Dietary Preferences of Captive Eastern Long-necked Turtles 'Chelodina Longicollis'

2007·1.2 MB·English

by  Catherine E, Reid Meathrel

Contributions
Dietary preferences ofcaptive
,
Eastern Long-necked Turtles Chelodina longicollis
CatherineEMeathrel and Sharon Reid
DepartmentofEnvironmentalManagementandEcology,LaTrobeUniversity,
Wodonga,[email protected]
Abstract
Fifteen adult Eastern Long-necked Turtles Chelodina longicollis (13 males, 2 females) were
removedfrom billabongs locatedwithintheOvensRiverfloodplain, Victoria,duringthesummerof
1997/98 to examine their feeding preference in captivity, away from competitors, in a controlled
environment. Fourprey items werepresented to thecaptiveturtles: naucorids (FamilyNaucoridae),
corixids (Family Corixidae), gudgeons (Hypseleotris klunzingeri) and caddisfly larvae (Family
Leptoceridae). In total, 12.9% ofpreywas consumedwithout any dietary preference shown bytur-
tles. Itwas concluded thatturtles in captivity fedpassively and opportunistically. Clearly, this may
differ from their foraging strategy in the wild where the anti-predator behaviour ofprey can differ
greatlyinthepresenceofrefugiainamuchlargerandmorecomplexhabitat. (The VictorianNaturalist
124(3),2007, 163-166).
Introduction Methods
Eastern Long-necked Turtles Chelodina Determination ofthe prey preference in
longicollis. Broad-shelled River Turtles C. captivity of Chelodina longicollis formed
expansa and Murray Turtles Emydura part ofa larger study on the habitat prefer-
macquarii in the Murray-Darling Basin ence and feeding ecology of freshwater
exhibit habitat overlap and compete for turtles in billabongs in northeast Victoria
prey. This is the case especially with the (see Meathrel etal. 2004). Permission was
carnivorous, long-necked species, Chelo- obtained from the Victorian Department of
dina longicollis and C. expansa (see Natural Resources and Environment (per-
Meathrel et al. 2002, 2004 and references mit RP-97-170) and La Trobe University
therein). The sympatric Murray Turtle has Animal ExperimentationEthicsCommittee
a shorter neck, is omnivorous, grazes on (permit LSB96/24/V2) to place into short-
periphyton, scavenges carrion and supple- term captivity (i.e. a period not to exceed
ments its diet with aquatic insects and fish three weeks) 15 adults ofthe most com-
carrion to meet metabolic requirements monly encountered species Chelodina
(Chessman 1986; Spencer et al 1998). longicollis [13 malesand2 females, sexing
Chelodina longicollis uses highly produc- followed Chessman (1978)] from the
tive, ephemeral water bodies more often Ovens River floodplain (146°14’44” E,
than the other two species (Chessman 36°14’05”S) between October 1997 and
1988), thereby lessening inter-specific February 1998.
competition(Kennettand Georges 1990). Laboratory housing consisted of 1.2 x 0.6
One ofthe major challenges for wildlife x 0.6 m fish tanks held at 22°C (i.e. the
ecologists is understanding how wild average temperature ofwater in the bill-
organisms perceive their environment and abongs, unpublished data) and filled to 2/3
howthatenvironmentmight bemeasured- capacity and containing charcoal and fiber-
the abundance and availability ofprey, the glass waterfiltersto maintainwaterquality.
level ofinter- and intra-specific competi- No substratum (i.e. gravel, vegetation) was
tion, etc. For freshwater turtles, the included in the tanks as this facilitated the
research relating habitat complexity with quantification ofprey items remaining fol-
habitat selection and dietary partitioning is lowing each trial. All turtles were in good
scarce. This study, therefore, examined the health and allowed to habituate in individ-
dietary preferences of C. longicollis under ualtanksfor 1 weekpriortothecommence-
controlled laboratory conditions away from ment offeeding trials. They received food
variableenvironmentsandcompetition. once every 24 hours forthisperiod, but not
within24hoursofthefeedingtrials.
Vol. 124 (3) 2007 163
Contributions
Prey offered were chosen on the basis of each turtle’s two separate feeding trials
that which was common in the stomach (T1i,ii3 = 0.023, p>0.05) and these data
contents ofturtles examined in the field were combined. Prey preference did not
(Meathrel etat. 2004). Naucorids (Naucor- differ between the various types of prey
idae) were collected from the Doctor’s fed to turtles in the laboratory (F =
Point billabong in Albury, New South 0.532, p>0.05) and suggested that t3hlel315
Wales; whereas corixids (Corixidae), gud- Chelodina longicollis in captivity were
geons (Hypseleotris klunzingeri) and cad- non-selective foragers (Fig. 1). Statistical
disfly larvae (Leptoceridae) were collected comparison ofpreypreference between the
from a man-made lake located on the genders was not warranted because the
Wodonga grounds ofLa Trobe University. data from only two females could not be
The amount of each prey type collected validly tested against that of 13 males.
was variable. Hence, we were unable to However, it appeared that there was little
replicate trialsby presenting equal numbers difference in preypreferencebetween male
ofeach prey type to each turtle for each of and female turtles. Generally, males ate a
theirtwofeedingpreferencetrials. greater proportion ofprey offered, but this
For each experimental feeding trial, each may have been caused by males 4, 9 and
turtle was presented with a range ofprey 10 (Table 1) all ofwhich ate approximate-
items, totalling 40 to 60 individual prey ly 35% ofthe prey offered as compared to
items each trial. The turtles were then just5% fortheotherturtles.
allowed 48 hours to consume prey (a pro-
cedure modified from Serrouya et al. Discussion
1995). After this time the turtles were Long-held ecological principles of habitat
removed and the tanks cleared of any use and foraging strategies suggest that
remaining prey. These prey were counted organisms use optimality decision rules (=
and any missing items were assumed to fitness maximising) to respond to changes
havebeen eatenbytheturtles. in their feeding environment (Pyke et al.
At the completion of the first feeding 1977; Chesson 1983). Changes can arise
trial, turtles were placed back into their from extrinsic, environmental factors such
individual tanks and starved for 24 hours as annual and seasonal variation in prey
to increase their appetite prior to presenta- abundance and availability, exacerbated by
tion ofa second group ofprey (Serrouyaet the presence of competitors for the food
al. 1995). Atthecompletion ofall 30 feed- resource, thereby rendering the study ofan
ing trials all turtles were released at their organism’s preferred diet difficult. Hence,
point ofcapture. No turtlewas held in cap- in this study 15 Chelodina longicollis were
tivity for more than two weeks in order to taken into the laboratory in an attempt to
meet the requirements ofthe animal ethics control for such change in order to simpl-
permit. ify the detection ofthe potential existence
Data were analysed with the statistical ofprey preference in this species offresh-
package SPSS for Windows v. 12 using waterturtle.
univariate ANOVA (Sokal and Rohlf The prey types and their relative propor-
1995) comparing the preference ofthe tur- tions offered to the captive turtles in this
tles for different prey types. As we could study were the same as those gathered for
not offer turtles the same number ofeach a reference collection ofaquatic fauna for
different prey type in both experimental the billabongs at Killawarra (Meathrel et
feeding trials, we converted the data to the al. 2004). The presence ofthese types of
proportion of prey eaten out of all prey prey within the guts ofwild turtles merely
offered(i.e. preference). may have represented the preys’ availa-
bility and accessibility ratherthan any spe-
Results cific preference the turtles may have had
Ofthe total 1232 prey items offered to the forthem in thewild. To determinewhether
15 Chelodina longicollis, 159 (12.9%) captive turtles were feeding preferentially
were consumed (Table 1). Preypreference, or opportunistically in this study, all prey
measured as the number eaten out ofthe items were presented to the turtles without
number offered, did not differ between refugia such as snags, macrophytes or tur-
164 The Victorian Naturalist
Contributions
Tdaatbalceo1m.bTihneednufomrbtehreotfwoprteryialosffpeerredinadnivdideuaatle.nFby=cfaepmtailvee,EMast=ermnalLeo.ng-neckedTurtlesincaptivity;
Turtle Preytype Naurcorids Corixids Gudgeons Caddisfly
larvae
F- 1 #preyoffered 26 40 9 20
#preyeaten 0 2 0 0
F-2 #preyoffered 16 40 9 8
#preyeaten 2 1 0 1
M- 1 #preyoffered 25 40 10 24
M #preyeaten 1 1 0 2
-2 #preyoffered 25 40 10 24
#preyeaten 10 5 0 2
M
-3 #preyoffered 15 40 15 9
M #preyeaten 0 1 0 5
-4 #preyoffered 17 40 7 7
M #preyeaten 3 22 0 1
-5 #preyoffered 16 40 7 7
M #preyeaten 1 1 0 0
-6 #preyoffered 16 40 7 7
M #preyeaten 1 3 2 2
-7 #preyoffered 15 40 14 5
M #preyeaten 1 1 0 2
-8 #preyoffered 16 40 19 4
#preyeaten 0 5 2 0
M
-9 #preyoffered 20 40 26 9
#preyeaten 10 0 21 2
M
- 10 #preyoffered 20 40 26 10
M #preyeaten 1 11 24 1
- 11 #preyoffered 20 40 21 10
#preyeaten 0 4 0 0
M
- 12 #preyoffered 16 40 9 6
#preyeaten 0 0 2 0
M
- 13 #preyoffered 16 40 8 6
#preyeaten 3 0 0 0
bidwater, andsowerereadilyavailablefor numbers (approx. 5%) ofprey were con-
capture. No prey preferences were shown sumed in the laboratory experiments, with
bythe turtles which suggested that the tur- turtles in some trials consuming nothing at
tles were passive, opportunistic feeders. all. This may have resulted from an insuf-
Chessman (1978, 1984) and Georges etal. ficient habituation time due to permit
(1986) also made this assumption, but restrictions. Although not observed, ifsub-
Onkonburi and Formanowicz (1997) cau- jectto stresstheturtlesmayhaveabstained
tioned that the distinction between passive from eating as they are capable ofwith-
and active prey preference may be altered standing long periods with limited food
by the interaction between forager behav- supplies(KennettandGeorges 1990).
iourandpreyantipredatorbehaviour. Although inferencesofdietarypreference
Meathrel et al. (2004) found that male in captivity can be applied only loosely to
and female turtles in the wild appeared to management of a species’ natural habitat,
consume the same prey items. Given the managers ofAustralia’s freshwaterecosys-
disparate sample sizes between the two tems should be aware that the level of
genders, no definitive statements werewar- ‘pristine’ habitat retained must not only
ranted in this study ofturtles in captivity. meetthe dietary metabolic requirements of
The female turtles brought into captivity species, but their ecological requirements
were approximately 15% larger and 40% as well. In the wild, with refugia and
heavier than males (unpubl. data), and changing environmental conditions, more
therefore should have consumed more than prey need to be retained to ensure that
males. However, the two females con- predators have the opportunity to encoun-
sumed less than the 13 males. In fact, low ter prey at levels sufficient to cover their
Vol. 124 (3) 2007 165
1 . 1 .
Contributions
0.25 - Gender
Female
9I |
Male
u
-
S3 0.20
o
Naurcorids Corixids Gudgeons Caddisflylarvae
Prey type
Fig. 1 Comparison ofmean (± 1 standard error) prey preference for 2 female and 13 male adult
Chelodinalongicollisheldincaptivity.
metabolic requirements. Further research Meathrel CE, SuterPJ and RadfordNM (2002)Niche
on the dietary preferences ofAustralia’s segregationbetweenthreespeciesoffreshwaterturtle
in a large billabong during flood. The Victorian
freshwater turtles needs to explore how Naturalist1 9, 160-173
prey preference may change over many Meathrel CE, Suter PJ, Reid S (2004) Habitat and
seasonsandyears. dbiieltlaarbyonpgrsefoenrentcheesOovfefnrsesRhiwvaetre,rntourrttlhe-seainstepVhicetmoerriaal.
TheVictorianNaturalist121 4-14.
AWcekanroewilndeedbgteedmteonNtischolasChapmanandLucy OnckhoonibcuerbiyJparenddatFoorsr:maenffoewcit,cozfDprReyJrvu(l1n9e9r7ab)ilPirteyy.
wWoirdkd.up fortheir invaluable help with the field- PyEktehoGlHog,yPEuclollioagmy>H&RE,voClhuatrionnov9,E1L9-2(51.977) Optimal
foraging: a selective review oftheory and tests.
References QuarterlyReviewofBiology;52, 137-149.
CCChhhEttiCWeeeuhnahmresssleytssstellisdhmmessomus,ea.adranniMiAMannnouuBaBsnsmrtCCBaolarruosCca(ant(qhlg1hy1uii-9(9Uaace817nonVra4i8l9asi)vl)W8tliieieF6lsrlEro(e)sdcn(GyioloSrit,dDlAfyahuoie,yaosVge)twfRiiMtrc)eceta(asthlloT(leeoibefTaraosess.ritutstnacurutha(hddunekUaiideene1nin)Mdpsend-uuse,onbsr5Nfel:7rci:e3fCaksC-hrweyhh5eedee7lsdl8iSthtid.uwuodParrautaetthlte)leeDh.e)r, SSSopecnBabFPkerairyrenan)reoelcdc.odccekeaauhCmrRedpteyaaidtiRmRanngioi-:aevntJsasedRu,tNn-trio,rrtidTenaylevhnRewaeRaz,oirJeonecmYeobhEddcopnrulmirseaPrmfaykrhnoar)ymdgaFnpdsueJluiistMrtoosi(uBlafAec1ro,ls,tZ9glsmo9yeaoH5osW,(cf)luDh(qoPrGmoauBagernrreaiyiatorisp7mAiIsAste3uiDektes.,n1er(tm2a2y1yry2C1,9asp3o9lFo8mg833i-lGp)e4a2yoe1an2mTdg-r(4o.hr3as31re4ah.t9(p9poi9dh.hWriv5aiteHe))--t
Australian WildlifeResearch13 65-69. ,
ChessmanBC(1988)Habitatpref,erencesoffreshwater
turtlesintheMurrayValley,VictoriaandNewSouth
Wales.Australian WildlifeResearch15.485-491
ChessonJ (1983) Theestimation andanalysisofpref-
erence and its relationship to foraging models.
Ecology64 1297-1304. Received28September2006;accepted8March2007
Georges A, N.orris RH and Wensing L (1986) Diet of
the freshwater turtle Chelodina longicollis
(Testudines : Chelidae) from the coastal dune lakes
ofthe Jervis Bay Territory. Australian Wildlife
Research13 301-308.
Kennett RM a,nd Georges A (1990) Habitat utilisation
anditsrelationshiptogrowthandreproductionofthe
eastern long-necked turtle, Chelodina longicollis
(Testudinata Chelidae), from Australia.
Herpetologica46 22-33.
,
166 The Victorian Naturalist