan our longestimmature specimen. Criteria employedfor determinationof sexual maturity in male cetaceansis variableamong workers, and we considered both the grosstestis size as well aspresence of spermusing Berg'sstain. Two of our animalswere classifiedas sexuallyimmature basedon grossexamination of reproductiveorgans although Berg's stain resultswere positive.As therequired density of spermnecessary to successfullyinseminate female dolphinsis unknown,the immatureclassification of thesetwo specimenswas leA unchangedby the histologicalresults. In our study,males with a testisweight greater than400 g werepresumed to besexually mature, and differences in seminiferoustubule diameter between immature and mature males obtained Rom histology can be used to accompanygross maturity assessments. Values for variousparameters where sexual maturityis attainedin malesof this specieswill be a usefulreference for caseswhere limited information is availableto assessmaturity in male S. bredanensis Appendix 1!. Female S. bredanensis are believed to reach sexual maturity in Japanesewaters around10 yearsof ageand at a totalbody lengthof 210 to 220cm unpublisheddata cited in Miyazakiand Perrin,1994!. Femalesgreater than 225cm and 17 yearsof age 68 have been assessedas maturebased on presenceof corpora Miyazaki, 1980!. Our resultsindicate that femalesreach sexual maturity between 212 and 217 cm, and our largestimmature individual was aged at 9 yearsand smallestmature at l0 yearsof age Table 2!. Our studyincluded histological examination of uterineand mammarytissue from severalindividuals as well as accompanyingdescriptive data on ovary size and shape.It appearsclear that S. bredanensisundergo a processof sexualmaturation which is evidencedboth in the size and shapechanges that occurin the ovariesas well as in changesin the uterineand mammary tissue. Many of our youngindividuals sampled werenear the age of 3 or4 years,and had already began to exhibitsome of thesechanges whereovaries were no longerflat. but folded, wrinkled. and larger in size. Additionally. when classifyingresults from our histologicalparameters, it becamenecessary to establisha maturing" categoryfor both uterineand mammarytissue. In fact. histologicalexamination of mammarytissue showed no diAerencesbetween parameters in a "maturing"female and a "mature"female that wasnot lactating Table 4!. These maturinganimals did notshow evidence of corpora,and accordingly were classified as sexuallyimmature, but thisstage was foundto be significantin X hredanensis.Our maturinganimals aged at 3 to 4 yearsand the youngest mature animal aged at 9 years wouldsuggest a sexualmaturation process lasting for upto a sixyear period. Thisis doubtful,and it is probablethat these findings instead demonstrate the dearth of samples Irom4 to 9 yearold femalesin ourstudy. This stresses the valueof youngfemales of thisspecies that may become available for futurestudies to determinehow long females undergothe processof sexualmaturation. 69 A primarygoal of our studywas to providea workingreference for scientistsand managersinterested in assessingsexual and physical maturity in the speciesX hredanensis. There is very little informationon reproductionin rough-tootheddolphins, and our determination of age, length and weight ranges of maturity attainment accompaniedby gonadmeasurements, descriptions, and histologyprovide a significant contribution to the availableknowledge for this species. Acknowledgements We would like to thank Dolphin Quest Inc. and the Sea Grant Industrial Fellowshipprogram for supportof the first authoras well as the SmithsonianInstitution Fellowshipsand Grants program. Thanks to HelenWimer at the SmithsonianInstitution for preparationof histologicalsections. Earlier versions of the manuscriptwere kindly reviewedby CauseyWhittow. 70 REFERENCES Berg,J.W. 1963.Differential staining of spermatozoain sectionsof testis.Am. J. Clin. Pathol. 23:513-515. Calzada,N.. A. Aguilar,C. Lockyer,and E. Grau.Patterns of growthand physical maturityin thewestern Mediterranean striped dolphin, .S'tenella coeruleoalba Cetacea:Odontoceti!. Chivers.S.J.. and A.C. Myrick. 1993.Comparison of ageat sexualmaturity and other reproductiveparameters for two stocksof spotteddolphin, .S'tenella attenuata!. FisheryBulletin 91:611-618. Ferrero.R.C., J. Hodder,and J. Cesarone.1994. Recent strand ings of rough-toothed dolphins,Steno bredanensis. on theOregon and Washington coasts. Marine Mammal Science10 !: 114-116. Hohn.A.A. 1985.Comparison of methodsto estimatethe average age of sexualmaturity in dolphins.Sixth Biennial Conference BioL Mar. Mamm.,Abstr. Klinowska,1991. Dolphins, Porpoises and Whalesof the World:ICI JN Red Data Book. Compiledby Klinowska,M. Gland,Switzerland. Miyazaki,N. 1980.Preliminary note on agedetermination and growth of the rough- tootheddolphin, Steno bredanensis, off thePacific coast of Japan.Reports of the InternationalWhaling Commission, No. 3, 171-179. Miyazaki,N. andW.F. Perrin.1994. Rough-toothed dolphin Steno bredanensis Lesson, 1828!In Handbookof MarineMammals Volume 5: The firstbook of dolphins. eds!Ridgway. S.H. and R.J. Harrison,Academic Press, London. UK. Myrick.A.C.. A.A. Hohn,J. Barlowand P.A. Sloan.1986. Reproductive biology of femalespotted dolphins, Stenella attenuata, Irom the EasternTropical Pacific. FisheryBulletin: VoL 84 !: 247-259. Perrin.W.F., andA.C. MyrickJr eds! 1980.Age determination of toothedwhales and sirenians.Rep lnt. Whal.Commn, Special Issue 3:viii+ 229 pgs. PerrinW.F., andS.B. Reilly. 1984.Reproductive parameters of dolphinsand small whalesof the familyDelphinidae. Rep Int. Whal.Commn, Special Issue 6: 97- 133. Stolen,M.K., W.E. McFee,and B.G. Mase.1999. Age and growth estimation of three massstrandings of rough-toothed dolphins Steno bredanensis! in the Gulf of Mexico, FL. ThirteenthBiennial ConferenceBiol. Mar. Mamm, Abstr. 72 73 Table 3. Uterinetissue measurements among various reproductive states as measured &om histologicalsections in X bredanensis. 'V Q CL Ill I 0 A C C E HJ l a5 Cl E C0 0 75 260 q 240 0 0 Q 0 220 E ~ 200 ~ 180 0 160 140 o Females dashed line! o Males dotted line! 120 10 15 20 25 30 Age GLG's! 76 260 240 E 220 O 200 Ol 180 ~ 160 O 140 lO O 120 100 line! 80 20 40 60 80 100 120 140 160 Total Body Weight kg! 77 45 1600 1400 40 1200 35 E 1000 P 30 Ol C 800 I I 25 600 e 0 I- 20 Ch 400 K 15 200 10 oP<> 160 170 180 190 200 210 220 230 240 Total Body Length cm! 78 0 ~ E ~ o Ol -3 0 0 P O! Q 4 lO 0 0 2 ~ ~ Ch O! K K 150 160 170 180 190 200 210 220 230 240 Total Body Length cm! 79 'oe e a Z.ZZZZZ I ZZ> > 0-Z>-ZO >-> ~ < C r< a Cl P C v, IPI e + I< IP t E r 92v< r< O oI v< r Jl r a e -' E t t U r< C PI ~ 0' t er v< PIvIv< P< C t C Xl C W r< I ~ ' E I I I/1 av, C C C C r tcop r' Z Z P E g c eE.- Zr ZZZZZZ ' p.Zvp.p r D' r~VP v r< e r P Vl oo ~ 4 Z >.Z> ' >Z!-r>.>.>b v >. C 0 Cl e=54 K a-o-a-a-< c E~ movie8 C 80 CI 0 CIC OCCOOO C H OCO ' ' OC CCC e t e IS W I CI CI n e P V C a' 00000000 C vi ae C a~~ C r+i a'Ce,, l,r ~ I C e ver ~ C I C ~ CIv.O ISCO I 0 CC0C CC 0 C v I rI C I CI I PX~r v.e Sr >V. m ~ I I t v, I xl 2 'p x rI, v; 0 c v I ~ I IS 5g Vl IO S rl V I V. V I I rI ~ . PI I.I r ~ t C u zzzzzzr r zzzzzzzzzz>r r r zzr zr I z z z z z z r z z i" z r z > z z z i r r z r z z z z r aa C S rI VIVICI- ' I-QISISCIvI ' ' ' vIa 0 0 g5 ~Ice I rI+r rIawr t oc rIwvlvICI cl ciao 0 C C~E vl 4 IS IS IS 0 0 0 0 0 0 rI rI rI rI V rI rI 6 rI rI 6 rI rI rI rI r4 rI rI rI IA + + 0 0+ ++atVI t + vi vi rI + CI CI t w 0 0 ~ C C="< C C 5I-- CW LLLC LLLLL + 5LLLLLLL g+LgLL QLL z~' IiC 0 eo ~ ~ ~ g~ rioo Is I eocIC 00 rI C~Cr Ovls- CP CCIIS rI I C 2 & % IS % CI I 0 rI 4 a'O'CCIOa I SKOAL<>C~ 5 v vl vI vl vI vl vI vI L vI vI vI vI vI vI C vI vI vI vI vI vI 5 8 vl vI C SI E rvv,nv r v. e CI n r'r v a v, v, e ev rr, gaoenr r v en c rr 4 Cere rr a ~ rr rr w vl ~ , m rl rr e ru ~ S C C S S C S C C rr S S C C 0 o rr e w o r 0 pv C II I I I z E0 CV I o w o Irl I N I rr t ~ I ~ ~ r Xl CI Cl Cl 2 0' a Cl ~ grvCngrg t oooooe>eo ooeo Oe oVl nnnnnou on nnon n n nnnnnnC nn nnnn CrIn n n n n n C C'.n n 4 Figure I. Total body lengthof male and femaleS. hredanensisplotted againstage GLG's!.The recession equation formale body length = 1$1 + 6.27male age 0.247 male age!; femalebody length= 16$ + 8.44 femaleage 0.302 female age! . The regressionequation for combinedmales and femalesbody length= 174 + 7.60 age 0.274 age! . Figure 2. Total body weight of male and femaleA'. bredanensis plotted againsttotal body length. The regressionequation for femalebody length= 76.1 + 2.53 female body weight 0.0110 female weight!; malebody length = 121 + 1.44 male body weight - 0.00475 male weight! . The regressionequation for combined malesand femalesbody length = 83.4 +2.20 weight 0.00836 weight squared!~. Figure 3. Right testis length cm! and weight g! plottedagainst total body length in male X bredaneravis.The regressionequation for maleright testis length =-105.316+ 0.592889male total length; log male right testis weight = -1.59273 + 0.0133394 maletotal length. Figure 4. Right ovary length crn! and weight g! plotted againsttotal body length in female8'. bredanensis.The regressionequation for femaleright ovary length= 1.40717+ 0.0119194female body length;female right ovary weight = -529757 + 0.0378120. Appendix l. Reproductiveparameters for all male5 hredanensisindividuals investigated.STD refers to averageseminiferous tubule diameters. Appendix 2. Reproductiveparameters for all femaleX bredanensisindividuals investigated. $3 CHAPTER 3 Hormone coaceatratioas during rehabilitatioa and in healthy rough- toothed dolphins, Steno bredanensis K.L. West'.S. Atkinson,C. Gaspar',C.A. Manire',H.L. Rhinehart',E. Hanahoe. J.C. Sweeney,and R. Stone' PROJECTED JOURNAL SUBMISSlON: Marine Mammal Science 01/31/03 Keywords: Endocrinology,hormone, progesteroae,testosterone, cortisol, thyroid, rehabilitation, stranding University of Hawaii, Departmentof Physiology,John A. Burns School of Medicine, 1960 East-West Road. Biomed T-608, Honolulu. Hl 96822 Alaska Sealife Center. P.O. Box 1329 Seward, AK 99664 ' DolphinQuest French Polynesia, BP 1021,Papetoai, French Polynesia 9872 Dolphin and Whale Hospital, Mote Marine Laboratoryand Aquarium, 1600 Ken ThompsonParkway, Sarasota, Florida 34236 DolphinQuest, Hilton WaikoloaVillage, 69-425Waikoloa Bch. Rd., Waikoloa,Hl 96730 DolphinQuest, 4467 Saratoga Ave., San Diego, CA 92107-2337 Dolphin Quest,P.O. Box 1068Middleburg, VA 20118 Correspondenceto: [email protected] 84 ABSTRACT Endocrinology parameterswere investigatedin the relatively unknown rough- tootheddolphin, Steno bredanensis. This study reports on cortisol.thyroid, progesterone and testosteronehormone concentrations.One hundred samples from seven different individuals housedat Mote Marine Laboratoryand Aquarium and Dolphin QuestFrench Polynesiawere analyzed for concentrationsnf cortisol and thyroid hormones,including Total and Free thyroxine, and Total and Free tri-iodothyronine. Testosterone concentrations were also determined in the four male rough-tootheddolphins and progesteroneconcentrations in the three females. Hormone concentrationswere profiled from strandedanimals during rehabilitationand over time in healthy captiveindividuals. Of five rehabilitated dolphins that were serially sampled for 62-100 days, two were releasedback to the wild, one continuedto thrive in captivity, and two died. All of these individuals initially had high cortisol concentrationsthat declined during treatment regardlessof outcome. Generally,thyroid hormoneconcentrations reflected the outcome of treatment for each stranded individual, increasing in the successfullyrehabilitated animals,and declining in those that died. Thyroid hormone concentrationsprofiled over several years in the 3 captive individuals generally declined with time. Testosterone concentrationsappeared to increaseduring treatment in the 2 successfullyrehabilitated malesas compared to the malethat died. Testosteroneconcentrations increased over a 6 year time period in the captive male. Progesteroneconcentrations were low in a juvenile and adult female. Progesteroneconcentrations reached a maximum of S.25ng/ml in the other female which may reflect pubertal status in this individual. 85 INTRODUCT1ON Steno bredanensis is wide-spread throughout most tropical and sub-tropical watersof the world. but few studieshave focusedon this speciesof cetacean.especially its biology Leatherwood arA Reeves. 1983; Miyazaki and Perrin, 1994!. The number of rough-tootheddolphins held in captiveconditions throughout history is very small comparedto bottlenosedolphins, which may be partially responsible for itspoorly known status. Cetaceanshave been known to strand throughout history, and relatively recent efforts to establishstranding networks involving professionalsand volunteershave concentratedon increasingthe survivalof strandeddolphins and whales. Hematology findings indicatethat most of theseanimals are in ill healthat the time of the event Walshet al., 1990!. Although.S. bredanensis only occasionallystrands Ellis, 1982!. there are records of both mass and individual rough-toothed dolphin strandings from severalareas. Mass strandings of X bredanensishave been reported from Florida Stolen et aL, 1999;Wells et al.. 1999!. Hawaii Mazzuca et al., 1999!, and Virginia West et al., in prep!. At leasta fewindividuals have stranded in Washingtonand Oregon Balcomb, 1980;Ferrero et al., 1994!,Japan Miyazaki, 1989!,Brazil Hetzeland Lodi, 1998!,the Galapagos Palacios, 1995; Smith and Whitehead. 1999!, Hong Kong Parsons,1998!, Thailand Chantrapornsylet al., 1996!,French Polynesia Gaspar et al., 2000! and the Indonesianarchipelago or from Senegal Cadenat, 1959; Duguy, 1975!. However,no accompanyinghematology or informationpertaining to theendocrinology of this species hasbeen previously investigated in associationwith theseevents. 86 Hormonal monitoring of captive animals provides a means to enhanceour understandingof biological and reproductive parametersin marine mammals. Serum analysisis a reliable methodthat is widely utilized for determiningcirculating hormone concentrations Hadley, 1994; St. Aubin. 2001!. Cortisol. a primary glucocorticoid secretedby the adrenalgland is o Aenconsidered as a stresshormone. Cortisol has been studied in the bottlenosedolphin St. Aubin et al.. 1996; St. Aubin. 2001!. and several other odontocetespecies Suzuki et al., 1998!. but levels have not been reported in X bredanensis. Glucocorticoidshave many functions. including involvement in immune response,reproduction, growth and development For, 1999!. However, investigations into the relationshipbetween cortisol concentrations.health and comparisonwith other hormonesin cetaceansis limited, especiallyon a speciesspecific basis St. Aubin. 2001!. Thyroid hormonesare secretedby the thyroid gland and are involved in many physiological functions Ganong, 1999!. The principal thyroid hormonesare thyro~ine T~! and tri-iodothyronine Tq!, with the free forms being active at the cellular level. Circulating concentrationsof T4 are higher than Tq, but Tq is believedto be the active form Ganong, 1999!. Total Ti decreaseswith acute illness or starvation in humans Larsen and Ingbar, 1992!, and both Tq and T~ valueshave been found to be higher in neonatesof severalpinniped species Engelhardt and Ferguson, 1980; Haulenaet al., 1998; Stokkan et al., 1995; Woldstad and Jenssen, 1999!. Consistently higher thyroid hormoneconcentrations in juvenile bottlenosedolphins versusadults is apparent Westet al., in prep.!. A wide range of Total T4 concentrationshave been reported for the bottlenosedolphin, pacific white-sided dolphin, river dolphin, pilot whale, killer whale, harbor porpoise and in beluga whales St. Aubin, 2001!. Thyroid hormone 87 concentrations Total T4, Free T4 Total T3 and Free Tq! all decline as pregnancy progresses West et al., 2001!. However, thyroid hormone trends sometimesdiffer according to the hormone of interest. St. Aubin et al. 996! determined that wild bottlenosedolphins had higher Total T4, Free T4, and Free T3 than captive animals. but this trend was oppositefor Total T3. Thyroid mechanismsare complex, involving several hormones,and a greatdeal is yet to be learnedabout thyroid function in cetaceans. Testosteroneis a primary reproductive hormone secretedby cells in the testes following puberty in males. ln cetaceans.baseline testosterone concentrations are only available for a few species.the bottlenosedolphin, spinner dolphin, and beluga whale Calle et al., 2000; Kirby, 1990!. Baseline testosteroneconcentrations in cetaceans appear to be very low in immature animals, less than 0.1 ng/ml, and range between approximately1 ng/ml and 70 ng/rnl in mature adult bottlenosedolphins Kirby. 1990!. Testosteroneconcentrations vary accordingto seasonin the bottlenosedolphin, as well as in spinner dolphins and beluga whales Calle et al., 2000; Kirby, 1990; Wells. 1984!. For example, in beluga whales mean testosteroneconcentrations obtained from greater than 20 sampleswere highest in March at 4.95 ng/ml and only 0.9 ng/ml in September Calle et al., 2000!. Spinnerdolphin testosteronevalues range between 0.1 ng/ml and 65 ng/ml. Testosteroneconcentrations in bottlenosedolphins are also known to decrease with illness, and appearto be affected by the health of individuals Kirby, 1990!. No testosterone concentrations are available for S. bredanensis. Femalereproductive hormones secreted by the ovaries include progesteroneand estrogens.Monitoring progesteroneconcentrations in serial plasmaor milk sampleshave been usedto assessovarian activity in captive bottlenose,spinner, striped and common 88 dolphins,pilot whalesand false killer whales Atkinson et al., 1999;Kirby andRidgway, 1984: Wells, 1984; West et al., 2000; Yoshioka et al., 1989!. Although baseline progesteroneconcentrations have beenestablished for six speciesof odontocetes.the rough-tootheddolphin is not among this group Kirby, 1990!. In Tursiopstruncutus, baselineprogesterone concentrations are consideredas anything below I ng/ml, ovulatory as fluctuations greater than I ng/mL and pregnancy characterized by progesteroneconcentrations sustained over 3 ng/mlfor an extendedperiod Kirby. 1990!. However.existing data does appear to indicatevariations in progesteroneconcentrations accordingto different cetaceanspecies. For example,progesterone values related to reproductivestatus in stripeddolphins and short-finned pilot whalescaught in Japanese cetaceant]sheries indicated a range of 30 to 40 ng/ml during pregnancyfor the striped dolphin,but only 3 to 9 ng/mlin the pilot whale Yoshiokaet al., 1989!. Totalestrogens have also been measured as an indicatorof reproductivecondition in bottlenosedolphins. Totalestrogen baseline values are estimated at 20 pg/mLranging between23 and90 ng/mlin pregnantdolphins Kirby, 1990!. Estradiolwidely fluctuates in dolphins,and a greaterserum sample is requiredto detectestradiol concentrations than progesteronevalues in bottlenosedolphins Yoshiokaet al., 1986;West, unpublished data!. METHODS Study Anissals Serumsamples were collected &om 7 individual rough-tootheddolphins to determine baseline hormonal values. From each of these individuals, a minimum of 6 $9 sampleswere includedto establishmeans and rangesfor this species.Four of the individualrough-toothed dolphins Dolphins F-l, F-2, F-3, and F-4! werepart of mass strandingevents occurring in Floridaduring late 1997 and 1998. Rehabilitationwas attemptedfor these stranded animals at theDolphin and WhaleHospital at MoteMarine 1 aboratoryand Aquarium MOTE! in Sarasota,Florida. During medicaltreatment and observationblood sampleswere collectedfrom each of these four individualson approximatelya weeklybasis for a 3 to 4 monthperiod. DolphinsF-I. F-2, andF-3 were adultmales and F-4 wasan adultfemale. Dolphins F-1 and F-2 werereleased back to the wild and DolphinsF-3 and F-4 died. While undercare, testesfor DolphinsF-1 and F-2 were describedas sexually matureduring ultrasoundexamination. Upon release, DolphinsF-1 hada bodyweight of 152.5kg andbody lengthof 236 cm andDolphin F-2 weighed160.4 kg with a lengthof 247 cm. Uponnecropsy, Dolphins F-3 weighed122.5 kg with a lengthof 236 cm andDolphin F-4 weighed107.2 kg with a lengthof 237.5 crn. Examinationof toothsections indicated 21+ growthlayer groups glg's! for F-3 and 18- 19glg's for F-4. All length,weight, and glg measurements indicate that Dolphins F-1. F- 2. F-3 andF-4 weresexually mature individuals West et al., in prep!. Three of the sampledanimals, Dolphins P, M, and T. were housedat Dolphin QuestFrench Polynesia DQFP! whichis a naturaloceanic environment located within the barrierreef on the islandof Moorea,French Polynesia. DolphinP was initially a youngstranded female estimated at lessthan a year of age that fullyrecovered and thrivedin captivityfor threeyears. Dolphin M was a malethat likely reached sexual maturityduring the study period based on the length and weight range of sexualmaturity attainmentfor S. bredanensis West et al., in prep.!. Dolphin T was an olderjuvenile femalewhere sexual maturity had not been attained at the time of samplecollection based onabsence of corporaupon necropsy. Serum samples were collected on an opportunistic basisfrom these three animals between the years1994 and 2000. DolphinP wasestimated at lessthan a year of age at the beginningof the samplingperiod 997! andweighed approximately 40 kg witha lengthof 178cm This juvenilefemale grew steadily throughout the studyperiod and by Novemberof 2000 DolphinP weighed82 kgand was 208 cm in length.Dolphin M weighedapproximately 81 kg witha lengthof 200 cm at thestart of sampling 994!. DolphinM increasedin weightand length to 103.5kg and232 cm by the year 2000. Dolphin T wasestimated at 91 kg witha lengthof 211cm at thestart of sampling 994! andweighed approximately 94 kg with a lengthof 222 cm at theend of samplingin 1996. SerumSamples and Analyses Serum sampleswere centrifugedto separateconstituents and the serumwas analyzedfor a panel of hormonalvalues for each of the dolphins.Direct radioimmunoassays Diagnostic Products Corp., Los Angeles,CA! wereperformed on eachsample to obtainconcentrations of cortisol, Total T4, FreeT4, Total T3, and Free T3 progesterone,estradiol and testosterone.All sampleswere analyzedin duplicate. Aliquotsof 50 ul of serumwere analyzed to obtaincortisol concentrations, between 25 and 100 ul of serumwere used for thethyroid hormones, 100 ul for progesterone,200 ul for estradiol.and 50 ul for testosterone.Serum samples were not extractedand mean nonspecificbinding was <1'/o for all assays.inter and intrassay coefficents of variation were <15 and <5~/o,respectively. Assay sensitivity for the cortisol,Total T4, FreeT4, 9l Total Ti and Free T3 assayswas 0.02 ng/ml, 0.025ng/ml, 0.1 pg/ml, 7.0 pg/ml, and 0.2 pg/ml, respectively. Sensitivity of each reproductivehormone assay was 0.02 ng/ml for progesterone,1.4 pg/ml for estradiol, and 0.04 ng/ml for testosterone.Concentrations of all hormoneswere read Rom a log-logit transformedstandard curve Rodbard, 1974!. Analysesincluded establishingoverall meansand rangesfor each of the separate hormone concentrationsassayed. Hormonal profiles were also separatedaccording to individual. and each hormone except estradiol plotted during rehabilitation of the strandedanimals. Femalesamples analyzed for estradiolconcentrations were too low to be detectableand were therefore not profiled over time. From the three individuals housedat DQFP, reproductivehormone concentrationswere profiled over a two to five year intervaL but frequency of serum sampling during this much longer period was substantially less than the dolphins at MOTE. Thyroid hormones, progesteroneand testosteroneconcentrations were profiled over time for Dolphins P, M and T. For Dolphin P, cortisol concentrationswere profiled during the first 100days of rehabilitation for comparison with the individuals at MOTE. However, thyroid hormones and progesteroneconcentrations in Dolphin P were profiled over a longer time intervaL between 1997 and 2000. Statistical Analyses Using MINITAB, a scientific statistical software package,regression analyses were performedon the concentrationof thyroid and cortisol hormonesin stranded individuals from MOTE to examine the influence of the length of rehabilitation on hormonal changes. For the 3 DQFP individuals sampledover a longer time period, 92 regressionanalyses were performed on thyroidhormone concentrations. MINITAB was usedfor ANOVA analysisof thyroid hormoneconcentrations between individuals and the Tukey's multiple comparisonstest used to determine if thyroid hormone concentrationsin thejuvenile S.hredunensis. Dolphin P. weresignificantly different than the adult individuals and two animals that likely were pubertal during part of the samplingperiod. Regressionanalyses were also performedon testosteroneand progesteroneconcentrations to examine rehabilitation efforts at MOTEand time at DQFP on reproductivehormonal changes. RESULTS Hormone Analyses: Hormone concentrations measured in X hredunensis are summarized in Table 1. Cortisolconcentrations averaged 18.6 ng/ml and ranged between 0 and130.4 ng/ml when including both strandedand healthy individuals Table I !. Thyroid hormone concentrationsalso showeda largerange of values.averaging 84.1 ng/ml for Total T4, 10.2pg/ml for FreeT~. 1071.7 pg/ml for TotalTi, and1.06 pg/ml for FreeT~ TableI!. Five of the 7 rough-tootheddolphins were initially strandedindividuals. These 5 animalswere in ill healthat the startof the samplingperiod and rehabilitationefforts attempted.In all of the strandedanimals, cortisol concentrations were higher at the beginningof thesampling period and dropped as rehabilitation continued Figure la and b!. For DolphinsF-l, F-2, F-3, F-4 and P therewas a generaldecline in cortisol concentrationsover time. This trend was statistically significant for Dolphin F-4 F = 16.6,df = 1,16, P < 0.001! and DolphinP F = 8.0, df = 1,20, P = 0.01! and nearly 93 significantfor DolphinF-1 F = 4.2,df = 1,6.P = 0.087!,Dolphin F-2 F = 4.9,df = 1,8, P = 0.057!and DolphinF-3 F = 4.8,df = 1,8. P = 0.059!. Peakcortisol concentrations reacheda high of 130.4ng/ml in one individualsoon aAer arrival at the Dolphin and Whale Hospital.but this value declinedduring treatmentand this animal ultimately survived and was released. ln all 5 animals, whether surviving or not, cortisol concentrationshad droppedto below 50 ng/ml by day 50 of rehabilitation.Cortisol concentrationson a longer time scale years! from the DQFPanimals were low for Dolphin P. M. and T. the highestvalue was lessthan 6 ngiml. The different thyroid hormone concentrations,measured as Total T~, Free T~. Total Tq or Free Ti tendedto follow the samegeneral pattern in eachof the individuals profiledduring rehabilitationfrom MOTE Figure2a andb, 3a andb, 4a and b. and Sa and b!. The overall health of each rehabilitatedanimal was closely reflected in thyroid hormoneconcentration trends. Dolphins F-1 and F-2 showedincreasing thyroid hormone concentrationsthrough the studyperiod. This trendwas statistically significant for F-1 with Total T< F = 13.0, df = 1.5, P = 0.015! and nearlysignificant for Total Tq F = 5.2, df =- 1.5, P = 0.072!; the trends for Free T~ and Free Ti were not statistically significant P's > 0.05!. For F-2 the trend was statistically significant for FreeTq F = 10.6, df = 1,8, P = 0.012! and Total Tq F = 8.6, df = 1,8, P = 0.019!; the trends for Total T~ and Free Tq were not statistically significant P's > 0.05!. Both of these individuals were successfullytreated and releasedback to the wild Figwe 2a and b, and 3a and b!. DolphinsF-3 and F-4 werealso treatedthroughout the studyperiod, but both of these animalsdied, andaccordingly a decliningtrend in thyroidhormone concentrations over the durationof samplingwas apparent Figure 4a andb, FigureSa and b!. Therewas a significantlynegative trend throughout treatment for F-3with FreeT< F = 10.6.df = I,S. P = 0.012! but the trends for Total T~, Total Tq and Free Tq were not statistically significant P's > 0.05!. For F-4 therewas a statisticallysignificant negative trend for Total T~ F = 13.8.df = 1.16. P = 0.002! and FreeT~ F = 4.6,df = 1.16. P = 0.048!; there was no statisticallysignificant trend for Total Tq or FreeTi P's > 0.05!. Thyroid hormoneconcentrations indicated a generaldropping tendencywith time in DolphinsP. M, and T Figures6a and b. 7a and b, and Sa and b!. This trend was statistically significant for Dolphin P for Free T~ F = 8.1, df = 1.16. P = 0.012!. For DolphinM thenegative trend was statistically significant for TotalT> F = 7.7,df = 1,13, P < 0.016! and FreeTi F = 5.70.df = 1.14, P = 0.032!. For Dolphin T the negativetrend was statistically significant for FreeTi F = 8.64, df = 1,4, P = 0.042!. The other trends were not statistically significant for the numberof samples-daysmeasured Figures 6, 7 and 8!. There were highly significant differences in thyroid hormonesconcentrations amongthe 7 individuals TotalT~.. F =-16.5, df= 6.80,P <0.001;Free T~.. F = 10.6,df = 6,75. P ~ 0.001: Total Ti..35.1. df = 6,83, P < 0.001; Free Tq. F = 17.6, df = 6.80, P < 0.001!. In all cases, Dolphin P had higher hormone values that were statistically significantwhen compared to DolphinsF- I, F-2,F-3, and F-4 using the Tukey'smultiple comparisontest followingthe ANOVA analysis.Dolphin P alsohad significantly higher hormone valuesthan Dolphins M and T for Total T~, Total Tq, Free Tq but not Free T~ using the Tukey's multiple comparisonstest. Testosteronevalues for male X bredanensisaveraged 1.93 ng/ml and ranged between 0.04 and 6.33 ng/ml Table I!. For Dolphins F-I and F-2 testosterone concentrationsappeared to increaseover rehabilitationand decreasefor Dolphin F-3 Figure9a!. This trendwas only statisticallysignificant for DolphinF-I F = 7.1.df = 1.5. P = 0.045!. In the rehabilitated males, testosteroneconcentrations only reacheda valuegreater than 5 ng/mlfor DolphinsF- I and F-2. Testosteroneconcentrations were profiledover a 6 yearperiod for DolphinM and increasedsignificantly F = 10.2,df = 1,31, P = 0.003!. The maximum recordedtestosterone concentration for Dolphin M was 5.79 ng/ml Figure 10!. Recordedprogesterone concentrations in X bredanenvisranged between 0 and 5.25ng/ml with an averagevalue of 0.38 ng/ml Table I!. Progesteroneconcentrations profiledduring rehabilitation for DolphinF-4 andover time for DolphinsP andT did not exhibita cleartrend Figure9b andFigure I la andb!. Valuesdid not riseabove baseline in either Dolphin F-1 or Dolphin P but fluctuatedfor DolphinT. reachinga recorded maximumof near5 ng/mlon two occasions Figure 9b and Figure I la and b!. DISCUSSION Cortisol concentrationsin rough-tootheddolphins showeda largerange of values, with concentrationsas high as 130 ng/ml which is higher than bottlenosedolphin ranges St. Aubin et al. 1996!, and much higher than averagesreported for severalodontocete speciesstudied from Japanesefisheries Suzuki et al., 1998!. In the bottlenosedolphin cortisol concentrations have been found to rise to more than three times baseline when the animalundergoes capture and handlingstress Thompsonand Geraci,1986!. Our averagevalue of 18.6ng/ml for the 89 samplesanalyzed for cortisolconcentrations in S. bredaneavisdoes not diAerentiate between animals in good versus poor health nor accordingto time spent in captivity Table 1!. A more useful baselinevalue for cortisol concentrationsin this speciesis obtained from valuesprofiled over long time periods in the 3 healthy animals at DQFP. During sampling of the healthy animals, cortisol concentrationswere generally < 1 ng/ml which agreeswith the lower values reached in rehabilitateddolphins as treatment progressedand the initially high cortisol values fell Figure la and b!. General health problems in the stranded.S. hredanensis individuals and profiled cortisol concentrations during rehabilitation are consistent with the relationshipbetween high cortisol and leukocytosisin most mammals St. Aubin. 2001!. Veterinary evaluation of the rehabilitated dolphins at the Dolphin and Whale Hospital indicatedneutrophilia and elevatedwhite blood cell counts in theseanimals upon arrival. The strandedrough-toothed dolphin housedat Dolphin Quest FrenchPolynesia was also in ill health at the time of arrival and clearly appearedto have undergonestarvation. Higher cortisol concentrations found in all of these individuals at the start ot rehabilitation is not surprising, and it is possible that cortisol does reflect a stress responsein X bredaneasis,at least in relation to eitherphysiological demands associated with ill health or acclimation to new environmental surroundings. Cortisol concentrationsat the start of rehabilitation may provide expectedvalues for this species when experiencingacute medicalstresses or starvation. However,although Dolphins F-3 and F-4 eventually died. cortisol concentrationshad stabilized about halfway through treatmentand did not rise near death Figure lb!. lt is unclear if the decline in cortisol concentrationsas treatmentprogressed in our studyanimals was related to health status, captivity acclimation or a combination of both factors. Further study to determine 97 cortisol concentrationsin healthy cetaceansimmediately after captive exposureis needed for clarification. Thyroid hormoneconcentrations obtained for S. hredanensiswere comparedwith publishedvalues for other odontocetespecies as summarizedby St. Aubin 001!. Total T< averagesare available for 7 otherspecies; Ti valueshave beenpreviously reported for only two species.Total T4 valuesamong the different specieswidely range between 15 and 192 ng/ml and FreeT~ between13.6 and 39.9 pg/ml St. Aubin. 2001!. Our findings which include both healthyand unhealthyX bredanensisindividuals are within the range of all the other species,between 26.2 and 197.7ng/ml for Total T4, but are in the lower than the end of the range available for Free T~ in other species Table I!. Total Ti and Free T3 values in X hredanensisalso exceededthe higher and lower end of the range available for cetaceans,but this is in comparisonto only 2 other species St. Aubin, 2001!. In order tor cetaceanvalues to have optimum usefulnessfrom a diagnostic perspective,the wide rangesavailable for all thyroid hormonesneed to be investigated accordingto factors likely to affect thyroid function. Many factors may affect baselinevalues as the thyroid gland is involved in a large number of physiological functions such as metabolism and growth. Even for the bottlenosedolphin, T. truncatus, which is often consideredthe model cetaceanspecies, valueshave only beenpresented as overall averagesand rangesexcept in one study by St. Aubin et al. 996!. The difficulties involved in obtaining large numbers of serum samples from cetaceansmakes distinctions between factors challenging, but averages from at least 50 samplesfrom each age/sexclass in bottlenosedolphins have shown considerabledifferences West et al., in prep.!. In T. truncatus, male and female 98 concentrationsare very similar for TotalT4, Free T4, and Total Tq, but concentrationsare significantlyhigher in bothmale and femalejuveniles compared with adults Westet al.. in prep.!. In our studyof X hredanensiswe had between81 and 89 serumsamples, dependingon the thyroidhormone, which providesa goodstarting point for establishing baselinevalues in this species.Although we only hadone young juvenile rough-toothed dolphinfor comparisonwith adultsand animals likely of pubertalstatus, higher juvenile valueswere found to bestatistically significant for X breJanensis which is similarto the bottlenosedolphin trends Westet al., in prep.!. Overallaverage Total T~concentrations in thejuvenile femaleX bredanensiswas 120ng/ml compared with an averageof 77.2 ng!mlin the adultsand pubertal animals. Findings were similar for FreeT4 and Total Ti, 14.2pg/ml vs. 9.2 pg/ml, and 1694pg/ml vs. 916 pg/ml. respectively,in the juvenile versusthe adults and pubertal animals. Our findings in healthy X hredanensis individualssampled over several years also indicated a slightdecline in thyroidhormone valuesas eachanimal aged which wasstatistically significant for FreeT4 in DolphinP. TotalT4 in DolphinM, andFree T3 in DolphinT Figures6, 7 and8!. Noneof these3 individuals were consideredadults at the start of sampling. Further studiesto establish valuesaccording to ageand sex classes in healthyindividuals of variousspecies will aid in providing narrowerthyroid hormonebaselines. Thyroidhormones in cetaceansare likely to reflectclinical conditions, but thereis only one report of this in a bottlenosedolphin wherea T4 drop was believedto be associatedwith candidiasis Greenwood and Barlow, 1979!. Determinationof cortisol concentrationsand thyroid hormonesin the harbor seal have suggestedstudies to compareclinical stateswith hormonalchanges which would beuseful in monitoringthe healthof captiveseals Riviere et al.. l 977!. Stresscertainly suppresses thyroid hormone concentrationsin the belugawhale. believedto be due to capture and handling St. Aubin and Geraci. 1988!. Low thyroid values were maintainedover time in the beluga, and reducedcaloric intake may also be a possible explanation as to why thyroid feedback mechanismsdid not re-elevateconcentrations in this study of thyroid function St. Aubin andGeraci, 1988!. Physiologicalstresses are diflicult to distinguishfrom each other. and alterednutritional state and specific clinical conditionshave similar effectson thyroid hormone values in humans Larsen and Ingbar, 1992!. Both chronic or acute illnesses and starvation change humanthyroid hormoneeconomy as indicated by declines in Ti concentrations. In humans,the responseof various thyroid hormonesis sometimes different accordingto the organsystem affected. but Ti concentrationsconsistently drop with illness Larsen and Ingbar, l 992!. Our findingsfor rehabilitatedS. hredanensisindividuals demonstrate a general relationshipbetween health status and thyroid hormoneconcentrations. The only individualwhere no significantrelationship was apparent is within the first 100days of samplingof DolphinP, whilethis strandeddolphin was rehabilitated. This is likely due to lesssamples and greater intervals between days sampled in this specificanimal. In the other 4 rehabilitateddolphins, thyroid hormonevalues did fluctuate, but increasinglyrose in those individuals that weresuccessfully treated and deemed releasable back to the wild Figure2a and b, Figure3a and b!, and decreasedin thosethat died Figure4a and b, FigureSa and b!. FreeTq valuesfollowed the sameoverall patternas the otherthyroid hormonesinvestigated during rehabilitation Figwe 2a and b, Figure3a and b, Figure4a and b, FigureSa and b!, but wasthe only thyroid hormoneof the four investigatedthat did not show a statistically significant incline or decline in at least one of the animals. Total T~ and Total Ti trends agreed with patterns in Free T~, which was statistically significant in most cases. Out of all thyroid hormonemeasurements. perhaps Free Ti is one of the better indicator".of generalhealth status for,S. bredanensis. Certainly. FreeT~ is the most commonly used indicator of the thyroid hormones in human medicine for diagnosingvarious forms of thyroid disease Fitzgerald, 2000; Ganong. 1999!. Testosterone concentrations for S. hredanensis exhibited a similar range to that available for other species such as the beluga whale and bottlenose dolphin. oAen measuringbetween 0 and 5 ng/ml Calle et al.. 2000; Kirby, 1990!. In T. truncate testosteronelevels in mature males tend to fluctuate between2 and 5 ng/ml. and rise to greater than 10 ng!ml during the breeding season Robeck et al.. 2001!. for 5 hredunensisall 4 males had values within this range but breeding seasonalitywas not investigated. The 3 rehabilitateddolphins were only under care betweenthe monthsof December and March. and testosterone levels reached maximum values in March for the animals that were released but a low in the dolphin that died Figure 9a!. ln the individual sampled opportunistically over several years. concentrationsgreater than 5 ng/ml were reached in October and November in two out of three years which may suggestseasonality trends. Regular sampling throughoutthe year for severalyears is neededto clarify potential seasonalpatterns in this species. Dolphins F- I, F-2, and F-3 were classified as mature adults accordingto both testicular size and length and weight measurementsconsistent with sexual maturity in this species West et al., in prep!. Dolphin M likely attained sexual maturity during the study period. puberty in T. truncatus is described by an increase in testosterone levels from less than 1 ng/ml to 10 101 ng/ml and the first rise to 2.79 ng/ml for DolphinM occurredin 1994. Dolphin M testosteronelevels were consistently recorded above I ng/ml towards the end of the study period Figure 10!. Although our sample size was small, findings from the 3 rehabilitated A. bredanensismales suggest that testosteroneconcentrations are suppressedwith illness in X bredaneacisas hasbeen found for T. truncatus. Dolphins F-I and F-2 were initially in ill health. Dolphin F-I presentedwith neutmphilia.elevated liver values,tachycardia and bilateral congestionof both lungs. Testosteronevalues were below 2 ng/ml during the initial treatmentof Dolphin F-I but had increasedto 5 ng/ml immediatelyprior to release. Necropsyfindings from Dolphin F-3 indicatedcardiac and intestinaldisease. and the maximum testosteroneconcentration of 3.4 ng~mlwas reachedrelatively early in the rehabilitationprocess. prior to seriousdeterioration. In X bredanensisprogesterone concentrations ranged between 0 ng/ml and 5.25 ng/mland this suggestsfollicular secretionof smallamounts of progesteronein Dolphin T Figure lib!. It is diflicult to detect ovarian cycling using only progesterone concentrationsunless samplesare collected on a frequent basis. Serum sampleswere collectedregularly during rehabilitationefforts for Dolphin F-4. Frequencyranged between3 days and one week for almost four months in this adult female but concentrations did not rise above baseline. It is unknown what effect the ill health of Dolphin F-4 may haveon reproductiveactivity, this individualwas euthanizeddue to extremepain from a non-responsiveintestinal infection. In thejuvenile female,Dolphin P, only baselineprogesterone concentrations were apparent which is in agreementwith an immatureanimal. Progesteroneconcentrations fluctuated above baselinefor Dolphin 102 T, spikingto 4.67 ng/mlin Januarywhich droppedand rose again in Juneof the same yearto 5.25ng/ml beforedropping to baselineagain. Unfortunately.samples were only opportunisticallycollected for this individual on an infrequentbasis. At the time of necropsy,no corporawere obvious. but a follicle wasobserved which may be responsible for small amounts of progesteronesecretion. It is possible that the progesterone concentrationspikes above I ng/ml areassociated with the gradualattainment of sexual maturityin S. bredanensis.The body length at the timeof deathin DolphinT waswithin the lengthrange of 212-217cm wheresexual maturity is reachedin this species West et al., in prep.!. Hormonalvalues presented for healthyand ill individualsof X hredunensiswill providea useful referencefor cliniciansand scientists. Additionally,baseline values determinedin our studycontribute towards a greaterunderstanding of the endocrinology of this relatively unknownspecies ot cetacean. ACKNOWLEDGEMENTS We would like to thankDolphin Quest lnc. and the SeaGrant IndustrialFellowship programfor financialsupport. Thanks to all staff andanimal trainers at DolphinQuest FrenchPolynesia and Mote Marine Laboratoryand Aquarium. Also thanksto Kate M'Callister and Anna Lindhjemfor their efforts with data. Earlier versionsof the manuscriptwere kindly reviewed by CauseyWhittow and Don Price. 103 REFERENCES Baluda, M.G., K.D. Duffield, and R.S. Sparkes. 1972. Cetacean hemoglobins: electrophoreticfindings in ninespecies. Comp. Biochem Physiol. 41B:647-653. Balcomb,K.C. 1980.Washington dolphin stranding. Cetus 2 !:8. Calle.P., S. Monfort, L J. Dunn. E. Jensen,J.R. Boehm, S. Young, and T. Robeck.2000. Seasonaltestosterone secretion in malewhite whales Delphinaprerusleucas!. 31 st Annual InternationalAssociation for Aquatic Animal Medicinejoint American Association of Zoo Veterinarians Conference. Abstracts. Cadenat, J. 1959. Rapport sure les petits Cetace's ouest-alricains.Re sultatsdes recherchesinterprises sw les animausjusqu'au mois de mars 1959. Bulletin de I'InstitutFrancais d'Afrique Noire, serie A, 21 !:1367-1409. Chantrapornsyl,S., K. Adulyanukosol,and K. Kittiwattangwong.1996. Recordsof cetaceansin Thailand.Phuket Marine Biological Center. Research Bulletin. 61:39- 63. De Monte, T., and G. Pilleri. 1982. Cetacean Hematology, lll. Plasma Proteins. Investigationson Cetacea, 14:197-229. Duguy, R. 1975. Contributiona 1 etudedes mammiferesmarins de la cote nord-ouest AIrique.Rev. Trav. Inst.Peches marit. 39 !:321-332. Ellis, R. 1982.Dolphins and Porpoises.Alfred A. Knopflnc., New York. 270 pp. Engelhardt,F.R., and J.M. Ferguson.1980. Adaptivechanges in harp seals.Phoca groenlandicus,and gray seals,IIalichoerus grypus, during the post-natalperiod. Gen.Comp. Endocrino1. 40:434-445. Ferrero,R.C., J. Hodder,and J. Cesarone.1994. Recentstrandings of rough-toothed dolphins,Steno bredanensis,on the Oregon and Washingtoncoasts. Marine Mammal Science 10 !: 114-116. Fitzgerald,P.A. 2000. EndocrinologyIn Currentmedical diagnosis and treatment. eds! Tierney,L.M., S.J. McPhee,and M.A. Papadakis.McGraw-Hill Companies,New York. Fox, S.I. 1999.Human Physiology Sixth Edition. McGraw-Hill Companies lnc. USA. Ganong,W.F. 1999. The Thyroid Gland In Reviewof MedicalPhysiology 19'" Edition, Appletonand Lange, USA. Gaspar,C., R. Lenzi, M.L. Reddy,and J.C. Sweeney.2000. Spontaneouslactation by an adult Tursiopstruncatus in responseto a strandedSteno bredanensis calf. Marine Mammal Science 16 !: 653-658. Greenwood. A.G. and C.E. Barlow. 1979. Thyroid function in dolphins: Radioimmunoassaymeasurement of thyroid hormones.Br. Vet. J. 135:96-102. Hadley. M.E. 1994.Endocrinology. 3 Edition., PrenticeHall, New Jersey. Haulena,M., D.J. St. Aubin, and P.J. Duignan. 1998.Thyroid hormonedynamics during the nursingperiod in harbourseals, Phoca vitulina. Can.J. Zool., 76:48-55. Hetzel, B., and I . Lodi. 1998. Investigatinga massstranding ot' rough-tootheddolphins in Brazil. Whalewatcher 31 !:18-20. Kirby, V.L. 1990. Endocrinology of marine mammals.In CRC Handbook of Marine Mammals: Health, Disease, and Rehabilitation. 2" Edition. ed! Dierauf, L.A. CRC Press, Inc., Boca Raton. Florida. Klinowska, 1991.Dolphins, Porpoisesand Whalesof the World: ICUN Red Data Book. Compiledby Klinowska, M. Gland, Switzerland. Larsen, P.R. and S.H. Ingbar. 1992. The Thyroid Gland In Williams Textbook of Endocrinology 8 Edition. eds! Wilson, J.D. and D.W. Foster.. W.B. Saunders Company,Philadelphia, Pennsylvania. Leatherwood, S., R.R Reeves, W.F. Perrin, and W.E. Evans. 1982. Whales, Dolphins, and Porpoisesof the EasternNorth Pacific and adjacent arctic waters: A guide to their identification. National Marine Fisheries Service, Circular 444, v + 245 pp. Leatherwood S. and R R. Reeves. 1983. The Sierra Club Handbook on Cetaceans. Sierra Club Books, San Francisco, CA. Mazzuca,L., S. Atkinson, B. Keating, and E. Nitta. 1999. Cetaceanmass strandings in the HawaiianArchipelago, 1957-199$.Aquatic Mammals 5 :103-114. Miyazaki, N. 1989. Strandingrecords of cetaceanson the Japanesecoasts. International Marine Biological ResearchInstitute Reports.1:21-25. Miyazaki, N. and W.F. Perrin. 1994.Rough-toothed dolphin Stenobredanensis Lesson, 1828!In Handbookof Marine MammalsVolume 5: The first book of dolphins. eds! Ridgway,S.H. and R.J.Harrison, Academic Press, London, UK. Palacios,D. 1995.Cetacean remains and strandingsin the GalapagosIslands. Abstract: 1995Northeast Regional Stranding Network Conference, Riverhead, New York, 28- 30 April. Ridgway.S. andM. Reddy.1995. Residue levels of severalorganochlorides in Tursiops truncatusmilk collectedat variedstages of lactation.Marine PollutionBulletin 30:609-614. Riviere,J.E., F.R., Engelhardt, and J. Solomon.1977. The relationship of thyroxineand cortisolto the moult of the harborseal Phoca vitulina.Gen. Comp. Endocrinol. 31:398-401. Robeck.T.R.. S. Atkinson.and F. Brooke.2001. ReproductionIn CRC Handbookof Marine Mammals:Health, Disease,and Rehabilitation,2 Edition. eds! Dierauf, L.A. and F.M.D. Gulland. CRC Press, Inc., Boca Raton. Florida. Rodbard, D. 1974. Statistical quality control and routing data processingfor radioimmunoassaysand immunoradiometricassays. Clin. Chem.20:1255-1270. St. Aubin,D.J., andJ.R. Geraci. 1988.Capture and handling stress supresses circultating levels of thyroxineand triiodothyroninein belugawhales, Delinapterus leucas. Physiol.Zool. 61:170-175. St. Aubin,D.J., S.H. Ridgway,R.S. Wells, and H. Rhinehart.1996. Dolphin thyroid and adrenal hormones:Circulating levels in wild and semi-domesticatedTursiops truncatus,and influenceof sex,age, and season. Marine Mammal Science., 12:1-13. St.Aubin, D.J. 2001. EndocrinoloJg InCRC Handbook otMarine Mammals: Health, Disease,and Rehabilitation,2 Edition. eds! Dierauf.L.A. and F.M.D. Gulland. CRC Press, Inc., Boca Raton, Florida. Stokkan,K A., M.K Vaughan,R.J. Reiter,L.P. Folkow,P.E. Martensson,G. Sager,C. Lydersen,and A.S. Blix. 1995.Pineal and thyroid functions in newbornseals. Gen. Comp.Endocrinol. 98:321-331. Stolen,M.K, W.E. McFee,and B.G. Mase. 1999.Age and growthestimation of three massstrandings of rough-tootheddolphins Steno bredanensis! in the Gulf of Mexico, FL. ThirteenthBiennial ConferenceBiol. Mar. Mamm., Abstr. Suzuki,M., T Tobayama,E. Katsuma,Y. Fuijise,M. Yoshioka,and K. Aida. 1998. Serumcortiosl levels in 14 cetaceanspecies. Thirteenth Biennial Conference Biol. Mar. Mamm., Abstr. Thompson,C.A., andJ.R. Geraci.1986. Cortisol, aldosterone, and leucocytesin the stressresponse of bottlenosedolphins, Tursiops truncatus. Can. J. Fish.Aquat. Sci. 43: 1010-1016. 106 Walsh. M. T.. D.K. Odell. G. Young. E.D., Asper.and G. Bossart.1990. Mass strandings of cetaceans In CRC Handbook of Marine Mammals: Health, Disease, and Rehabilitation. ed! Dierauf,I..A., CRC Press,Inc.. Boca Raton.Florida. Wells. R.S. 1984.Reproductive behavior and hormonal correlatesin Hawaiian spinner dolphins,,Stenellalongirostris. In Reproductionin Whales.Dolphins and Porpoises. eds! Perrin, W.F.. R.L. Borwnell, Jr., and D.P. Demasters.International Whaling Commission. Cambridge, UK. Wells, R., C. Manire, H. Rhinehart, D. Smith, A. Westgate,and F. Townsend. 1999. Rangingpatterns of rehabilitatedrough-toothed dolphins,,S'teno bredanensis. released in the northeastern Gulf of Mexico. 13 Biennial Conference of the Biology of Marine Mammals. Abstracts. West. K.L., S. Atkinson, M. Carmichael. J.C. Sweeney, B. Krames, and J. Krames. 2000. Concentrationsof progesteronein milk from bottlenosedolphins during different reproductivestates. General and ComparativeFndocrinology: 17! 218-224. West. K.L., S. Atkinson, C. Schwetz, J.C. Sweeney,and R. Stone. 2001. Thyroid hormone concentrations during different reproductive states in adult female bottlenosedolphins Tursiops truncate!. 32 nd Annual International Association for Aquatic Animal MedicineConference. Abstracts. West, K.I., J.M. Mead, and C. Potter. 2002. Reproductionin the rough-tootheddolphin: Stenobredanensis: Attainment of sexualmaturity. In Preparation. West, K.L.. S. Atkinson, E. Hanahoe,C. Schwetz,J.C. Sweeney,and R. Stone. 2002. Thyroid hormone baselinevalues in the bottlenosedolphin, changesaccording to age/sexclasses and reproductivestate. In Preparation. Woldstad.S., and B.M. Jenssen.1999. Thryoid hormonesin grey sealpups flulichoerus grypus!. Comp. Biochem.Physio k A 122:157-162. Yoshioka, M., E. Hori, T. Tobayama,K. Aida, and I. Hanyu. 1986.Annual changesin serumreproductive hormone levels in captive femalebottlenosed dolphins. Bull. Jpn. Soc.Sci. Fish. 52:474-478. Yoshioka, M., K. Aida, and I. Hanyu. 1989. Correlation of serum progesteronelevels with reproductivestatus in femalestriped dolphins and short-finnedpilot whales. Nippon SuisanGakkaishi, 55:474-478. I07 Table l. CortisoL Total T4, FreeT4, Total T3, and FreeT3, progesteroneand testosterone in serumsamples collected from sevenSteno hredanensis individuals. 108 140 -4 Ootphin F-1 o Ootphin F-2 OotphinP 120 -,' 1 -' F M ~ l ! / 0 0 Rehabilitation Oay 140 +- Oolphin F-3 ~ Oolphin F% 120 100 ~ E F e eoj I 7 / 0 0 10 20 30 40 50 80 70 80 Rehabilitation Oay l09 Total T, Free T, 175 150 i' 1 100 e -4 e 5 25 0 10 20 30 40 SO 60 70 80 90 Rehabilitation Day TotalT, Free T, 2.0 i 1 2000 E 1.S-I 1500g g I- 10 i 1000 o.s-I 00+ ~~ 0 10 20 30 40 50 60 70 80 90 Rehabilitation Day 110 25 Total Te Fno Tc f 15» 100 ~ I ! 8 75 ~ 50 5 w 0 0 0 10 20 30 40 50 60 70 80 90 100 RehabititabonOaf 35 Total T~ Ftee T! 30" 25 L 1500 f R ~ g. 10-ty 71 0.5 " 004 0 0 10 20 30 40 50 M 70 M 90 100 RehabltttationDay 25 2tm Total T, i Free T, 1 25 g 15 C CS CS CL - 1CO e e I-0 0 1O- 75 pO 5 25 0 0 0 10 20 30 40 50 60 70 RehabeitationOay 35 Total T, Free T, 30- 2.5 15OO$ 2.0 8 I- g 1.5 -' 1 18$ 0 o 10 05 00 r -- <-0 0 10 20 30 40 50 60 70 Rehabilitabon Oay Tel T, . FreeT, ' 175 125 15 E a 100 I b ! 10 75 5 0 0 10 20 30 40 50 80 70 80 RehsbfSsbonDsy 35 2500 TceslT! Free T~', 30 2.5 1500 E 20 ~ t ee1.5 4. .-lac 5 1.0 q - 500 05 I 00 0 0 10 20 30 40 50 80 70 80 RohabiQs8onOsy 113 TotalT, Free T, ~ 175 150 l 100 a Q- ~, ~ O 75 SPA ~S Date 35 TotalT, Free T3 30 25 g 20. g 15 e - 1000~5 10 ' 500 05 ~ 00 -~ 25 2B! TattttT, FreeT, 150 125 E it a 100 10 75 5 35 2500 TrrtetT, Free 30- 'r 5 $ 20 15CN g 15 1app 15 I 10 4 05 00 =4- p 10 ~ Ootphn F-1 0 Ootpten F-2 DolphinF-3 p 6 C 0 e 0 4- N I 5y 0 4.... W 4 a 0 10 20 30 40 50 60 70 80 90 100 RehabditationDay 10 0 Ootphet F-4 8 a C 0 0ta G. 0 ~~~~~ 0 10 20 30 40 50 80 70 RehabilitationOay t0 8 E e 2 e I 4 e ~ N SP'ÃH ~ fP. %~8 Date 118 10 E g e' 2 o ~ o i- ~=as-oo-m e 8 Pf gP'P' 10 S- E s IL 2 t / 0 i00=-=- fà e 119 Figure I: Cortisol concentrationsprofiled againsttime in rehabilitation for five stranded ,S'.hredanensis individuals. Animals which survived a! and those that died b! following rehabilitationefforts. The regressionequation for Dolphin F- I cortisol = 33871.6- 0.944358date: Dolphin F-2 cortisol = 14595.6- 0.406131 date; Dolphin P cortisol= 318.066- 0.0086683date; Dolphin F-3 cortisol = 22725.8- 0.627146date: Dolphin F-4 cortisol = 33739.3- 0.931488date. Figure 2: Thyroid hormoneconcentrations profiled againsttime in rehabilitationin Dolphin F-l, releasedback to the wild 6om Mote Marine Laboratoryand Aquarium. Thyroxineconcentrations against time a! and tri-iodothyronine concentrationsagainst time b!. The regressionequation for Dolphin F-1 Total T4 = 52.2365 + 0.167031 date; Free T4 = 7.10583 + 0.0382268 date: Total T3 = 514.319 + 2.44515 date; Free T> = 0.646245 + 0.0042829 date. Figure 3: Thyroid hormoneconcentrations profiled againsttime in rehabilitation in Dolphin F-2, releasedback to the wild from Mote Marine Laboratoryand Aquarium Thyroxineconcentrations against time a! and tri-iodothyronine concentrationsagainst time b!. The regressionequation for Dolphin F-2 Total T4 = 74.8109 + 0.0749809 date; Free T4 = 8.57937 + 0.0267552 date; Total Ti = 691.561 + 2 45624 date; Free Ti = 0 489810 + 0.0051158 date. Figure 4: Thyroid hormoneconcentrations profiled againsttime in rehabilitationin Dolphin F-3. Thisanimal later died at Mote Marine Laboratoryand Aquarium. Thyroxine concentrationsagainst time a! and tri-iodothyronineconcentrations againsttime b!. The regressionequation for Dolphin F-3 Total T4 = 62.2815- 0.451051 date; FreeT4 = 9.22222 - 0.105741 date; Total Ti = 809.320 - 1.20067 date; Free Ti = 0.811967 - 0.0071656 date. Figure 5: Thyroid hormoneconcentrations profiled againsttime in rehabilitationin Dolphin F-4. This animal later died at Mote Marine Laboratoryand Aquarium. Thyroxine concentrationsagainst time a! and tri-iodothyronineconcentrations againsttime b!. The regressionequation for Dolphin F-4 Total T4 = 106.247- 0.836680 date; Free T4 = 10.5310- 0.0772159 date; Total T> = 1001.09- 1.89330 date; Free Ti = 1.30326 - 0.0050205 date. Figure6: Thyroidhormone concentrations profiled against time for DolphinP. This dolphinthrived in captivityat DolphinQuest French Polynesia beyond initial rehabilitationefforts. Thyroxine concentrations against time a! andtri- iodothyronineconcentrations against time b!. Theregression equation for DolphinP TotalT4 = 135.1980- 0.728710482date; Free T4 = 17.8965- 0.1681965959date; Total Ti = 1446.8482- 9.13187date; Free Ti = 1.5360 0.009083 date. Figure7: Thyroidhormone concentrations profiled against time for DolphinM. This captivedolphin was sampled for thyroid hormone concentrations for2.5 years at DolphinQuest French Polynesia. Thyroxine concentrations against time a! 120 and tri-iodothyronineconcentrations against time b!. The regressionequation for DolphinM TotalT~ = 95.8264- 0.9267date; Free T4 = 12.8212- 0.1183710534 date; Total T3 = 995.0345 + 3.4012401694 date; Free T> = 1.05498 0.0118243728 date. Figure8: Thyroidhormone concentrations profiled against time for DolphinT. This captivedolphin was sampled for thyroidhormone concentrations for 2 yearsat DolphinQuest French Polynesia. Thyroxine concentrations against time a! and tri-iodothyronineconcentrations against time b!. Theregression equation for DolphinT TotalT4 = 98.0426- 1.2615date; Free T~ = 12.5866- 0.1249145461 date; Total Ti = 713.2837 4.8803646031 date; Free Ti = 1.1858 0.0518837828 date. Figure9: Testosteroneand progesterone concentrations protiled against time in rehabilitation for 4 strandeddolphins at Mote Marine Laboratoryand Aquarium. DolphinsF-1 and F-2 survivedand were eventually released and F-3 and F-4 died. Testosteroneconcentrations against time a! and progesterone concentrationsagainst time b!. The regressionequation for Dolphin F-1 Testosterone= 0.4505 + 0.0386774749date; Dolphin F-2 Testosterone= 1.7246 - 0.0400074405 date; Dolphin F-3 Testosterone = 1.4806- 0.0148206048date; Dolphin F-4 Progesterone= 0.1203 0.0016844240483 date. Figure 10: Testosteroneconcentrations profiled againsttime for Dolphin M. This captive dolphin was sampledfor testosteroneconcentrations for 6 yearsat Dolphin Quest FrenchPolynesia. The regressionequation for Dolphin M Testosterone= 0.2708 + 0.0010516426106 date. Figure 11: Progesteroneconcentrations profiled againsttime for Dolphins P andT. Thesecaptive dolphins were sampledfor progesteroneconcentrations for 2 yearsfor DolphinP a! and2.5 years for DolphinT b!. Theregression equationfor DolphinP Progesterone= 0.12219 0.000079344973259date; DolphinT Progesterone= 0.1203 0.0016844240483date. SYNTHESlS The overall goal of this studyon X bredanensiswas to provide scientific data that will directly contribute to quantification of characteristicsrepresentative of ecologically healthyand unhealthypopulations of this species.Establishing basic information on a relatively unknown dolphin and advancing the current state of knowledge for .S. hredanensiswill lead to assessmentmeans and ultimately proactive measuresfor addressingenvironmental impacts to this cetacean.This study included research into live wild populations,dead specimens,and captive individuals. Current knowledgeon distributional patterns and status of free-ranging populations of X hredanensis is comparedwith information available for other relatively well known speciesof small dolphins. The adaptationalresponse of delphinid populationsto exploitation pressures and environmental stressors is illustrated with examples of distributional and reproductiveparameter shiAs that havebeen demonstrated in otherspecies. A discussion of populationlevel effects is followed by an analysisof endocrinologydata from individuals of the species 3'. hredanensis. The individual rough-toothed dolphins investigatedinclude both healthy animals and those experiencingphysiological stress. providingan opportunityto obtaincomparative data. Establishingbaseline biological and ecologicalparameters for X hredanensison a populationand individual scale contributeto neededspecies knowledge for this relatively unknowncetacean. Adaptationof 8"ddPopulations The first responsephase which includes alarm and adaptation is a stage where definition of speciesspecific characteristicsto stressorsis extremely valuable. Stressors 122 do not causethe samedegree of stressamong cetacean species St. Aubin, 2001!. If it is possibleto detectsuspected stressors early in the overall biological responsepattern of a species,the likelihood of implementingpreventative measiues to avoid the final stageof this destructiveprocess increases. However,at the time of the actual effect of a given stressoron individuals, direct application of techniquesfor assessingthe initial alarm responsepresents difficulties. Cardiac tissue contraction band necrosis and ischemic injury to organs in many speciesof cetaceansuggests initial alarm but histological analysesonly becomepossible once theseanimals have already died and carcassesare subsequentlyrecovered Cowan, 2000!. Additionally. histological investigationinto the alarm responsehas not yet been undertaken for the species X hredanensis. The adaptationresponse which is also apparentduring this first stagemay provide an easier meansto assesspopulation responsesin live, wild individuals. Once patternsand normal behaviorsfor the speciesX hredanensisare determined,populations can he monitored and comparativeresearch techniques used to detectchanges in theseestablished patterns. It is probablethat changesin the distributional pattern of cetaceanpopulations in responseto environmentalstressors commonly occurs, but is not documentedat the time of the event. It is rare to find available baseline data prior to impact effects. Additionally, it is difficult to prove causeand effect when environmentalchanges are suspectedas the causativeagent resulting in populationpattern shifts. A combinationof these factors likely manifests in undetectedstress to cetaceanpopulations during early stagesof the biological stressresponse, in which case the problem does not become apparentuntil it hasreached a moreadvanced and severestage. It is likely that the adaptationresponse of wild populationsof S. bredanensisto environmental stressorswould be reflected in changesof habitatuse. As cetaceansspend their entire lives in the marine environment and the sourceof known potential stressors all involve the immediatesurroundings of these animals,a shift in habitat preference should be detectableon a population level. Defined distributional patterns for S. bredanensisprovide a starting point for initial assessmentsof the population adaptation responseto environmentalchanges. This researchon X bredanensisdescribes preferred habitat characteristicsand distributional patternsfor this type of dolphin in a small oceanic island environment. This providesa significantcontribution to availablespecies knowledge since establishing distributional patternsfor X bredanensisis challengingas live, wild populationsof this dolphin tendto be somewhatunaccessible for study. Theconsistent presence of rough- tootheddolphins in Polynesianwaters was first identifiedby Leatherwoodet al. 982!. The relatively frequentsighting rate of S. bredanensisin FrenchPolynesia provided a uniqueopportunity for wild populationstudy as in thesewaters this speciesis rankedas the secondmost commonly sighted cetacean Gannier, 2000!. The ability to generatea distributional data set for S. bredanensisin French Polynesiais advantageousto wildlife managersin this areaand will additionally providequantitative baselines for other oceanic regionswith similar oceanographicconditions. A descriptionof preferredhabitat characteristics is neededfor any givencetacean species.This informationcan then be usedin comparativeanalyses to assesschanges in naturalhabitat use that mayoccur due to factorssuch as habitatdegredation, increased humanencroachment, or pollution. Certainly,cetaceans can be negativelyimpacted by l24 their surroundingsand adapt to environmentaldisturbances. For example.cetaceans are likely to flee &om areaswhen alarmed.noise pollution from oil rigs has been shown to havethis effect on belugawhales, and tuna fishery operationsalter movementsof spinner dolphins Norris and DohL 1980;Thomas et al.. 1990!. A significant number of studies have focused on defining horne ranges and normal activity patternsfor other speciesof small dolphinsthat are relatively well known such as the bottlenoseand spinnerdolphin from many different regions Ballance, 1990; Corkeron, 1990;Hansen. 1990; Kenney, 1990:Klinowska, 1991;Norris and Dohl, 1979; Pryor and Norris, 1991: Reynoldset al., 2000; Saaymanand Tayler, 1973: Scott et al., 1990; Shane, 1980; Shane, 1990; van Waerebeek et al., 1990; Wells et al.. 1990; Wells and Scott, 1999; Wursig and Harris, 1990; Wursig et al., 1994!. Coastalhabitats have beenthe focus of theseinvestigations because of researcheraccessibility and concern for habitat vulnerability as the inshore waters are especially susceptibleto anthropogenic impacts. For spinner dolphin populations using near-shorewaters as a resting area, human disturbances within the preferred habitat are believed to have resulted in distributional patternshifls as the populationadapts to the disturbance Klinowska. 1991; Perrin, 1989!. For bottlenosedolphins, a population in San Diego Bay illustrates dramatic distributional changesbelieved to be attributableto pollution effects. Until the 1960's,a bottlenosedolphin populationwas consistently present in SanDiego Bay before disappearingfrom the area for almost a 10 year period. The return of these resident animals coincided with improved water quality of the Bay, and although other factors may have been responsibleor contributed to these observed distributional changes, pollution is the likely culprit Klinowska, 1991!. Prior to the current study on S. bredanensis,no distributional or preferred habitat usage findings have been available from near-shorewaters for this dolphin species. The only detailed distributional map for X bredanensisis from the Eastern Tropical Pacific ETP! region. where survey effort covering over 25' of latitude and longitudeover severalyears in large and expensiveNOAA ships were usedto obtain this data Au and Perryman, 1985; Hewitt, 1985; Wade and Gerrodette, 1993!. hredanensiswas sightedon a relatively infrequent basis in the ETP, rankedas the ninth most likely cetaceanto be sighted at a density of only 0.0047 individuals/km. Similar large-scalesurvey efforts haveincluded study of.S'.bredanensis distributional patterns for other wide regionsof open oceansuch as the WesternTropical Indian Ocean Ballance and Pitman. 1998!. X hredanensishas been describedas a primarily pelagic species found in off-shore waters throughout tropical and subtropical oceans of the world Klinowska, 1991;Leatherwood et al., 1976;Leatherwood et al., 1982;Leatherwood and Reeves,1983; Miyazaki and Perrin, 1994!, and accordingly speciesdistributional data in off-shore environments is useful. Although not to the extent of the spinner and bottlenosedolphin, the rough-tootheddolphin has been sighed in near-shorewaters in addition to the described off-shore habitat Klinowska, 1991; Leatherwood et al., 1976; Leatherwoodet al., 1982; Leatherwoodand Reeves,1983; Miyazaki and Perrin, 1994!. S. bredanensisis also presentaround oceanic islandsthroughout its range, for example theCanary Islands have been identified as a regionwhere this speciesis frequentlyfound Ritter, 2002!. In French Polynesia,rough-toothed dolphins are most commonly found in water depthsbetween 1000 and 2000 m at distanceof 1.8 to 5.5 km from the reef barrier. l26 Resultsalso indicate the trequencyof sighting at various distancesfrom shoreand that the averagegroup size of X bredanensisis 12.1individuals. This specieswas observed in French Polynesianwaters throughout the year and no clear variations in seasonal distributionwere apparent. These tmdings provide reference values and insightinto the habitat preterenceof S. bredanensisin an oceanic island environment. However,many questionson rough-tootheddolphin ecology remain, and further investigationsof live wild populationsin this regioncould lead to a determinationof site fidelity. homeranges. social structurefluidity and reproductiveseasonality in this species. Through continued populationmonitoring of French Polynesia's rough-toothed dolphins.comparisons can bemade to theseestablished distributional patterns in orderto assesspotential threats. Fishery interactions have been identified by fisherybiologists in FrenchPolynesia. for examplevideo footagedocuments rough-toothed dolphins stealing bait at depthin excessof 200m Bach.P., pers.comm.!. lt is not knownwhich specific speciesof cephalopodsand fish that S bredanensisfeeds upon in French Polynesiabut any marineecosystem threats in this areaare likely to affect all levels of the food chain. Direct impactson rough-tootheddolphin populationsmay also result from human activities associatedwith the tourism industry in French Polynesia. Dolphin watchesare a populartourist attractionin manyregions of the world. The focusof this activity in FrenchPolynesia is currentlyon spinnerdolphins, but the preferredhabitat of rough- tootheddolphins overlaps with thatof S. longirostris.5 bredanensishas been held under captiveconditions and the removalof this speciesfor public display purposesmay present a future threat to populations. Defined distributional patterns and group characteristicsof S. bredanensisin French Polynesia will aid wildlife managersin 127 determiningif populationadaptational responses are occurring as a resultof thesety~s of potentialstressors into the future. Adaptatiottof ReproductiveParameters The adaptationof populationsto stressorsinclude biological and physiological changeson an individualscale. Severelydepleted population numbers may resultin a densitydependent response. where individual reproductiveparameters change in an attemptat compensationfor greatlyreduced population sizes. Evidenceof the density dependentresponse is reflectedin birth rate, ageof first birth, and juvenile andadult survival. This responseis basedon changesin recruitmentprocesses and the interplay betweenreproduction and survival,and hasbeen documented to occur in 16 speciesof marine mammals Fowler, 1984!. For most cetaceans,the density dependentresponse usuallymanifests in a changeof the agethat femalesgive birth to their first calf and in overall birth rates Fowler, 1984!. ln this event, physiological mechanismscome into play to increasepregnancy rate by decreasingage of sexual maturity attainment, deceasinglactation length. and decreasingthe restingperiod. The densitydependent responseof cetaceanpopulations appears to dependnot only on predationand available resources,but the social and behavioralcharacteristics of a speciesalso appearto play an importantrole Fowler, 1984!. A density dependentreproductive response to exploitationpressures has been documentedin the killer whaleand sperm whale which represent two speciesof cetacean withhighly organized social structures Fowler, 1984!. The killer whaleis consideredto be a gregariousspecies where strong social bonds and the maintenanceof complex relationshipsis paramountin group structuring Leatherwoodand Reeves,1983!. 128 Comparisonswere made betweenkiller whales found off of the West coast of Canada, and birth rate was found to be higher in pods where young animals had been taken for display as opposedto pods where no animals had been removed Biggs, 1982: Fowler, 1984!. The suggestedexplanation for this differencewas an increasein the reproductive activity of older females as a direct result of social and behavioral changes which occurred in responseto the removal of the young animals Biggs, 1992; Fowler, 1984!. Spermwhales have showeda similar reproductiveresponse to exploitation pressuresthat has been demonstrated in killer whales Best et al., 1984; Fowler et al.. 1984!. Sperm whales off of South Africa were comparedduring two different time periods where the later investigationsfollowed exploitation of the population. Similarly for sperm whales. an increasein pregnancyrate followed exploitation effects and was most evident in the older female age classes Best et al., 1984!. This increasein pregnancyrate reflects in a decreasedcalving interval. and in this casethe calculateddecline was from an averageof 6 to 5.2 years Best et al.. 1984!. Sperm whales exhibit a highly organized social structure, and provide another example of the density dependentreproductive response for a speciesof cetaceanwith a high interdependenceof school members Best et aL 1984!. Not all exploited cetaceanspecies that have been previously investigatedhave demonstratedthe density dependentreproductive response to population declines. The fin whale is an example of a specieswhere greatly reducedpopulations resulted &om extensivecommercial whaling. Reproductiveparameters in fin whaleswere describedin detail in the 1950'sas this baleenwhale becamethe focal whaling species,constituting a major portion of the world's catch after blue whale stocks weredepleted Macintosh and Wheeler. 1952; Mizroch and York, 1984!. Pregnancyrate in the fin whale has been analyzed,and a high degreeof variability was detectedamong regions, time periods.and seasons. No increase in pregnancy rate for fin whales was apparent according to exploitationeffects and if a trend is present,it is maskedby high variability Fowler. 1984; Mizroch and York, 1984!. As fin whalesare often sightedas solitary individuals or in relatively small group sizesrarely exceeding7 animals,it is possiblethat the lack of evidence to support a density dependent reproductive responseis related to social structures that differs &om that found in odontocetes such as the killer whale and sperm whale. Regardlessof the reasonwhy heavily exploited fin whale populationsdid not demonstratea densitydependent reproductive response, clearly reproductiveparameters need to be investigated on a species specific basis. The population responseto exploitation is not consistentamong all cetaceansbut needsto be determinedfor effective speciesmanagement efforts aimedat the speciesof interest. The spinnerdolphin providesan exampleof a small dolphin speciesthat inhabits similar environmentsto the rough-tootheddolphin where heavy exploitation has been documentedand responsive reproductive parametersinvestigated Leatherwood and Reeves, 1983; Klinowska, 1991!. Spinner dolphin pod sizes range a great deal, sometimesnumbering into the thousands,and this speciesexhibits fluid social groupings Klinowska, 1991; Leatherwoodand Reeves,1983!. In the ETP, life-history parameters in S. longirostrishave been documented to shift to the extentthat the ageof sexual maturity attainment in femalesis decreasedby several years Klinowska, 1991; Perrin and Henderson.1984!. Over a twenty year period, the spinnerdolphin populationin the ETP region was estimatedto have been reducedto only 20'/0 of pre-exploitation l30 abundances Klinowska, 1991!. This extreme exploitation is evident in the reproductive parametersof ETP spinnerdolphins. believed to be responsiveto the lossof huge numbersfrom this population Perrin and Henderson,1984!. Attainmentof sexual maturityin the FTP spinnerdolphins were compared with an unexploitedpopulation in the Gulf of Mexico where both malesand femalesreach sexual maturity at a longer and heavier size Perrin and Henderson, 1984!. Female spinner dolphins in the eastern tropicalpacific reach sexual maturity at an averageof 5.5 growthlayer groups GLG's! andat an averageof 7-10GLG's in the Gulf of Mexico Klinowska,1991; Perrin and Henderson.1984!. Potentialproblems with this dataset needto be takeninto account althougha comparisonof these spinner populations with varyingdegrees of exploitation do supporta densitydependent reproductive response to extremepopulation decline. Firstly. as opposed to a direct comparisonbetween exploited and unexploited populations,in this specificstudy of spinnerdolphins, both populations had experienced differentlevels of pressure.The degree of populationexploitation was estimated to have resultedin a much lower fraction of the original size ir. one population 7-25'/o! versus the other 8-72'/o!. Anotherconsideration with this data set is that the two spinner dolphin populationsinvestigated represent two different forms from different geographicallocations. There are at leastfour different forms of spinnerdolphin stocks in the easterntropical pacific, with morphologicaland size variationsbetween these stocks Gray, 1846!. However,this still providesa measureof comparisonfor a small speciesof dolphinand supportsmanifestation of the densitydependent reproductive responseevidenced by decreasedage of sexualmaturity attainment. ln anothersmall dolphinspecies, a shortercalving interval, an excessively short lactation period and short restingperiod have beensuggested as an exploitation result for stripeddolphins exploited oA of Japan Kasuya, 1984!. For thesmall species of cetaceanwhere exploitation effects have beeninvestigated. findings indicatethat reproductiveparameters are altered,which is also in agreementwith the generallysocial natureof small dolphins. When thousandsof individuals are available for reproductivestudies and the level of exploitation is known for a given dolphin species,it becomespossible to compare reproductiveparameters among populations. For the species5'. bredanensis,reproductive information is lacking and it has not previously beenknown for any region at what age and size range these animals attain sexual maturity. No estimatesof calving interval, length of lactation. length of resting period or other important reproductiveparameters are available from any region Klinowska. 1991; Leatherwood and Reeves, 1983; Miyazaki, 1994!. Reference baselines are needed to assess potential shiAs in reproductiveparameters that mayresult from futureexploitation pressures. The difficulty in obtaining a suAicient sample size to investigatereproduction in S. bredanensishas precludedextensive study of this species. The need for knowledge on reproductive biology has been emphasizedas paramountfor conservationmeasures aimed at the rough-tootheddolphin Klinowska, 1991!. Only one study has addressedreproduction in X bredanensisto date. Miyazaki 980! examined reproductive organs and aged adult specimenscaught in a Japanese drive fishery. His findingsindicate that at 14years of ageand a lengthof 225cm males are sexually mature, and females at an age of 17 years of age and length of 225 cm. However, no immature specimenswere aged, and even basic information such as what age or size range the rough-tootheddolphin attains sexual maturity is not yet known for this species. Miyazaki s work 9SO! suggestedthat S. bredanensismay undergoseveral simultaneousovulations by examinationof corpora, but there havebeen no investigations into reproductiveseasonality. reproductive rate. or any other specific studies on reproduction. Resultsfrom this study provide basic knowledgeon reproductiveparameters and characterizethe age and size range where sexual maturity is attainedin X bredanensis. Sexualdimorphism is apparentas X hredanensismales grow to a longerand heaviersize. males attain sexual maturity between 5 and 10 years of age. Femalesreach sexual maturity by at least9 yearsof age,and undergothe gradualprocess of sexual maturation which may begin as early as 3 or 4 yearsof age. Due to the diAiculty in obtaining access to a largeenough sampling of dead specimensto investigatereproduction in this species, specimenswere combinedfrom both the Atlantic and Pacific oceansto establishbaseline values. Significant differencesbetween the Pacific and Atlantic populationswere not obvious, but the combinationof sampling needsto be recognized. lt is also possiblethat the age and size range ot sexual maturity attainment characterizedby this study may representpopulations already experiencing some degree of environmentalor exploitation stress. IIowever, this is the first study to investigatesexual maturity attainment in this species.and these refereiice values serve as the only baselinevalue available. Monitoring of these establishedparameters will provide a means to aid in the detection of future changesto thecurrent status of populationsfound in bothoceans. 133 Horssonal Events The secondphase of the biologicalstress response in individualsfollowing alarm andadaptation is hormonalevents. In orderto beable to detecthormonal changes which signify that a speciesis experiencingsignificant stress, baseline endocrinology data from healthyindividual animals is first required.Originally from FrenchPolynesian waters. 3 individualrough-toothed dolphins housed in a naturallagoon environment in this same regionallowed opportunity to obtainserum samples over time. Two of theseanimals werehealthy and the third wasoriginally a strandedindividual that wasrehabilitated and thrived in captivity for threeyears. In additionto the ability to establishdistributional information from live. wild rough-tootheddolphins in French Polynesia,captive animals from this samearea were accessiblefor this study to determinebiological parametersin healthyindividuals. Although captive animals may not be entirelyrepresentative of live. &ee-rangingindividuals, this still providesa startingpoint for establishingbaselines in individuals from a speciesof cetaceanwhere so little is known. There is no indication that French Polynesia'srough-toothed dolphins currently representan ecologically unhealthypopulation. This relatively isolated island environment in the South Pacific does not have continental scale impacts from land use change,nor doesthe humanpopulation on thesesmall islandsexceed a million people like some other oceanic islandsof similar size West and van Woesik, 2001!. Human activitiesin this regionvery well maybe affectingS. bredanensispopulations, however, baseline distributional and hormonal data established &om this area are assumed to representa populationthat wasnot undersignificant stress at thetime of datacollection. l34 This assumptionis basedon the lack of evidenceto supporta high mortalityrate for S. bredanensisin FrenchPolynesia. If S. hredanensispopulations in FrenchPolynesia were underextreme stress, phase three of the biologicalstress response would beapparent and deadindividuals would be muchmore prevalent.The rough-tootheddolphin has been identified as the second most abundant cetaceanspecies to be sighted in French Polynesianwaters. yet therehave been only a fewindividual stranding reports for this area Gaspar et al.,2000!. Thisarea has not been identified as one where rough-toothed dolphinsare taken for humanconsumption like Japan. the West Indies, and the Solomon Islands Klinowska, 1991!. This supportsthe likelihoodthat 5'. bredanensispopulations in FrenchPolynesia are ecologicallyhealthy. at least in comparisonto other regions wherelive rough-tootheddolphins are eithertaken or rarely sightedand only stranded individualsare generally reported Ferrero et al., 1994;Klinowska. 1991; Mazucca at al.. To definecharacteristics representative of an ecologicallyhealthy population of this species,French Polynesia provided the ideallocation. French Polynesia may be regardedas a pristineenvironment for manycetacean species. Both live, wild population distributionalpatterns could be determinedand hormonal baselines established &om healthy captive animals in these waters. Investigationof hormonalevents in S. bredanensis,which signify the secondstage of the biologicalstress response, requires comparativehormone concentrations. Serum samples are requiredfor hormonal analyses,and collectionof blood samplesfrom Iree-ranginganimals is not possible withoutimplementation of captureprocedures. It is necessaryto takeadvantage of biologicalsampling opportunities that arisewhen cetaceans are held in a captive environment. Very few individuals of the speciesS. bredanensishave been held under captiveconditions throughout all of history.and currentlythere is only one individual housedwithin the United States. This necessitatesopportunistic sampling, and this study includedserum sampling of 4 individualsin Florida,housed at MoteMarine Laboratory andAquarium MOTE! over a 3-4 monthperiod. At the time of early sampling.the S. bredanensisindividuals at MOTE were representativeof generallyunhealthy animals that were part of massstranding events where the majority of pod membersdied. This providedan opportunityto obtain endocrinologydata &om an unhealthypopulation under considerable stress. 1-lormonaldata from animals in French Polynesia and Florida allowed for comparisonsbetween individuals believed to be representativeof both ecologicallyhealthy and unhealthy populations of X hredanenvis. Cetaceansare oRen in ill-health at the time of massstranding events Walsh et al.. 2001!. The Floridastudy animals that survived the initial massstranding events and were transportedto the Dolphin and Whale Hospital at MOTE for rehabilitationare no exception. From one of thesestranding events in Decemberof 1998a total of 12 3". bredanen~isindividuals stranded,2 died enrouteto the Dolphin and Whale Hospital and only two whichbecame study animals survived long enough to bemedically treated. The two individualswere diagnosed with elevatedwhite bloodcell countsand elevated liver valuesupon arrival at theDolphin and Whale HospitaL Although treated with antibiotics andantifungal medications, both animalseventually died and necropsyresults revealed gastrointestinaldisease, cardiac fibrosis, and bronchopneumoniain one animaland intestinal and cardiac disease in the other. The other two stranded study animals housed at MOTE initially exhibitedneutrophilia, elevated liver values,bilateral congestion of l36 bothlungs, and tachycardia. Both of theseanimals were treated throughout rehabilitation and eventuallyreleased. Early veterinarydiagnoses certainly indicatecompromised health in all MOTE animalsat the beginningof rehabilitation efforts and that biological samplesduring this timeframeare reflectiveof dolphinsin ill-health. Physiological stressesassociated with clinical state and transport and acclimation to the treatment facility was evident. Frequentserum samples collected from unhealthyX bredanensis individuals allowed for determinationof hormonal values representativeof this species while experiencingsignificant stress. Endocrinologydata from the captiverough-toothed dolphins housed in French Polynesiaand at MOTE in Floridawas separatedaccording to healthstatus for direct comparisonof values. The initially strandedindividual from FrenchPolynesia was not includedin theseanalyses. Cortisol, thyroid, progesteroneand testosteronevalues for unhealthyindividuals were defined as thoseconcentrations obtained within one month followingthe Florida massstranding events. All 4 individualswere in ill-healthduring this period of the rehabilitationprocess. Healthy values were categorizedas concentrations obtained within the last month of rehabilitation for the 2 Florida individuals that were successfullytreated and released. Healthy valuesalso included all concentrationsobtained from the 2 healthy captive rough-tootheddolphins in French Polynesia. T-tests assumingunequal variances were performedbetween these two groups,and cortisol and Free T~ concentrations were found to besignificantly different betweenhealthy and unhealthyindividuals. For cortisol,mean = 8.6 ng/ml in healthy individualsand mean= 46.2ng/ml in unhealthyindividuals t = 5.24,df = 24, P <0.05!. For FreeT4, mean = 11.2pg/ml in healthyindividuals and mean = 8.7pg/ml in unhealthy l37 individuals t = -2.89, df = 40, P .05!. Total T4, Total Tq, Free Tq, progesteroneand testosteroneconcentrations were not significantly different betweenthe two groups P s ! 0.05!. Thesedirect comparisonssupport resultsobtained by profiling cortisol and thyroid hormonal concentrationsduring rehabilitation in the stranded animals. Corti~I was initially high which likely indicatesa stress responsedue to compromisedhealth. acclimationto captivity or a combinationof thesetwo influences. Resultsfrom this study on S. bredanensissuggest that baselinecortisol concentrationsare lessthan 10 ng/ml. and valuesover 25 ng/ml likely signify an individual under stress. Thyroid hormoneresults also demonstrateda relationship with health status of individuals. and acclimation to captivitydid not appearto affectthese hormones. From the four rehabilitateddolphins, thyroid hormonesgenerally increased as medicalcondition improved in the two that were released.The other two rough-tootheddolphins that died demonstrateda declining trend in thyroid hormone concentrations as clinical state deteriorated. Free Ti values immediatelybefore death reached a low of only 1.3 pg/ml which is well outsideof the rangeavailable for otherspecies. Free T4 valuesbetween 16.6 and 39.9 pg/ml havebeen summarizedby St. Aubin 001! as the overall rangeavailable for cetaceanswhich includesthe pacificwhite-sided and bottlenose dolphin and beluga, pilot andkiller whale. This study is the first to investigatedifferences in thyroid hormoneconce.itrations accordingto healthstatus. Only immediatelyprior to deathwere extremely low levels apparentfor S. bredanensis,most samples were within the rangegiven for otherspecies of cetacean. Resultsfrom this investigationprovide cortisol and thyroid referencevalues andwill havediagnostic usefulness in assessingclinical stateand relative stress level of S. bredanensis. 13$ Hormonalconcentrations are also effected by environmentalsurroundings and the topic of endocrinedisruption has recently received a greatdeal of attention. Pollution levels are believed to be reflected in thyroid hormone values in marine mammals St. Aubin. 2001!. In the gray seal thyroid hormonesact as biomarkersand in the common seal a thyroid deficiency has been found to be associatedwith environmental contamination, specifically polychlorinated biphenyls PCB s! Brouwer et al.. 1989: Hall et al.. 1998!. Evaluationof thyroid glands from dead harbor sealsin the North Sea are thought to be morphologicallydefective due to pollution in this area Schumacheret al., 1993!. Abnormal histological findings from beluga whale thyroid glands also correlatewith high organochlorinecontamination in the St. Lawrenceestuary De Ciuise et al., 1994!. To be able to accuratelyassess the effects of environmentalstressors such as contaminates on hormonal events, baseline concentrations from animals from an ecologically viable populationare first needed. Concentrationsfrom captive X bredanensisindividuals from FrenchPolynesia provided the most feasiblealternative to multiple capturesfor obtaining the necessarybiological samplesfor establishing baselinesrelating to endocrinology. Although many factors influence hormonal concentrations,further researchis underway to establishto what degreenatural factors play a role in the thyroid hormoneconcentrations provided by this investigation St. Aubin et al., 1996; West et al., 2001!. Bottlenose dolphin adults tend to have lower thyroid hormonevalues than juveniles which is in agreementwith findingsfor neonatal sealsof severalspecies Engelhart and Ferguson,1980; Haulenaet al., 1998;Stokkan et al., 1995;West et al., 2002: Woldstadand Jenssen,1999!. The heaithy individuals which served as the control samplesfor reporting thyroid hormonesin S. bredanensiswere either all adults or appearedto have approachedsexual maturity during the sampling period. Althoughthe pubertalstatus of 2 individualsmay have slightly affectedthe baselinemeans reported, differences among all individuals sampledwas only apparentin the concentrationsof the very young juvenile when compared to the other 6 study animals. Regardless,these results do provide a starting point for assessingthe statusof potentialenvironmental stressors on hormonalchange in X bredanensis.This is the only cetaceanspecies to date where direct comparisonsin endocrinologicalparameters have beenmade according to physiologicalhealth. Testosteroneconcentrations are negatively related to blubber PCB's and DDE, which is transformedfrom DDT concentrationsin Dali's porpoise Subramanianet al., 1987!. Decreasesin testosterone concentrations were found to be statistically significant with increasinglevels of DDE in this speciesof small cetaceanfrom the North Pacific Subramanianet al., 1987!. Implicationsof alteredreproductive hormones in cetaceansis far reaching,any factor which threatensthe ability of populations to reproducecould have devastatingeffects on maintenanceof population numbers. Determinationof baselinereproductive hormone concentrationsaccording to various reproductivesaates are required to be able to accurately assessthe estrogenic effect of environmental stressorssuch as contamination. For the speciesS. bredanensis,this study provides rangesfor testosteroneand progesterone concentrations, and will allow for furtherstudies and futurecomparisons with thesenewly established baselines. Testosterone values in adult S. bredanensisindicated suppressedvalues in ill animals which was evidencedby low concentrations that increased with successful rehabilitation but remained low in adults that died. Suppressionof testosteroneconcentrations with illness has been 140 previouslydocumented in the bottlenosedolphin, as an individual close to deathand another with Lobo's diseasehad significantly lower reproductivehormone levels Kirby, 1990!. The ability to impregnatefemales is a key component in reproductive success, and in bottlenosedolphins testosterone levels profiled throughoutthe year showeda similar patternto spermproduction although spermatogenesis lagged behind by several weeks Schroederand Keller. 1989!. In the T. rruncatus individual with suppressed testosteroneduring Lobo's disease,a femalewas impregnatedwhile this male was ill Kirby, 1990!. Perhapsthis was due to the lag in spermproduction. further studies are neededto quantify concentrationsof reproductivehormone concentrations needed in cetaceansfor successfulreproduction. This would be usefuldata in determiningthe ultimate effect of environmental stressorsthat alter reproductive hormonal events. For females,this studyof S'.bredanensis did not demonstrateany unusualfindings for the reproductivehormone progesterone, where values were often at levelsthat indicatea lack of ovarian activity. It is unfortunatehowever, that estradiol concentrationswere undetectablein this species,as femaleestrogens may serve as key hormonesfor studying pollution effectsand endocrinedisruption. Assaydevelopment in dolphinsfor other forms of estrogenwould be useful in further study of reproductivehormones in cetaceans,as total estrogensand estradiolare the only formspreviously reported, and only for the bottlenosedolphin Kirby, 1990;Yoshioka et al., 1986!. It is likely that reproductivehormone concentrations obtained Rom live animals are reflectiveof populationlife-history parameters determined by the analysesof large samplesizes of deadspecimens. Testosterone concentrations obtained from the 5 captive malesprovide ranges between 0 and 6 ng/ml for S. hredanensisadults. Adult statusin these individuals was established by comparison of individual length and weight measurementswith deadspecimen data determinedin this study of .S'.bredanensis. Availablegonad data from the captive males was also compared with deadspecimen data from 22 males and the relationship to total body length, total body weight and sexual maturityin the species.Results agreed that by the endof the samplingperiod, all males hadattained sexual maturity. Bothgonad measurements and testosterone concentrations suggestthat the captivemale in FrenchPolynesia reached sexual maturity during the studyperiod. Testosteroneconcentrations have been demonstrated to increaseat thetime of sexualmaturity attainment in manyspecies of seals Atkinsonand Ciilmartin, 1992; Noonanet al., 1991!,similar to the pattern in this X bredunensisindividual. Over the 6 year study period, testosteroneconcentrations fluctuated, but increasedto values consistentlygreater than 2 ng/ml near the end of sampling,demonstrating similar concentrationsto those of mature adult bottlenose dolphins Kirby, l990!. Dead specimendata Irom X bredanensismales indicates that individualswith testeslengths greaterthan 25 cm are sexuallymature. An increasein testeslength from the captive male Irom less than 25 cm to 33 cm was observed on ultrasound near the end of the study period. Femalereproductive data obtained from 27 deadspecimens suggests a gradual processof sexualmaturation for S bredanensis.This processis describedby gross physicalchanges in thestructure of theovary as it becomesmore folded and wrinkled andshape changes occur. Findingsindicate that sexual maturation in femalesmay take up to 5 years,although this is probablyan overestimate due to a lackof availablefemale deadspecimens representing young adult age classes in thisstudy. In thehealthy captive 142 femaleS. bredanensis,progesterone values fluctuated above baseline on a few occasions. This individual died suddenlyand no corpora were apparentwhen ovarieswere examined at the timeof necropsy,and accordingly this animalwas classified as sexually immature Perrin and Reilly, 1984!. However.small amountsof progesteroneare secretedby follicles,and a follicle wasapparent upon examination of the ovary. It is possiblethat thesereported progesterone fluctuations are due to follicularactivity. The reproductive stateof this individualmay represent an animalundergoing the gradualprocess of sexual maturation.Total body lengthand weight rangesfor this individualare also in close agreementexamined female dead specimens that were categorized as maturing animals upongross and histological examination of reproductiveorgans. Thisinvestigation provides a significantcontribution to characterizingbaseline values and hormonal events associated with the biology and clinical state of .S. bredanensis. This foundation will be useful for further studies of individual S. hredanensisand thesereference values for variousreproductive states and conditionswill be valuablein assessingstressor eAects to this species.Much remains to be learned aboutthe speciesS. bredanensisand baselinefindings resulting from this study emphasizeareas where future research would be especially useful to aidin conservation. Wildpopulation research focusing on the long-tenn study of photographicallyidentified individualsshould be possible in near-shoreareas of relativelyhigh species abundance suchas FrenchPolynesia. Photographic identification would allow for a greater understandingof behavioralcharacteristics, social fluidity and relationshipswithin S. bredanensispods. The application of technology advancements usedfor other species of cetaceanssuch as tagging and tracking techniques would lead to definitionof movement patterns,home range, and acti'ities at depth. This informationwould contribute significantlyto availableknowledge on theecology of thisspecies. On a world-wide populationlevel, biopsy sampling of wildindividuals would provide the necessary tissue for geneticanalyses. A greatdeal is yetto belearned for S. bredanensis and the baseline speciesdata generated bythis research should serve to stimulatefurther work in areas needed for more effective conservation efforts. REFERENCES Aldridge,B.M, King, D.P., Kennedy-Stoskopf, S.,and J.L. Scott. 2001. Immunology. In CRC Handbookof Marine Mammals:Health, Disease, and Rehabilitation.2 Edition. eds!Dierauf, L.A. and F.M.D. Gulland. CRC Press. Inc., Boca Raton, Florida. Andrews,B. 1999.MMTAG: Bottknose Dolphin Management; Past, present, and future. Reportfrom the Bottlenose Dolphin Reproduction Workshop. Pgs. 7-16. Anonymous.1998. Rare dolphins returned tothe wild. Mote News, 43 !:6-7. Atkinson,S., and W.G. Gilmartin. 1992. Seasonal testosterone pattern in Hawaiianmonk seals.Journal of Reproductionand Fertility. 96:35-39. AuD.W.K.. and W.L. Perryman. 1985. Dolphin habitats in the eastern tropical Pacific. FisheryBulletin 83: 623-643. Baker.A.N. 1983. Whales and dolphins ofNew Zealand and Australia - an identification guide.Victoria University Press, 133 pp. Balcomb,K C. 1980.Washington dolphin stranding. Cetus 2 !:8. Ballance,L.T. 1990. Residence patterns, group organization, and surfacing associations ofbottlenose dolphins inKino Bay, Gulf of California,Mexico. In Thebottlenose dolphin. eds! Leatherwood, S.and R.R. Reeves. Academic Press, San Diego. BallanceL.T., andR.L. Pitman.1998. Cetaceans of the WesternTropical Indian Ocean: distribution,relative abundance, andcomparisons with cetacean communities o ftwo othertropical ecosystems. Marine Mammal Science, 14 !: 429-459. Best,P.B., Canham, A.S., and N. Macleod.1984. Patterns of reproductionin sperm whales,Physeter macrocephalus. Reports of the InternationalWhaling Commission,Special Issue 6:51-79. Bigg,M. 1982.An assessmentof killer whale Orcinus orca! stocks off Vancouver Island,British Columbia. Reports of theInternational Whaling Commission, 32: 655-666. Brager,S., B. Wursig,A. Acevedo,and T. Henningsen.1994. Association patterns of bottlenosedolphins Tursiops truncatus! in Galveston Bay, Texas. Journal of Mammalogy,75 !: 431-437. 145 Brouwer,A., PJ.H.Reijnders, and J.H. Koeman. 1989. Polychlorinated biphenyl PCB! inducesvitamin A and thyroid hormonedeficiency in the commonseal Phoca vitulina!. Aquatic Toxicology, 15:99-106. Bruyus,W.F.J.M. 1971. Field guide of whalesand dolphins. N.v. vitqevevj, Amssterdam. Pgs93-95. Carwardine,M.. Hoyt,E., Fordyce,R.E., and P.Gill. 1998.The NatureCompany Guides:Whales, Dolphins and Porpoises. Time Life Books.Sydney. Australia. 288 pgs. Chantrapornsyl,S.,K Adulyanukosol,and K. Kittiwattangwong.1996. Records of cetaceansin Thailand.Phuket Marine Biological Center. Research Bulletin. 61:39-63. Corkeron,P.J. 1990. Aspects of thebehavior al ecologyof inshoredolphins Tursiops truncatusand Sousa chinensis in MoretonBay. Australia. In Thebottlenose dolphin. eds! Leatherwood, S.and R.R. Reeves. Academic Press, San Diego. Cowan,D.F. 2000. The histopathology of the arlarm reaction in dolphins. 31" annual InternationalAssociation for AquaticAnimal Medicine, joint with American Association for Zoo Veterinarians Conference. Abstracts. Cuvier,G. 1817.Le regne animal distribue d'appres son organisation. pour servir de base al'histoire naturelledes animauxet d'introductiona I'anatomiccomparee. Avec figures,desinees d'appres nature. Tome I, contenantI'introduction, les mammifereset lesoiseaux. Detervoie. aros. 1817, 8 vo.Pp. I-xxxvii, 1-540. Cuvier,G. 1823.Recherches sur lesossemens fossiles, ou 1'onretablit les caracteres de plusieursanimau don't les revolutions duglobe ont detruit les especes. Novelle edition,Tome cinquieme, Ire Partie,Paris. Pg 400 Addition Importante. DeGuise,S. A. Lagace,and P. Beland. 1994. Tumors in St. Lawrencebeluga whales Delphinapterusleucas!. Veterinary Pathology, 31: 444-449. DeGuise,S., K. Erickson,M. Blanchard,L. Dimolfetto,H. Lepper,J. Wang,J.L. Stott, and D.A Ferrick. 1998.Characterization of a monoclonalantibody that recognizes a lymphocytesurface antigen for the cetacean homologue to DC45R. Immunology,94: 207-212. Dohl,T.P., Norris, K.S., and I. Kang.1974. A propoisehybrid: Tursiops X Steno.Journal of Mammalogy.Vol 55 !:217-221. Ellis,R 1982.Dolphins and Porpoises. Alfred A. KnopfInc., New York. 270 pgs. l46 Engelhardt.F.R.. and J.M. Ferguson. 1980. Adaptive changes in harpseals, Phoca groenlandicus,and gray seals, Halichoerus grypus, during the post-natal period. Gen. Comp. Endocrino1.40:434-445. Ferrero.R.C.. J. Hodder,and J. Cesarone.1994. Recent strandings of rough-toothed dolphins..Stenohredunensis, on theOregon and Washington coasts. Marine Mammal Science 10 !: 114-116. Fowler.C.W. 1984.Density dependence in cetaceanpopulations. Reports of the InternationalWhaling Commission, Special Issue 6:373-379. Fraser.F.C. and P.E. Purves. 1960. Hearing in cetaceans:Evolution of theaccessory air sacsand the structureand functionof the outer middle ear in recentcetaceans. Brutish Museum of Natural History 7 !. 140 pgs. Gannier.A. 2000. Distributionof Cetaceansoff the SocietyIslands FrenchPolynesia! as obtainedfrom dedicatedsurvey. Aquatic Mammals26 !.111-126. Gannier. O. and A. Gannier. 1998. First results on the distribution of cetaceansin the societyislands French Polynesia!. 12 BiennialConference of theBiology of Marine Mammals. Abstract. Gaspar.C.. I.enzi. R., Reddy, M.L, and J.C. Sweeney. 2000. Spontaneous lactation by an adult Tursiopstrunca.'us in responseto a strandedSteno hredanensis calf. Marine Mammal Science, 16 !:653-658. Gray.J.E. 1846.On theCetaceous animals. Pp. V-xii, 13-53,plls 37 in: VolumeI, part III, Mammaliaof Richardson.J. and Gray, J.E. eds! The Zoology of the voyage of H.M.S. Erebusand Terror, underthe commandof CaptainSir JamesClark Rossduring the years 1839 to 1843.2 vols,quarto, 1844-1875. Longmans, Brown, Greenand Longmans, London. Gotch,A.F. 1995.Latin Names Explained: A guideto thescientific classification of reptiles,birds and mammals. Facts on File, New York. Hadley,M.E. 1994.Endocrinology. 3 Edition., PrenticeHall, NewJersey. Hall, A.J., N.J.L. Green,K.C., Jones,P.P., Pomeroy, and J. Harwood. 1998.Thyroid hormonesas biomarkers in greyseals. Marine Pollution Bulletin, 36:424-428. Hansen,L.J. 1990.California coastal bottlenose dolphins. In The bottlenosedolphin. eds!I eatherwood,S. and R.R. Reeves. Academic Press, San Diego. Harrison,R., andM.M. Bryden.1988. Whales, Dolphins, and Porpoises, 1988. Facts on File. New York. Haulena.M.. D.J.St. Aubin, andP.J. Duignan 199$.Thyroid hormone dynamics during thenursing period in harbourseals, Phoca vifulina. Can. J. Zool.,76:48-55. HetzeLB., andL. Lodi. 1998.Investigating a massstranding of rough-tootheddolphins in BraziL Whalewatcher 31 !:18-20. Hewitt,R.P. 1985. Reaction of dolphinsto a surveyvessel: effects on census data. Fishery Bulletin, 83 !: 187-193. Hoyt,E. 1984.Whale watchers handbook. Doubleday Press, New York Pgs78-79. Kasuya,T. 1984.Effect of exploitationonthe reproductive parameters of two species of Stenellaoff thePacific coast of Japan.Reports of theInternational Whaling Commission.Special Issue 6: 478. Kenney,R.D. 1990.Bottlenose dolphins off theNortheaster United States. In The bottlenosedolphin. eds! Leatherwood, S. andR.R. Reeves. Academic Press, San Diego. Kirby.V.L. 1990.Endocrinology of marine mammals. In CRCHandbook of Marine Mammals:Health. Disease,and Rehabilitation,2 Edition. ed! Dierauf,L.A. CRC Press, Inc.. Boca Raton, Florida. Klinowska,1991. Dolphins, Porpoises and Whales of theWorld: ICUN Red Data Book. Compiledby Klinowska,M. Gland,Switzerland. l.eatherwood,S.. Caldwell,D.K., and Winn, H.E. 1976.Whales, Dolphins, and Porpoises of the Western North Atlantic. A Guide to Their Identification.NOAA Technical ReportNMFS CIRC-396, iv +176 pp. LeatherwoodS.. and R. R. Reeves.1983. The SierraClub Handbookon Cetaceans.Sierra Club Books, San Francisco, CA. Leatherwood,S., Reeves,R.R., Perrin,W.F., and W.E. Evans.1982. Whales, Dolphins, andPorpoises of the Eastern North Pacific and adjacent arctic waters: A guideto their identification. National Marine FisheriesService, Circular 444, v + 245 pgs/ Lesson,R.P. 1828.Cetaces. Volume I in: Lesson, R.P. 1828-1844.Histoire naturelle, generaleet particulieredes mammiferes et desoiseaux decouverts depuis 1788 jusqu'anos jours. Baudouin Freres, Editeurs, Paris, 10 volumes, 8 toiii. Seuleed. Complete,20 vols, 8 vo,pl l. Maps,Bruxelles!. Lodi, L. 1992.Epimeletic behavior of free-rangingrough-toothed dolphins, Steno hredanensis,from Brazil. Marine Mammal Science, !:284-287. Mazzuca.L.. Atkinson. S., Keating,B., and E. Nitta. 1999.Cetacean mass strandings in the HawaiianArchipelago, 1957-1998. Aquatic Mammals 5 :103-114. Mead.J.M. and C.W. Potter. 1990.Natural history of bottlenosedolphins along the centralatlantic coastof the United States.In The BottlenoseDolphin eds! Leatherwood,S. and R. Reeves.Academic Press, San Diego. Pgs. 165-195. Mizroch.S.A..and A.E. York. 1984.Have pregnancy rates of southernhemisphere Fin whales,Balaenoptera physalus, increased? Reports of the InternationalWhaling Cmmission,Special Issue 6:401-410. Miyazaki,N. 1980.Preliminary note on age determination and growth of therough- tootheddolphin, Steno bredanensis, off the Pacific coast of Japan.Reports of the InternationalWhaling Commission, No. 3, 171-179. Miyazaki.N. 1984.Further analyses of reproductionin the striped dolphin, Stenella coeruleoalba.off the Pacific coastof Japan.Reports of the InternationalWhaling Commission.Reproduction in whales, Dolphins and Porpoises. Special Issue 6, 343-353. Miyazaki.N. 1989.Stranding records of cetaceanson theJapanese coasts. International MarineBiological Research Institute Reports. 1:21-25. Miyazaki.N. andW.F. Perrin. 1994. Rough-toothed dolphin Steno bredanensis Lesson. 1828!.Pages 1-21 in S.H.Ridgway and R.J. Harrison, eds. Handbook of Marine Mammals.Volume 5: The first bookof dolphins. AcademicPress, London, UK. Mobley,J.R., S.S. Spitz, K.A. Forney, R. Grotefendt. and P.H. Forestell. 2000. Preliminaryresults of 1993-98aerial surveys. Southzest Fisheries Science Center AdministrativeReport U-OO-14C. 26 pp. Available Rom SWFC, P.O. Box 271, La Jolla. CA 92038. Noonan,L.M., K. Ronald,and J.l. Raeside.1991. Plasma testosterone conentrations of captivemale hooded seals Cytophora cristata!. Canadian Journal of Zoology, 69:2279-2282. Norris,K.S. The Porpoise Watcher. 1974. W.W. Norton and Company, New York Norris,K.S., Baldwin,H.A., andD.J. Samson.1965. Open ocean diving testwith a trainedporpoise. Deep Sea Research, 12:505-509. Norris,K.S., and T.P. Dohl.1979. Behavior of theHawaiian spinner dolphin, Stenella longirostris.Fishery Bulletin 77 !:821-849. Norris,K.E., and T.P. Dohl. 1980.The behaviorof the Hawaiianspinner dolphin, Stenellalongirostris. Fishery Bulletin, 77: 821-849. 149 Ott, P.H., and D. Danielwitz. 1996.Southward range extension of Stenobredanensis in the Southwest Atlantic and new records of Stenella coeruleoalba for Brazilian waters.Aquatic Mammals22 !:185-189. Parsons,E.C.M. 1998.Strandings of small cetaceansin Hong Kong territorial waters.J. Mar. Biol. Ass. U.K.. 78:1039-1042. Palacios,D. 1995.Cetacean remains and strandingsin the GalapagosIslands. Abstract: 1995Northeast Regional Stranding Network Conference, Riverhead, New York. 28-30 April. Perrin,W.F. 1989.Dolphins, porpoises and whales. An actionplan for theconservation of biologicaldiversity: 1988-1992. ICUN Gland,Switzerland. 30 pgs. Perrin,W.F. and R.L. Brownell.1984. Reproduction in whales,dolphins and porpoises. Reportsof theInternational Whaling Commission. Reports of theInternational Whaling Commission.Special Issue: 6, xii and 495. Perrin. W.F., and J.R. Henderson.1984. Growth and reproductiverates in two populationsof spinnerdolphins, Stenella longirostris, with different histories of exploitation.Reports of theInternational Whaling Commission. Reproduction in Whales,Dolphins and Porpoises.Special issue 6, 417-430. Perrin,W.F. and A.A. Hohn.1994. Pantropical spotted dolphin Stnella attenuata Gray, 1946!In S.H.Ridgway and R J.Harrison, eds. Handbook of MarineMammals. Volume 5: The first book of dolphins. Academic Press.London, UK. Perrin, W.F.. E.D. Mitchell, J.G. Mead,D.X. Caldwell, M.C. Caklwell, P.J.H.Van Bree, and W.H. Dawbin. 1987.Revision of the spotteddolphins, Stenella spp. Marine Mammal Science 3 !:99-170. PerrinW.F., and Reilly, S.B. 1984.Reproductive parameters of dolphinsand small whalesof the family Delphinidae.Rep Int. Whal.Commn, Special Issue 6: 97- 133. Pryor,K. 1975.Lads before the wind: Adventures in Porpoise Training. Harper and Row, Publishers, New York. 278 pgs. Pryor,K andK.S. Norris. 1991. Dolphin Societies, Discoveries and Puzzles. eds! Pryor K andK.S. Norris.University of CaliforniaPress, Oxford England. Pgs. 79-112. Reynolds,J.E. III, R.S.Wells, and S.D. Eide. 2000. The Bottlenose Dolphin: Biology and Conservation.University Press,Gainsville, Florida. 150 Richards,J.M. 1934.Resultats des campagnes scientifiques accomplies sur son yacht par Albert ler princesouverain de Monaco.Fascicule LXXXIX. 472 pgs. Ritter, F. 2002. Behavioralobservations of rough-tootheddolphins Steno bredanensis! off La Gomera,Canary Islands 995-2000!, with specialreference to their interactionswith humans.Aquatic Mammals,28 I !: 46-59. Romano,T.A., S.H.Ridgway, and V. Quaranta.1992. MHC class II moleculesand immunoglobulinson peripheralblood lymphocytes of the bottknosedolphin. Tursiopstruncatus. Journal of ExperimentalZoology, 263: 96-104. Romano, T.A., S.Y. Felten, J.A. Olschowka, and D.L. Felten. 1994. Noradrenergic and petidergicinnervation of lymphoidorgans in the beluga,Delphinapterus leucas: An anatomicallink betweenthe nervousand immunesystems. Journal of Morphology, 221:243-259. Romano,T.A., Ridgway, S.H., Felten,D.L, and V. Quaranta.1999. Molecular cloning and characterizationof CD4 in an aquaticmammal. the white whale Delphinapterusleucas. Immunogenetics, 49:376-383. Rudolph,P., Smeenk,C.. andS. Leatherwood.1997. Preliminary checklist of Cetaceain the IndonesianArchipelago and adjacent waters. Zoologisches Verhandelingen, 312:1-50. Saayman,G.S., and C.K. Tayler. 1973. Socialorganisation of inshoredolphins Tursiops a Juncusand Sousa!in the Indian Ocean.Journal of Mammalogy,54 !:993-996. Schlais,J.F. 1984.Thieving dolphinsa growingproblem in hawaii's fisheries.Sea Frontiers, 30 !:293-298. Shallenberger,E.D. 1981.The statusof Hawaiiancetaceans. Marine Mammal CommissionReport, MMC-77/23 NTIS PB82-109398!.79 pages. Scott, M.D., Wells, R S. and B. Irvine. 1990. A long-term study of bottlenosedolphins on the west coastof Florida. In The bottlenosedolphin. eds! Leatherwood,S. and R.R. Reeves.Academic Press, San Diego. Schroeder,P.J., and K.V. Keller. 1989. Seasonalityof serumtestosterone levels and spermdensity in Tursiopstruncatus. Journal of ExperimentalZoology, 249:316- 321. Schumacher,U., S. Zahler, S., G. Heidemann,K. Skirnisson,and U. welsch. 1993. Histologicalinvestigations on the thyroidglands of marinemammals and the possibleimplications of marinepollution. Journal of Wildlife Diseases,29: 103- 108. 151 Shane.S.H. 1980.Occurrence. movements, and distribution of bottlenosedolphin. Tursiopstruncatus, in SouthernTexas. Fishery Bulletin 78 !: 593-599. Smith,T.D. 1983.Changes in sizesof threedolphin Stenellaspp.! populations in the easterntropical Pacific. Fishery Bulletin, 81:1-14. Smith. D.S., and H. Whitehead.1999. Distribution of dolphins in Galapagoswaters. Marine Mammal Science 15 !:550-555. Southern.S. 2000.Molecular analysis of stress-activatedproteins and genes in dolphins andwhales: A newtechnique for monitoringenvironmental stress. 31" annual InternationalAssociation for AquaticAnimal Medicine,joint with American Association for Zoo Veterinarians Conference. Abstracts. St. Aubin.D.J., and J.R. Geraci. 1988. Capture and handling stress supresses circultating levelsof thyroxineand triiodothyronine in belugawhales, Delinapterus leucas. Physiol.Zool. 61:170-175. St. Aubin. D.J.. and J.R. Geraci. 1992.Thyroid hormonebalance in belugawhales. Delphinaptert~cleucas: Dynamics aAer capture and influence of thyrotropin. CanadianJournal of Veterinary Research,56:1-5. St. Aubin,D.J., S.H. Ridgway,R.S. Wells, and H. Rhinehart.1996. Dolphin thyroid and adrenalhormones: Circulating levels in wild and semi-domesticatedTursiops truncatus.and influence ot sex,age, and season. Marine Mammal Science.. 12: I- 13. St. Aubin,D.J. 2001. Endocrinology In CRCflandbook of MarineMammals: Health, Disease.and Rehabilitation.2" Edition. eds! Dierauf.L.A. and F.M.D. Gulland. CRC Press. Inc., Boca Raton, Florida. St. Aubin, D.J. and L.A. Dierauf. 2001. Stressand Marine Mammals.In CRC Handbook of Marine Mammals:Health, Disease,and Rehabilitation,2 Edition. eds! Dierauf, L.A. and F.M.D. Gulland. CRC Press,Inc., Boca Raton,Florida. Stokkan,K.A., M.K. Vaughan,R.J. Reiter, L.P. Folkow, P.E. Martensson. G. Sager,C. Lydersen,and A.S. Blix. 1995.Pineal and thyroid functions in newbornseals. Gen.Comp. Endocrinol.98:321-331. Stolen,M.K., McFee,W.E., and B.G. Mase. 1999. Age and growth estimation of three massstrandings of rough-tootheddolphins Steno bredanensis! in the Gulf of Mexico, FL. ThirteenthBiennial ConferenceBiol. Mar. Mamm.,Abstr. Subramanian,A., S. Tanabe,R. Tatsukawa,S. Saito, N. Miyazaki. 1987.Reduction in the testosteronelevels by PCB'sand DDE in Dali's porpoisesof NorthwesternNorth Pacific. Marine Pollution Bulletin 18 2!: 643-646. 152 Thompson,C.A., andJ.R. Geraci. 1986. Cortisol, aldosterone, and leucocytes in the stressresponse of bottlenosedolphins, Tursiopstruncatus. Can. J. Fish. Aquat. Sci. 43:1010-1016. Thomas.J.A.. R.A. Kastelein.and F.T. Awbrey. 1990.Behavior and blood catecholaminesof captivebeluga whales during playbacks of noise&om an oil drillingplatform. Zoo Biology,9: 393-402. Van Waerebeek.K.. J.C. Reyes,A.J. Read,and J.S. McKinnon. 1990. Preliminary observationsof bottlenosedolphins &om the Pacificcoast of SouthAmerica. In The bottknosedolphin. eds! Leatherwood,S. and R.R. Reeves.Academic Press, San Diego. Wade P.R. and T. Gerrodette. 1993. Estimates of cetacean abundance and distribution in the EasternTropical Pacific. Reportsof the InternationalWhaling Commission 43: 477-493. V ~lsh.M. T., R.Y. Ewing, D.K. Odell. and G.D. Bossart.2001. Massstrandings ot cetaceans In CRC Handbook of Marine Mammals: Health. Disease, and Rehabilitation, 2" Edition. eds! Dierauf, L.A. and F.M.D. Gulland. CRC Press, Inc., Boca Raton, Florida. Watkins, W.A., P. Tyack, and K.E. Moore. 1987.Steno hredaneasis in the Mediterranean Sea. Marine Mammal Science 3 !: 78-82. Wells,R.S. andM.D. Scott.1999. Bottlenose dolphin Tursiops truncatus Montagu, 1821. In Handbookof MarineMammals Volume 6 The SecondBook of Dolphinsand Propoises. eds! Ridgeway,S.H. andR.J. Harrison. Academic Press, New York. Wells,R.S., L.J. Hansen.A. Baldridge,T.P. Dohl,K.L. Kelly, andR.H. De&an.1990. Northwardextension of the rangeof bottlenosedolphins along the California coast.In The bottlenosedolphin. eds! Leatherwood, S. and R.R. Reeves. AcademicPress, San Diego. Wells,R., Manire,C., Rhinehart,H., Smith,D., Westgate,A., andF. Townsend.1999. Rangingpatterns of rehabilitatedrough-toothed dolphins. Stneo bredanensis, released in the northeastern Gulf of Mexico. 13 Biennial Conference of the Biologyof Marine Mammals.Abstract. West,K.L., S. Atkinson,E. Hanahoe,C. Schwetz,J.C. Sweeney,and R. Stone.2002. Thyroidhormone baseline values in thebottlenose dolphin. changes according to age/sexclasses and reproductivestate. In Preparation. 153 West, K.L. and R. van Woesik .2001.Spatial and temporalvariance of river dischargeon Okinawa Japan!: Inferring the temporalimpact on adjacentcoral reefs. Marine Pollution Bulletin: 42 0! 864-872. West. K.L.. S. Atkinson. C. Schwetz,J.C. Sweeney,and R. Stone.2001. Thyroid hormoneconcentrations during different reproductive states in adultfemale bottlenosedolphins Tursiops truncatus!. 32 nd AnnualInternational Association for AquaticAnimal Medicine Conference. Abstracts. Whittow.G.C., Hampton,I.F.G., andC.A. 1978.Ohata. Body temperature of therough- tootheddolphin. Journal of Wildlife Management,42 !:184-185. Woldstad,S., andB.M. Jenssen.1999. Thryoid hormones in greyseal pups Halichoerus ~pus!. Comp.Biochem. Physiol. A 122:157-162. Wursig,B., andG. Harris.1990. Site and association fidelity in bottlenosedolphins off Argentina.In The bottlenosedolphin. eds! Leatherwood, S. andR.R. Reeves. AcademicPress, San Diego. Young.N.M, Ludicella,S., Evans,K. andBaur, D. 1993.The incidentalcapture of marine mammalsin U.S. Fisheries.Center for Marine Conservation.Washington, D.C., xvi +415 pp. Yosihoika. M., E. Mohri, T. Tobayama.K. Aida, and I I-Ianyu. 1986.Annual changesin serumreproductive hormone levels in the captive female bottlenoseddolphins. Bulletin of the JapaneseSociety of Scientific Fisheries,52 I 1!: 1939-1946. 154