Reproductive Strategies in a Population of Gobiosoma bosci (Osteichthyes: Gobiidae) with Slow and Fast Maturing Individuals

6 Pages·1996·4 MB·English

by  Candace H Conn

GulfResearchReports,Vol.9,No.3,177-182, 1996 ManuscriptreceivedJune2,1995;acceptedAugust14,1995
REPRODUCTIVE STRATEGIES IN A POPULATION OF
GOBIOSOMA BOSCI (OSTEICHTHYES: GOBIIDAE)
WITH SLOW ANDFASTMATURINGINDIVIDUALS
CandaceH.Connand DavidL.Bechler*
LamarUniversity,DepartmentofBiology,P,O.Box10037,Beaumont,Texas,77710,USA
ABSTRACTThereproductivebiologyofGobhsomaboscicollectedfromNovember1986toOctober1987inthe
McFaddinWildlifeRefugeinsoutheastTexaswasstudiedbyusingmorphometricdata. Malesachievedgreater
weightsperunitlengththanfemales,andlongevitywasabout12to13months. GSIvaluesandmeanmonthly
ovumdiametersindicatedthatthebreedingseasonranfromApriltoSeptember,withamajoractivitypeakin
MayandaminorpeakinSeptember. Significantdifferencesinmaleandfemalestandardlengths(SL),ovum
diameter, and egg number existed for sexually mature specimens between the first and second peaks of
reproductive activity. Aneggversuslengthanalysisproducedapositivelinearrelationship. Anaccessory
gonadalstructureindex(ASGI)wasdevelopedandrevealedthatmaximalAGSdevelopmentcorrespondedwith
themaleGSI,butdidnotproducediscemablepeaks. Tworeproductivestrategieswerefollowedanddepended
upontimeofhatchingandgrowthrate. Someindividualsthathatchedearlyinthebreedingseasongrewrapidly
andwerecapableofegglayingbyAugustorSeptember. Individualshatchedlateinthebreedingseasondelayed
breedinguntilthefollowingseason.
Introduction Gobiosoma bosci, a cavity nester preferring hard
substrates(Bechleretal.1990),hasbeenstudiedextensively
Thedistributionofthenakedgoby,Gobiosomabosci, alongtheAtlanticCoast. DahlbergandConyers(1973),
extends from Long Island, New York, to the state of who reviewed much of the literature, postulated that
Campeche,Mexico(HoeseandMoore 1977). Inviewof spawning seasons for different populations of G. bosci
themarkedenvironmentaldifferencesthroughoutitsrange, were variable, and depended on location. They related
G.boscimaybeexpectedtovaryitsreproductivestrategies. initiationandterminationofspawningtowatertemperature
Inareviewoflifehistoryphenomena.Cole(1954)stated andindicatedpeakspawningactivitywasinthewarmest
thattheageatwhichreproductionbeginsisoneofthemost months.MaythroughAugust. However,theydidnotdiscuss
significantcharacteristics,andassuchwillinfluencethe anyvariationsinbreedingstrategies.OtherreportsonG.ho.^ci
reproductivesuccessofanindividual.Therefore,southerly byNeio(1976),CrabtreeandMiddaugh(1982),andFitzsimons
populations of G. bosci have at least two potential andSeok(1989)revealedmanyfactsaboutthelifehistoiyand
reproductive strategies available to them (Steams and ecdogy ofG. bosci. However, none ofthe above studies
Crandall1984).First,anindividualmaygrowthroughout examinedreproductivestrategiesofthenakedgobyindetail
thenon-reproduedveseasonafterhatchingandthenengage along the Gulf Coast. In this study, we concentrated
in breeding during the subsequent breeding season. specificallyonlifehistorydatarelatedtothereproductive
Alternatively,young-of-the-yearmaygrowrapidlyenough biologyofG.bosciinasoutheastTexassaltmarsh.
toreproducebeforetheendofthebreeding season,and
therebygainabreedingseasonnotavailabletofi^with DescriptionoftheStudyArea
delayedbreeding. Assuch,individualswithinthesame
populationmayemployonestrategyortheotherdepending TheMcFaddinNationalWildlifeRefugeislocatedin
upontheirtimeofhatchingandenvironmentalconditions. southernJeffersonCounty,20kmsouthwestofSabinePass,
Texas(GiiffithandBechler1995),asubtropicaltotemperate
regionoftheUnitedStates. Therefugeisabrackishmarsh
* Correspondingauthor;currentaddressisDept,ofBiology, dissectedbymeanderingcreeks,man-madecuts,andshallow
ValdostaStateUniversity.Valdosta.GA31698 lakes;thelargest,CIamLake,wastheprimaiysiteofthissmdy.
177
Conn andBechler
ClamLakeisrelativelyshallow,1.0to1.3mindepth,with Results
thedeepestarea2.8m. Thesubstrate^dbanksalongthe
edgesofthe consistofafirmclay. Afinesiltsubstrate
lies1to5mfromshore.Theclayhanksareriddledwithholes GeneralEcology
andtunnelsinhabitedbyRhithropanopeusharrisi,themud
crab,twospeciesofVca, thefiddlercrabs,andCaUinectes Salinityduringtheyearinwhichthefishwerecollected
sapidus,thebluecrab.Thepredominantflorasurroundingthe rangedfrom0to12ppt. Watertemperaturerangedfrom
lake is Phragmites australis, a reed grass, and Spartina 7 to 32®C. During the months ofhighest reproductive
altemiflora,asaltmarshgrass. Tidalinfluences inClam activity,ApriltoJune,salinityvariedfrom2to6pptand
Lakeareminimal.Wind-generatedwaveactionundercuts meanwatertemperaturewas28’’C. Daylengthincreased
thebanksofthelakeandresultsinthegradualerosionof from12.48hourson1April1987to14.05hourson30June
the shoreline and a subsequent exposure of roots of 1987 (United States Naval Observatory 1965). In
PhragmitesandSpartina.Theserootsarewashedfreeof September, when there was a smaller peak in
soilandextendintothewaterexceptduringdroughtand reproductive activity, the salinity was 12 ppt and
lowtide. LargeclumpsofPhragmitesrootsbreakoffand water temperature was 28®C. Day lengthdecreased
fallintothewater.Theseremainedclosetothebanksofthe from 12.81hourson 1 September1987to11.90hours
lakebelowthewaterandprovidedhabitatforG.bosci. on30September 1987.
Specimens forpreservation were collected in clam
MethodsandMaterials shells,algalmats,grassclumps,andPhragmitesaustralis
rootsalongtheedgeofthelake.Seiningoversoftsediments
BeginninginNovember1986,monthlycollectionsofG, producedfewspecimenscomparedtothenumbercollected
bosci weremadeinClamLakefw*aperiodof12months, closetothebanks.
exceptforJuly.whennocollectionwasmade.Sixtyto100fish
werecollectedeachmonthbyusinga3.23mmmeshseine. LifeHistory
Specimens wereliardened in 10% fornialin for 24 hours,
washedinwaterfor24hours,andpreservedin55%isopropyl Length-frequency relationships examined by month
alcohol. Aminimumof28specimens(14males,14females) indicatedthatG.bosci lifeexpectancywasnotmorethan
wereexaminedfromeachmonthlycollection.Standardlength 12to13months(Figure1). Thisconclusionwasbasedon
(SL)wasmeasuredtothenearest0.01mmwithaMitutoyodial thefactthatduringthemonthsofAugustandSeptember
mm
caliper.Totalbodyweightwasmeasuredtotheneaiest0.(XX)l specimensgreaterthan25and29 respectivelywerenot
gwithaMetllerAE 100analyticalbalance. Gonadswere collected. However,from NovembertoJuneindividuals
removedandwetweighedtogetherasaunitforuseinobtaining exceeding29mm(range=30.()-44.4mm)werecommon.
Gonadal Somatic Indices (GSI). Ten accessory gonadal AbroadrangeofsizeclassesfromDecembertoApril
structures(AGS),whicharerelatedtosexualmaturationof indicatedthatG.bosci mostlikelyreproducedthroughout
male gobies (Miller 1984, Cole and Robertson, 1988, the spring and summer. The smallest specimens
Lahnsteineretal.1992),wereremovedandwetweighedeach (minimum=ll.l mm) were collected from August to
month. TheweightoftheAGSwasdividedbythetotalwet Octoberandrepresentedofferingnewlyrecruitedintothe
bodyweightandmultipliedby 100toproduceanAccessory populationduringlatespringandsummer. Themaximum
GonadalStructureIndex(AGSI)similartothatoftheGSI.hi sizeofspecimensprogressivelyincreasedfromOctoberto
females,lOovaccntrallylocatedonthesuxfaceoftheovaryand April when a maximum size of44.4 nun was reached.
randomlyselectedweremeasuredin10animalseachmonth From April to June, the beginning of the reproductive
withanocularmicrometer. DuringthemonthsofApriland seasonasdiscussedbelow,thesizeofG.boscidecreased
May,thebeginningofthebreedingseascai,thediametersof fromthemaximumto40.0mm(Figure1). Themaximum
^175to200ovaonthesurfaceoftheovariesweremeasuredin sizeofG.bosci inAugustwas24.8mm,inSeptember,
fishtodetermineifmorethanonesizeclassofovawere 29.7 mm,and inOctober, 36.6 mm. Theincreasein
presentEggslargeenoughtobeteasedfromtheovarieswoe maximumsizebetweeneachpairofconsecutivemonths
countedbyhandwithanOlympuszoomstereomicroscope was4.9mmand6.9mmwithanaverageof5.9mm/month.
system. Theseovawereclassifiedastotheirdevelo{Hnental Thus if a growth rate of approximately 6 mm/month
stagebyfollowingHeinsandRabito(1986).Salinityandwater throughthesummerisassumed,thelargestindividualsin
tOTip^atureweretakeneachmonthwithaYellow Springs August would be no more than about fourmonths old.
TSC meter. All statistical analyses employed MINITAB Thesefishindicatethatallgobiescollectedinthismonth
(1986)software. representedanewgeneration.
178
ReproductiveStrategiesinG.bosci
Length-weight relationships comparing males and Weighty=0.281+0.0364SL+0.0018SL2
females showed that males notonly reached a greater Weight,^=0.291+0.0309SL+0.0008SL2
maximum standard length (44.4 mm) thanfemales Thesetworegressionmodelsweresignificanllydifferent
(34.7 mm), but were heavier per unit body length (N^367; R^O.958: DF=3, 363; F=6.1891;P<0.005) and
(Figure2). Acurvilinearrelationshipbetweenstandard showedthatmaleswereheavierthanfemalesperunitlength.
lengthand weightexisted formales andfemales. The The results of l-tests indicate that the quadratic (t=4.03,
regressionanalysisproducedthefollowingequations: P<0.01) and linear (t;=;3.27, P<0,01) components were
responsibleforthedifferencesbetweenthemodels.
Reproductiveadults
Ovaeasily teasedfrom theovarieswereclassifiedas
4-5monthsold
maturetoripe(HeinsandRabito 1986). Meannumberof
maturetoripeovaperfemalewas466(N=40,SD=178.2)
witharangeof116to1030. Theseovamadeuptwodistinct
sizeclassesineachofsixfemalesexaminedfromApriland
May(Figure3).
MeandiameterforallstagesofdevelopingovainMay
was0,45 mm(N=12,SD=0.1153). TheSeptembermean
ovumdiameterwasmuchlower.0.19nun,butthevariance
greater(N=8 SI>=0.1538).Meanovumdiameterofmature
andripening,eggsinAprilandMaywas0.49mm(range:
0.35to0.56mm)and,inSeptember,0.42mm(range:0.34
to 0.50mm). A two sample t-test showed a significant
differenceinthediameterofmatureandripeningovaforthe
twoperiods(N=23;1^2.75;DF=21;P=0.012).
Not only were ovum diameters smaller during the
secondbreedingpeak,butovumnumberperfemalewasalso
Months
significantly lower(N=29: DF=3,25:RM).73; F=22.42;
P=0.0001). Aregressionanalysiscomparingeggnumber
Reproductive adults
12-13monthsold Nolargereproductive versus SL for females from April and May against
adults females from August and September produced the
Figure 1. Length-frequency histogram for the months followingmodels:
involvedinthebreedingseasonofG.bosciintheMcFaddin Bggs^,^-577+45ASL
National Wildlife Refhge, Sabine Pass, TX. Size class EggSp^i=-382+34.0SL
incrementsarebasedon2mmintervals.Dataforthemonth
ofJulyaremissing. ThemonthofOctoberisgivenforsize
comparisons,butisnotpartofthebreedingseason.
SL(mm)
Figure2. Length-weightrelationshipsformaleandfemale Figure3.Ovumsizeclassfrequenciesforsixfemalescollected
G.bosciaregiven. The367datapointsusedtocomputethe in April and May. Size class increments are based on
mm
regressionlinesarenotshown. 0.05 intervals.
179
Conn andBechler
During December the AGS were small and difficult to
locateandremove.Inmonthsofhighreproductiveactivity,
the AGS were large and completely filled the ventral
portionoftheabdominalcavity.
AGSIvalues(Figure5)followedapatternsimilarto
GSIvaluesformales. AGSIvalueswerelowNovember
throughDecemberandroseslowlyfromJanuarytoMarch.
UnlikeGSIvalueshowever.AGSIvaluesdidnotpeakin
AprilbutcontinuedtoclimbandpeakedinMay,remained
highthroughAugust,andthendeclineduntilOctober.
OvarianmaturationinfemaleG.bosciwasdetermined
by ovum diameter. Mean ovum diameters (Figure 6)
remainedlowOctoberthroughFebruary,andthenbeganto
Figure4. LinearrelationshipbetweenSLandeggnumber increase in March. The increase in ovum diameter in
fMocrFaaldldfienmaNlaetsicoonlallecWtielddldiufreinRegftuhgee,brSeaebdiinnegPsaesas,soTnXi.nthe Marchsignaledtheonsetofovummaturation. Therefore,
themeanovumdiameterforMarchwasusedasacriterion
againstwhichovumdiametersforfishexaminedinother
Thedifferencebetweenthetwomodelsresultedfrom monthswerecompared. Allfemales whosemeanovum
significant differences in the coefficients of the slopes diameterwaslargerthan0.1924mm,themeanovum
(t=2,22. P=0.036), but not in the intercepts (t=0.51, diameter for March, were considered to possess
P=0.617).Themeaneggnumberduringthefitstpeakwas505 maturingova.
(SD=137.1)and252(SI>=95,2)duringthesecondpeaL Of32femalesexaminedinAugustandSeptember,13
Thetotalnumberofmaturingtoripeovaintheovaries hadameanovumdiametergreaterthan0.1924mm,the
ofall females produced the following regression model minimummeansizeofmaturingova.ThemeanSLofthese
(Figure4): 13femaleswas20.8mm(range=16.9to24.0mm). Ofthe
Fecundity=-416+39.5SL 40fishexaminedinAprilandMay,31hadovumdiameters
ThepositivecorrelationbetweenSLandovanumber exceeding0.1924mm. Themeanstandardlengthofthese
was significant (N=37; DF=1, 35; F=45.836; P<0.001; 31femaleswas24,1nun(range=18.9mmto29.6mm). A
R^O.567). comparisonofSLindicatedthatsexuallymamrefemales
MeanmaleGSI values werelow fromNovemberto fromAugustandSeptemberweresignificantlyshorterthan
January,begantoriseinFebruaryandMarch,andpeaked females from April and May (N=22; DF=20; t;=4,92;
in April (Figure 5). GSI values then declined slightly P=0.0001). In addition, significantly fewer females
duringJune,producedasecondpeakinAugust,andthen possessedmaturing ovaries during the second breeding
declinedinSeptember. TheAprilpeakhadameanvalue peak(N=32,DF=1,X^24.953,P<0.001).
of1.266andtheAugustpeakameanof1.284. MaleGSI
values peaked one month prior to the female peakGSI 15
valuesforbothreproductiveperiods.
FemaleGSIvaluesshowedabimodaldistribution,but
with a more dramatic change than males (Figure 5).
Female GSI values remained low November through o
December,beganagradualincreaseinJanuary,increased CoO
rapidlyFebruarythrough April, andpeaked in May ata
mean value of 14.26. GSI values declined in August,
peakedagaininSeptemberatameanvalueof4.10 and
declinedinOctober.
0
InmaleG.bosci theAGSwasmitten-shapedwitha
smalllobeextendinglateraltothemainlobe,attachedtothe
Month
bodywallnearthevent,andextendeddorsallybetweenthe
innerbody walland theintestinal tract. Each AGS was Figure5. MonthlymeanGSIandAGSvaluesformalesand
attachedto the caudalendofthe testisby a small duct. females. DataforthemonthofJulyarenotgiven.
180
Reproductive Strategies inG.bosci
Malesexualmaturitywasdeterminedonthebasisof
GSIvalues(Figure5). GSIvalues*whichbeganincreasing
in March, were used as an indicator ofmaturing testes
(N=16,^=0,5624).Of23malesexaminedinApril,20had
aGSIexceeding0.5624. ThemeanSLofthese20males
was30.6mm(range: 19.5to44.4mm). GSIalsopeaked
in August Mean SL ofmales with aGSI value above
0.5624 was 20.5 mm (range: 17.3 to 24,6 mm), A two
samplet-testcomparing maturingmalesfromeachtime
periodindicatedthatmalesmaturedatashorterSLinthe
latesummer(N=61;DF=56.9;1=6.10;P=0.0001).
Discussion
According to conclusionsbyDahlbergandConyers Figure 6, Monthly mean ovum diameters. Datafor the
(1973),populationsofG.bosci inthenorthernpartsofits monthofJulyarenotgiven.Openportionsofbarsrepresent
standarddeviations.
rangeshouldreproducelateinlifeafterhavinggrownand
maturedoverthewinter. Populationsatthesouthernend
oftherange,suchasinsouthFloridaandMexico,would ovulated and laid, and indicating that the majority of
beexpectedtomatureearlierandpossiblyatasmallersize individualsdidnotreproduceuntilthefollowingspring.
duringtheprotractedbreedingseason. Theprimaryfactor Fromthisanalysisweconcludethatindividua!members
controlling the age at first reproduction, and thus the ofthe G. bosci population in Clam Lake followed two
reproductivetacticemployedbyanindividual,wouldbethe distinctreproductivestrategiesbasedupontheirtimeof
lengthofthebreedingseasonincombinationwiththetime hatching. Some individuals that hatched early in the
ofhatching. breedingseasonmaturedrapidly,reproducedattheendof
TheonsetofthereproductiveseasonintheClamLake thebreedingseason,andlaidalownumberofeggs. These
populationofG.bosci,determinedbychangesinGSIand same individuals then had the opportunity for further
AGSlvalues,andovumdiameters,indicatedthefollowing: growth over the winter with a second opportunity to
(1)thereproductiveseasonextendedfromatleastlateApril produceagreaternumberofeggsintheearlypartofthe
to September, a period of five to six months; (2) the next breeding season. Individuals hatched later in the
reproductiveseasonpossessedbimodalpeaksinspawning reproductiveseasonweretoosmalltoreproducebytheend
activity; (3)thepresenceofmatureandripeningovafrom oftheseasonanddelayedbreedinguntilthefollowingyear.
April to September indicated that breeding occurred These conclusions support Steams and Crandall (1984)
throughout the summer; and (4) individual females andSteamsandKoella(1986),whofoundthatplasticityin
possessedmultiplesizeclassesofovasuggestingthatthey reproductivestrategiesallowsindividualstovarytheirage
might breed more than once during the season, a fact andsizeatfirstreproduction. Ourdataalsoindicatethat
supportedbyfieldexperiments(Conn 1989). timeofhatchingisacriticalfactorindeterminingwhich
Individualsinvolvedinthefirstreproductivepeakin reproductive strategy individual G. bosci employ in
MayweresignificantlylargerthanthoseintheSeptember protractedbreedingseasons.
peak.Theabsenceoflargeindividualsinthelasthalfofthe
summer indicated that the oldest individuals were not Acknowledgments
Livingmorethan 12to 13monthsandweredyingoutby
August. Therefore,atleastsomeindividualsinvolvedin TerryLingandRoyKingarethankedforassistancein
thesecondbreedingpeakhadhatchedoutduringtheearly thefield. JimKiakowoskiandthestaffoftheMcFaddin
part ofthe breeding season, matured rapidly, and were NationalWildlifeRefugearethankedforprovidingaccess
breedingbytheendofthebreedingseason.Amongthese to the refuge. John Sullivan is thanked forreading an
individuals,only40.6%produced eggscapable ofbeing earlierdraftofthismanuscript.
181
,
ConnandBechler
LiteratureCited
Bechler,D.L.,C.H.Conn,andT.S.Ling. 1990. Anovelnest- Hiens.D.C.andF.G.Rabito,Jr. 1986. Spawningperformance
trapforecologicalandbehavioralstudiesofcavity-nesting in North American minnows: direct evidence of the
fishes. JFishBiol36:277-278. occuirenceofmultipleclutchesinthegenusNotropis. J
Cole, K.S. and D.R. Robertson. 1988. Protogyny in the FishBiol28:343-357.
Caribbean reefgoby, Coryphopteruspersonatus: gonad Hoesc,H.D, and R.H. Moore. 1977. FishesoftheGulfof
A&M
ontogenyandsocialinfluencesonsex-change.BullMarSci Mexico,Texas UniversityPiess,CollegeStatiem.327p.
42:317-333. Lahnsteiner,F.,M, Sciwald, and R.A. Patzner. 1992. The
Cole,L.C. 1954. Thepopulationconsequencesoflifehistory seminal vesicles of the male grass goby Zosterisessor
phenomena. QuartRevBiol29:103-137. ophiocephalus(Tclcostei,Gobiidae). Finestructure and
Conn,C.H. 1989. ThereproductivebehaviorofGobiosoma histochemistry.Zoomorphology111:239-248.
hosci^ the naked goby. [Unpublished masters thesis] Miller,P.J- 1984.TheTokologyofGobioidFishes. Pages119-
CrabtEsrleelceep,catRri.toEmn.enoatfnodfsMpBiaidwodnlaoiugnygg,h.sLiat1ems9a8ro2f.UnCOihyvesartssemirotysd,heeTslXl.bsoi5zse0quapi.nadnutsh^e MINTs1Tt5rA3aBtien,gGiIc.nscW.a.n1Pdo9t8tt6asc.taiMncisd.nRi.AtJca.baWdoReoemtfteiorcenPn,rceeedssMs.aFnIinusca.hl,r,Lep5ortnohddoruencl.teiaosne,.
Hypleurochilusgemmatus,Hypsoblennuisionthas(Pisces Nero,StLa.tLe.Col1l9e7g6e.,PaT.h2e36napt.ural history ofthe naked goby,
Blenniidae) andGobiosomabosci (Pisces,Gobiidae) in Gobiosomabosci(Pcrciformes; Gobiidae). [Unpublished
twoSouthCarolinaestuaries. Estuaries5:150-155. masters thesis] DepartmentofBiology, Old Dominion
Dahlberg,M.D.andJ.C.Conyers. 1973. Anecologicalstudyof University,VA.86p.
GobiosomabosciandC.ginsburgi(Pisces,Gobiidae)on Stearns,S.C.andR.E.cirandall. 1984. Plasticityforageand
theGeorgiacoast. FishBulll\:n9~'2Xn, size at sexual maturity: A life-history response to
Fitsimons,JJML andK.Seok. 1989. Theeffectofsize,sex,prior unavoidablestress.In:Potts,G.W.andR.J.Wootton(eds),
tcsidoncy,andpaste^xrioioeontemtoialddtnscbythenaked Fish Reproduction, Strategies and Tactics, p 13-33.
g!±>y,Gobiosormboxi. PtocLouisianaAcadSd52:3542. AcademicPress,NewYork.
Griffith,S.AandD.L.Bechler. 1995. The distribution and Steams,S.C.andJ.C.Koella.1986.Theevolutionofphenotypic
abundanceofthebayanchovy,AnchoamUchiUi,inaSoutheast plasticitymlife-historytraits:predictionofreactionnorms
TexasMarshLakeSystem. GulfResRep9;117-122. forageandsizeatmaturity. Evolution40:893-913.
182