Document 6521740

JournaloJ Animal
Ecology 2007
76,881-887
Why do femalelizards lay their eggsin communalnests?
RAJKUMAR
S . R A D D E R a n dR I C H A R D S H I N E
Schoolof Biological SciencesA08, Universityof Sydney,NSW 2006, Australia
Summary
l. In many reptile species,femalesoviposit communally (i.e.many clutchesare laid
within the samenest).This behaviourmight resultfrom constraint(scarcityof nest-sites
offeringsuitableincubationconditions)or adaptation(directfitnessbenefitsaccruing
from the proximity of other eggs).
2. To testbetweenthesealternatives,
we gatheredfield and laboratorydataon montane
scincidlizardsBassianaduperreyifrom south-eastern
Australia.Our data supportthe
adaptationist
hypothesis.
3. In the field, communalvs.solitaryclutcheswerelaid in similar sites,and the relative
frequencyof communalnestingwasnot predictablef,romnest-siteavailability.Thermal
regimesfor incubationdid not differ betweencommunalvs.solitarynests,nor between
eggsat the corevs.peripheryof a communalnest.In the laboratory,femalesselectively
ovipositedbesideexistingeggsratherthan in otherwiseidenticalpotentialnestingsites.
4. From cycling-temperature
incubation in the laboratory,eggsincubatedwithin a
clusterof other eggstook up lesswater,but producedhatchlingsthat werelargerand
faster-runningthan werehatchlingsfrom eggsincubatedalone.
5. Hydric modificationsof incubationconditionswithin a clusterof tightly packedeggs
thus may providea direct fitnessbenefltto communaloviposition.
Key-words:aggregation,Bassianaduperreyi,reproduction,Scincidae,sociality.
Journalof Animal Ecology(2001)76, 881 887
doi: 10.I | 1| I i.1365-2656.2007
.01219.x
Introduction
Attempts to explain the causalmechanismsresponsible
for generatinglile-history diversitygenerallyfall into
one of two main categories:adaptationor constraint
(Williams 1966; Uller 2003; Wilkinson & Gibbons
2005).The former typeof explanationinvokesselective
advantages,
and infersthat the current distribution of
a trait largelyreflectsfitnessbenefits.In contrast,the
'constraint'
approachinfers that phenotypicdiversity
is restrictedby factorssuchasinflexibiedevelopmental
trajectories,physical laws (e.g. allometry of surlace
area-volumerelationships),or environmentalconditions
(e.g. small size due to lood scarcity).Clarifying the
rolesof constraintvs.adaptationin the elaborationof
specifictraits remainsa major challengefor evolutionary
biology (Williams 1966;Congdon & Gibbons 1987;
Amold 1992;Leimar,Van Dooren& Hammerstein2004).
Life-historyvariableshaveattractedparticular ilterest
in this respect,because
of their tight linkagewith reproO 2007 The Authors,
Journalcompilation
O 2007British
EcologicalSociety
Correspondence:
RichardShine,Schoolof Biological
Sciences
A08, Universityof Sydney,
NSW 2006,Australia.
Tel.:+61293513772.Fax:+61293515609.
E-mail:ncs@bio.usyd.edu.au
ductivesuccessand thus,organismalfitness(Roff 2002).
That linkagefacilitatesempiricalmeasurementof the
fitnessconsequences
of the trait in question(Endler1986).
In many cases,interspecificand/or intraspecificvariation
in life-historytraits is bestexplainedby a combination
of adaptationistand constrainthypotheses.
For example, the upper limit lor total reproductiveoutput of a
femalereptile may be set by the maternal abdominal
volume availableto hold the clutch (a constraint),but
adaptationmay regulatereproductiveinvestmentbelow
this upper boundary becauseof selectivelorcesrelating
to aquaticor arboreallocomotion (Vitt & Congdon
1978; Shine 1992;Pizzattq Almeida-Santos& Shine
2007). Within the overall reproductiveallocation, a
trade-off betweenclutch sizeand offspring sizemakes
it impossibleto simultaneouslyoptimize both variables;
and thus,adaptationmay optimizeone(suchasoffspring
size)suchthat variationin the other (in this case,clutch
size)is driven largely by constraint (i.e. finite energy
resourcesor maternal abdominal volume: Sinervo&
Licht 1991;Brown & Shine 2007). Similarly, lessthan-annualreproductionby femalereptilesplausibly
reflects a combination of adaptation (to delay
reproduction until a large clutch can be produced)and
environmentalconstraints(short summer seasonsin
882
R. S. Radder&
R. Shine
O 2007 The Authors.
Joumal compilation
O 2007British
EcologicalSociety,
Journal of Animal
Ecology,76,
88r-887
cold climates:Bull & Shine1979;Bonnet,Naulleau&
Mauget 1992;Naulleau & Bonnet 1996).
The challengesof disentanglingthe rolesof adaptation
vs. constraintare evengreaterlor simplebehavioural
variables.Functional correlatesof complexmorphological or physiologicaltraits imply a fitness-driven
processlor their acquisition(Gans 1988;Blackburn
1992;Rose& Lauder 1996).In contrast,it is more
difficult to support an adaptationistinterpretationfor
behaviouraltraits that areflexiblewithin the lifetimeof
a single individual. The obvious alternativeis that
organismsrespondflexibly to local conditions for
example,to lood availabilityor predatorthreat - and
hence,observedpatternsof dietsor activity levelsare
driven by a mixture of adaptive (geneticallycoded,
fitness-enhancing)
factors and,/ortemporallyor spatially
variablelocal conditions(constraints).
In this paper,we explorethe rolesof adaptationand
constraint with respectto a widespreadbut poorly
understoodbehaviouraltrait: the tendencyof female
reptilesto aggregatefor oviposition,such that a high
proportion of eggswithin a singlepopulation are laid
within a few communalnestsrather than many single
ones (Graves& Duvall 1995;Bonfiglio, Balestrin&
Cappellari2004;Oda 2004;Doody 2006).The phylogenetic distribution of this trait suggestsmultiple
evolutionary shifts both in the occurrenceand the
relativelrequency of communal oviposition (Graves&
Duvall 1995).Clearly, there are two different (but
potentialreasonsfor femalesto oviposit
nonexclusive)
communally:'constraint'[nestsproviding optimal
incubation(thermal, hydric, antipredator)conditions
lor eggsare scarce,so many femalesare forced to use
the samesite]or 'adaptation'(thereis a fitnessbenefit
to eggslaid within suchlargeclusters,or to femalesthat
aggregatefor this purpose:Blouin-Demers,Weatherhead
& Row 2004;James& Henderson2004).
We can testbetweenthesetwo hypotheses
in several
ways.For example,the 'constraint'hypothesis
predicts
that the lrequencyof communalovipositiondepends
upon nest-siteavailability;thus communal nestsare
expectedto be more common in placesand at times
where potential nest-sitesare scarce.In contrast,the
'adaptation'hypothesispredicts
that (1) femaleswill
activelyselectoviposition sitescontaining previously
laid eggsevenil many empty sitesare available,and (2)
offspringsomehowbenefitfrom communalincubation
(e.g. by metabolic heating from other eggs:Godley
et al. 2001;Ewert & Nelson 2003;Zbinden, Margaritoulis & Arlettaz 2006). We have gathered extensive
field and laboratory data on communally nesting
montanelizardsto test suchpredictions.
Methods
that is widely distributedthroughcool-climatehabitats
in south-easternAustralia (Cogger 2000). In the
Brindabella Range 40 km west of Canberra, in the
AustralianCapitalTerritory (148"50'E,35'21' S),fanales
of this speciesproducea singleclutch of threeto nine
eggseach year during early summer (Shine,Elphick
& Harlow 1997;Shine1999;Shine& Elphick 2001).
Oviposition is concentrated in sun-exposedsites,
typically clearingswithin the eucalypt lorest (Shine,
Barrott & Elphick2002).Both the developmental
rates
and phenotypictraits of hatchlingsare influencedby
incubation temperatues but are insensitiveto substrate
moisture content (Flatt et al. 2001). Eggs are laid
beneath rocks and logs, facilitating the search for
natural nests(Shineet a|.2002).
FIELDMETHODS
Eachyear since1994,we havesearchedclearingsin the
Brindabella Range for Bassiananests,and recorded
the numberof eggsper nest,the depth of eggsbeneath
the soil surface,and the sizesof rocksand logscovering
those nests.Miniature data-loggers(iButtons,Dallas
Semiconductor,Dallas, TX, USA) within each nest
haverecordedthermal profilesof nestsat CoreeFlats
(1050m a.s.l.),Picadilly Circus (1240m), and Mount
Ginini (16 15 m) (Shine2002).For the presentstudy,we
alsoquantifiedthermalregimesat the top, bottom and
middle section of each communal egg cluster, to
explorethe possibilityof metabolicheatingby eggsin
clusters.For analysis,we classifyall nestswith lessthan
10 eggsas 'solitary'(i.e. representinga singleclutch),
althoughsomeof thesepresumablyrepresenttwo small
clutches.All nestswith 10 or more eggsare definitely
communal, becausefield-capturedlemalesnever lay
more than nine eggs.
C O L L E C T I O NA N D I N C U B A T I O NO F E G G S
We collected eggsof Bussiqnadupeteyi from communal
nests in the Brindabella Range (see above) in early
December,when all nestswere < 1 week postlaying.
The eggswerepackedin moist vermiculite( 200 kPa)
and transportedto the University of Sydneywherethey
wereweighed,marked with pencil on the shell and randomly distributedto experimentalgroupsfor incubation.
Eggswereincubatedin vermiculite (-200 kPa). To keep
water availability constant acrosstreatment groups,
we used7 g of vermiculiteper egg for all incubations
( e . g .5, e g g s x T= 3 5 g o f v e r m i c u l i t el 0; e g g s x i = 7 0 g
of vermiculite.
E X P E R I M E N T AP
L R O C E D U R ETSO E X A M I N E
EFFECTS
O F C O M M U N A LO V I P O S I T I O N
ON
EGG HYDRIC EXCHANGE AND HATCHLING
S T U D YS P E C I E S
AND AREA
TRAITS
Bassianaduperreyi (Dumeril & Bibron 1839) is a
medium-sized(to 175mm total length) scincidlizard
Eggs were obtained as described above, and randomly
divided into five treatment groups (1, 5, 10, 15 or 25
883
Advantagesof
communal
oviposition
eggsper group), each with at least two replicates.All
treatment groups wereplaced in a cycling-temperature
incubator with a mean temperat\re of 22 'C and a diel
cycle+ 7 5 'C. We weighedeggsat 10-dayintervals.As
soonashatchlingsemerged,wemeasuredand weighed
them [snout-ventlength (SVL), tail length, and body
massl.Each hatchlingwas then kept in an individual
plastic box (22x13 x 7 cm) with a 12 h light : 12 h
dark photoperiod. An underfloor heating element
provided a thermal gradientfrom 20 to 35 "C within
eachcagelor 8 h per day; ambienttemperaturefell to
20 "C overnight(seeShine& Harlow 1996for details).
Each box contained a sheltersite and a water dish.
The hatchlings were fed on live crickets three times a
week. At 7 days post-hatching, we measured the
young lizards'running speedsat 24 "C on a I -m raceway containing infrared sensorsto quantify elapsed
times(seeShine& Harlow 1996lor details).Following
thesetrials,the hatchlingswerereturnedto the field lor
release.
E X P E R I M E N T APL R O C E D U R ETSO E X A M I N E
MATERNAL
NEST-SITE
SELECTION
To determinewhether communal oviposition reflects
an active preferenceby ovipositing femaleslor sites
alreadycontainingeggs(vs.a consequence
of nest-site
scarcity), we collected 26 gravid B. duperreyifrom the
Brindabellapopulation in November,a few weeksprior
to ovipositionin the wild. Theseanimalsweretransported
to Sydney within 24 h of capture,and maintained
thereafter in individual plastic bins (60 x 40 x 20 cm)
containingmoistsand(100mm deep)and lourpotential
nestingsites.Each suchnestingsiteconsistedof moist
vermiculite120mm deepunderan upside-downplastic
box (12 x72x2.5 cm), with a small doorwayto allow
ingress.We placed25 dummy eggs(empty egg-shells
from the previousyear'segg incubation, cleanedand
then filled with moist cotton-wool to resemblelive eggs)
in one of thesefour nestingsites(randomly selected);
the otherswereempty.Ambient temperaturewaskept
at20'C, but an underfloorheatingelementprovided
a substratethermal gradientfrom 20 to 35'C within
the bin for 8 h per day. Gravid femaleswere fed live
cricketsthrice weeklyand waterwasprovidedad libitum.
Following oviposition, femaleswere returned to the
BrindabellaRangeand released.
Results
F I E L D D A T AO N F R E Q U E N C Y
OF COMMUNAL
NESTING
O 2007 The Authors.
Joumal compilation
O 2007British
EcologicalSociety,
Journal of Animal
Ecology,76,
881 887
We obtaineddetaileddata on 367 natural nestsof B
duperreyiin the BrindabellaRangeover I I years,containing a total of 5407eggs(about 900 clutches).Of
these367 nests,at least234 (64ok)werecommunal (i.e.
contained more than nine eggs; see Fig. 1). Mean
number of eggswas6.04 in solitary nests(range 1 9)
30
a Z J
@
o
-d 1s
E
Z
<a
10 30 50 70 90 110 130 150 170 190
Totalnumber
ol eggsin nest
Fig, l. Frequencydistributionof numbersof eggsfoundin
natural nestsof scincidlizards Bassianaduperreyiin the
Brindabella
Range.Individualclutches
nevercontainmore
thannineeggs(andrarelymorethansix),sothatallnests
with
morethannineeggsmustrepresent
communalclutches.
vs. 18 06 (range10-200) in communalnests;and thus,
more than 80%of all eggswerelound in communalnests.
The 'constraint'hypothesispredictsthat changesin
nest-siteavailability should generatechangesin the
frequency of communal nesting. Thus, we predict
significantyear-to-yearvariationif nest-siteavailabilityat
a given site changes through time; and significant
spatial heterogeneityif nesting areas dilfer in the
numbersof suitablecoveritems.Our study sitesin the
BrindabellaRange fulfil theseconditions:some sites
havemany more exposedrocksthan do others,and the
number of such rocks has changedthrough time as
a result of anthropogenicdisturbance(e.9.vegetation
clearing by hydroelectricity contractors removed
> 90%o
of surfacerocksat onesitein 200I ) and wildfireinducedshifts in vegetationcover and thus,degreeof
shadingof rocks (critical for nesttemperatures:
Shine
et a\.2002). Despite this variation, contingency-table
testsrevealno significantspatialor temporal(annual)
variation in the frequencyof communal nestingfor
testsacrossyearswithin eachof our threemain sites,all
P > 0'05;for comparisonsamongthesesitesin eachof
I 1 years(1995-2005),all P > 0.051.Detailedinspection
ol the data revealedno patternfor increasedfrequency
of communal nestingfollowing reductionin nest-site
availability (e.g.decreasefrom 90% to 8070communal
after anthropogenic rock removal at Coree Flats study
area in 2001), nor for major decreasefollowing
increasednest-siteavailability (from 50% to 42o/ocommunal after reduction of shadingdue to wildfires at the
Picadillystudy site in 2003).
Another reason for nest-siteavailability to vary
might be the numbers of other femaleslaying eggs
within a site, a parameter that varied over a 20-fold
rangeacrossyearswithin our studysites.However,the
number of nestsfound within a siteeachyear did not
correlatewith the proportion of thoseneststhat were
communal(r = 28 sitex year samples,r = 0.06,P = 0.75).
In summary the proportion of neststhat werecommunal
was relativelystabledespitetemporaland spatialvariation in factors that modifv nest-siteavailabilitv.
884
R. S. Radder &
R. Shine
F I E L D D A T A O N P H Y S I C A LA T T R I B U T E SO F
NESTS
All nests were located beneathcover items (usually
rocks, but sometimeslogs) and we measured the
dimensionsof all suchitems.If nest-sites
arescarce(as
predicted by the 'constraint' hypothesis),we would
expect femalesto utilize all opportunities and thus,
larger cover items should attract more clutches(and
thus,eggs).In practice,we found no significantcorrelation betweenthe sizeof a cover item and the number
of eggsbeneathit [overall,n= 284coveritems,multiple
logisticregressionwith singlevs. communalclutch as
the dependent variable, and independentvariables
cover-itemlength : likelihood ralio X2= I .84,d.f. = 1,
P = 0'11(means63.5vs.42.2cm, respectively,);
width,
X2= 1'05,P = 0'31 (means18.3vs. 19.1cm); depth,
X'1= 0'17,P = 0'69 (means10'0vs. 9 9 cm)1.Instead,
communal clutcheswere fitted under evensmall rocks
by extendingdeeper undergroundthan did solitary
nests(overall,comparingsolitaryvs.communal:mean
depthsbelowsoil surlaceof uppermoste gg,X2= 0.96,
P = 0 33;of lowestegg,X2=26'2, P < 0 0001).
450
€
4oo
'6
o
JCU
6
E
o
U
300
250
200
(b) Egg massat hatching
1000
900
E
3 aoo
U
700
F I E L D D A T A O N T H E R M A L A T T R I B U T E SO F
NESTS
Thermal regimesmeasuredinside 110 natural nests
were examinedusing mean nest temperaturesas the
independentvariable,and 'communal vs. solitary' as
the dependentvariable.There was no significant
differencein mean temperaturebetweensinglevs.communal nestsoverall(20.31vs.20.09'C; likelihoodratio
from logisticregression
X2= 0'03,d.f. = 1, P = 0'87)or
separatelywithin each of our three main sites (all
x2<o.80,P>0.39).
Simultaneousmeasurements
of temperaturesat the
top, middle and bottom of the eggclusterinsidenine
communal field nests were unaffected by probe
placement (repeated-measuresANovA with probe
positionasthe factor and trimesterof incubationasthe
repeatedmeasure:probe position, Fr.rr=0.29,P = 0 76;
interactionprobe position vs. time, Fo,,=0.43, P=
0 79).In summary we lound no consistentthermaldifferencesamong or within nestsrelatedto the number
of eggspresent.
EFFECTS
O F C O M M U N A LO V I P O S I T I O N
ON
E G G H Y D R I C E X C H A N G EA N D H A T C H L I N G
TRAITS
@ 2007 The Authors.
Journalcompilation
@ 2007British
EcologicalSociety,
kturnul oJ Animul
Ecology,76,
8 8I - 8 8 7
Eggsincubatedin largergroupstook up lesswaterper
(Fig. 2a; Fo.,= 18.16,
eggduringincubation
P < 0.001)
and thus weighedlessimmediatelyprior to hatching
than did their singly incubatedconspecifics(Fig.2b;
nestedANovA,Ft.t= 4'60, P < 0.04).The communally
incubatedeggsnone the lessproducedlarger hatchlings
(Fig. 2c; nestedlNovl for SVL, Fr.t= 6.03,P < 0.02;
for mass, Ft,t= 4'07, P < 0'05) with significantly
F
2 6
c
c
o
c
j
zs.s
o
z
E r u
24.5
1
5
t
0
1
5
Number
of eggspernest
2
5
Fig.2. Effectsof communalincubationon wateruptake
by eggsand phenotypic
traitsof hatchlingscincidlizards
Bassianaduperreyi.Eggs were incubatedeither singly or
withingroupsof five,I 0, I 5 or 25eggs.Thefigureshowsmean
valuesandassociated
standard
errors.
shortertails relativeto SVL (nestedeNcovewith SVL
as covariate,Fn.,= 4'26, P < 0'05) that tendedto run
more rapidlyalonga 1 m racewaythan did conspecifics
from solitary incubation (nested.lNovr., Fr:=4'07,
P < 0.05).
MATERNAL
NEST-SITE
SELECTION
Captivefemaleslaid 20 of 26 clutchesin the nest-box
containingdummy eggs,out ol four nest-boxes
available
per cage(three of which did not contain such eggs).
Thus,femalesovipositedbesidedummy eggslar more
often than would be expectedbasedon the null hypothesisof equal probabilitiesof utilization of eachbox
( x ' z =3 7 ' 3 ,d . f .= 1 , P < 0 ' 0 0 0 1 ) .
885
Advantagesof
communal
oviposition
@ 2007 The Authors
Journalcompilation
@ 2007 British
EcologicalSociety,
Journal of Animal
Ecology,16,
8 8I - 8 8 7
Discussion
More than 80%of eggslaid by .Rassianaduperreyiinthe
Brindabella Range are depositedwithin communal
clutches (probably more, becauseour criterion for
recognizingcommunal nestswas conservative).This
high incidence reflects active maternal preference:in
the laboratory,nestinglemalesovipositedin sitesthat
alreadycontainedcuesfrom previouslylaid eggs.The
alternativehypothesis- that communal oviposition
resultslrom a scarcity of suitablenest-sites- is not
supportedby our field data, becausethe lrequencyof
communal oviposition was unaffectedby spatialand
temporalvariationin nest-siteavailability.Our laboratory incubationstudiessuggestthat ovipositionbeside
an already-laidclutchmay enhancematernalfitnessby
affectingthe phenotypictraits of hatchlings,possibly
mediatedvia effectsof communalovipositionon water
uptakeby eggs.
Communal egg-layingoccursin a diversearray of
reptile species,takesmany forms, and has evolvedand/
or been lost many times within reptile phylogeny
(Graves& Duvall 1995).Thus, no singlehypothesisis
likely to explain the functional significanceof such a
diverseand labiletrait. For example,metabolicheating
by late-stageembryosmay play a major role in species
that producelargeclutchesof largeeggs(e.g.somespeciesof turtles: Ewert & Nelson 2003; Zbinden et al.
2006)but betrivial for themajority ol squamates
which
(llke B. duperueyl)produce such small eggsthat metabolic heat production is insignificant.Similarly, the
selectiveforces operating on communal oviposition
may differ substantially betweenB duperreyi(mt:Jliple
clutchesintermingled beneathexisting cover items)
comparedwith taxa in whichmany lemalesnestwithin
a small area,but in speciallydug burrowsrather than
under an existingcoveritem, and with no direct physical
contactbetweenclutches(Graves& Duvall 1995).
Given this diversity, the relative importance ol
constraintsvs. adaptiveadvantagesmust differ also.
There are many situationsin which suitablenest sites
are limiting in spaceor time; for example,iguanas
migratelong distancesto areasthat providethermally
(Rand 1972;Bock,
optimal and predator-lreenest-sites
Rand & Burghardt 1985);and suitablehydric conditions may be scarcein arid or seasonallyinundated
habitats(Muth 1980;Brown& Shine2004).Under such
conditions,femalesof some squamatespeciesdelay
ovipositionuntil conditionsaresuitable(Andrews1997)and
hence,limited nest-siteavailability may enlorce communal oviposition in time and spacein reptiles(BlouinDemerset a1.2004;Brown & Shine 2004) as well as in
othertypesof animals(e.g.moorhens,McRae 1996).
However,simplenest-sitescarcitycannotexplainall
casesof communal oviposition:nestingB duperreyi
seekout siteswhereeggshavebeendepositedpreviously.
The same behaviour occurs in snakeslTropidonophis
mairii (Gray, 1841)land may be widespreadin com(Brown& Shine2005).The
munailyovipositingspecies
selective lorces lor such maternal behaviour have
attracted considerablespeculation,involving putative
fitnessbenefitssuchaseasiernestexcavationfor mothers,
predator satiationat the time of hatching,identification of suitableincubation conditionsbasedon nest
choice by more experiencedconspecifics,and social
facilitationof offspringdevelopment(Graves& Duvall
1995;Hayes2000;Marco, Diaz-Paniagua& HidalgoY1la2004;Branchi & Alleva 2006).Our data suggesta
simpler alternative:communal incubation enhances
the body sizeand locomotor speedol offspring that
hatch from thoseeggs.
At first sight, the most likely reason for eggs to
benefit from communal incubation would appearto be
heatingthrough the metabolicoutput of neighbouring
eggs,because( 1) somereptileeggsgeneratesubstantial
(Godley et al. 2001;
heat during late embryogenesis
Ewert & Nelson 2003;Zbinden et a\.2006),and (2) in
B. duperreyi,higher incubation temperaturesaccelerate
embryonic development and produce larger faster
hatchlings(Shine,Elphick & Harlow 1995).Heating
due to the presenceof conspecificsalso has been
identified as a probable lorce for communal nesting in
frogs (Waldman 1982)and mammals (Rhind 2003).
Our data militate against this explanation. First,
measurements
of nesttemperatures
failedto revealthe
predicted shifts in thermal regimes(either in comparisonsbetweencommunallyincubatedvs. solitary eggs,
or between eggs in central vs. peripheral positions
within the communalcluster).Although the proximity
of other eggs increasedmean temperature,the effect
was trivial (< 0 25 'C). Also, hatchingoccurredat the
same time in eggs from each of the incubation
treatments(Radderand Shine,unpubl. data),a result
inconsistent with thermal differentials (given the
strongthermaldependence
of incubationperiod:Shine
et al. 1995).
Why, then, did communal incubation generate
larger, faster hatchlings? Our data on water uptake
show that close apposition to other eggs modifies
hydric exchangebetweenthe egg and its environment,
and thus the effectson hatchlingtraits might well be
mediatedby suchhydric factors.Previousresearchhas
shown that dry incubation substratesreduce water
intakeby embryos,and reducehatchlingsizeasa result
(e.g.Brown & Shine2004;Deeming2004).Remarkably,
however,we lound the opposite relationship between
water uptake and offspring size:communallyincubated
eggstook up lesswater,but thesesmallereggsyielded
larger not smaller hatchlings.Phenotypic traits of
hatchling B. duperreyi are relatively unaffected by
hydricpotentialof the incubationsubstrate(Flatt et al.
2001),so the mechanismsby which proximity of other
eggsinfluenceswateruptakeand hatchlingphenotypes
warrantsmore detailedinvestigation.In the oviparous
lacertidlizard Lacerto schreiberi(Bedriaga,1878),eggs
incubatedwithin a cluster of other eggstake up less
waterthan do solitarilyincubatedeggs,in somehydric
conditionsbut not others(Marco et a|.2004).As in our
886
R. S. Radder&
R. Shine
own study, the magnitude
of this effect was great
enough to affect hatchling phenotypes (Marco et al.
2004). In combination,
communal
oviposition
these results suggest that
often may affect hatchling
viability, and hence impose selection on maternal
nest-site preferences.The sensitivity of developmental
processesto relatively minor modifications ol incubation conditions thus may help to explain an otherwise
puzzling phenomenon: the widespread occurrence of
communal oviposition
in squamate reptiles.
Acknowledgements
We thank M. Elphick for assistance in fieldwork,
laboratory studies and manuscript preparation. J. Thomas
helped with animal collection and husbandry. The
study was funded by the Australian Research Council,
and all procedures were approved by the University of
Sydney Animal Care and Ethics Committee.
References
O 2007 The Authors.
Joumal compilation
@ 2007 British
EcologicalSociety,
kturnal of Animal
Ecology,76,
881 887
Andrews, R.M. (1997) Evolutron of viviparity: variation
betweentwo sceloporinelizardsin the ability to extendegg
retention. Journal of Zoo logy, London, 243, 579 - 595.
Amold. S.J. (1992) Constraints on phenotypic evolution.
AmericunNaturalist,140,585-S107.
Blackburn, D.G. (1992)Convergentevolution of viviparity,
matrotrophy,and specializationsforfetal nutrition in reptiles
and other vertebr^fes.AmericanZoologist,32,313-321.
Blouin-Demers,G., Weatherhead,PJ. & Row, J.R. (2004)
Phenotypicconsequences
of nest-siteselectionin black rat
snakes(Elapheobsoleta).CanadianJournal of Zoology,82,
449-456.
Bock, B.C., Rand, A.S. & Burghardt,G.M. (1985)Seasonal
migration and nesting site fidelity in the green iguana.
Contributionsin Marine Science,27. 435-443.
Bonfiglio,F., Balestnn,R.L. & Cappellari,L.H. (2004)Hemtdactylus mabouia (Tropical house gecko). Herpetological
Reyieyv.35.39l.
Bonnet,X., Naulleau,G. & Mauget,R. (1992)Cyclesexuelde
la femellede Viperaaspis(Reptilia, Viperidae). Importance
des rdserveset aspectsmdtaboliques.Bulletin de la Sociiti
Zoologique de France, ll7, 279-290.
Branchi, I. & Alleva, E. (2006)Communal nesting,an early
socialenrichment,increasesthe adult anxietyJike responseand
shapesthe role of socialcontext in modulating the emotional
behavior. BehsviouralBrain Research.172. 299-306.
Brown,G.P & Shine,R. (2004)Matemal nest-sitechoiceand
offspring fitness in a tropical snake(Tropidonophismairii,
Colubridae).Ecology,85, 1627 1634.
Brown, G.P & Slnne,R. (2005)Nesting snakes(Tropidonophis
mairii, Colubridae) selectivelyoviposit in sitesthat provide
evidenceof previous successfulhatching. CanadianJournal
of Zoology,83, I 134-l 137.
Brown,G.P & Shine,R. (2007)Repeatabilityand heritability
of reproductive traits in free-ranging snakes.Journal of
Evolutionary Bio Iogy, 20, 588- 596.
Bull, JJ. & Shine, R. (1979) Iteroparousanimals that skip
opportunities for reproduction. American N aturalist, ll4,
296-316.
Cogger,H.G. (2000)ReptilesandAmphibiansof Austalia,6th
edn. ReedNew Holland, Sydney.
Congdon, J.D. & Gibbons, JW. (1987)Morphological constraint on egg size: a challengeto optimal egg size theory?
Proceedingsof the National Academy of Scientes of the
United Statesof America, 84, 41 45-41 47.
Deeming, D.C. (2004) Post-hatching phenotypic effects of
incubation on reptiles. Reptilian Incubation. Enyironment,
Evolutionand Behatiour(ed.D.C. Deeming),pp.229-251.
Nottingham University Press,Nottingham.
Doody, J.S. (2006) Communal nesting in the red{hroated
skink, Bassianap latynota. HerpetoJauna,36, 23-24.
Endler, J.A. (1986) Naturul Selection in the Wild. Princeton
University Press,Princeton,NJ.
Ewert,M.A. & Nelson,C.E. (2003)Metabolicheatingof embryos
and sex determination in the American alligator, Alligator
mississippiensis.
Journal of Thermal Biology,28, 159 165.
Flatt, T., Shine, R., Borges-Landaez,PA. & Downes, S.J.
( 2 0 0 1 )P h e n o t y p i cv a r i a t i o ni n a n o v i p a r o u sm o n t a n e
lizard, (Bassiana duperreyi): the effects of thermal and
hydric incubation environments. Biological Journal oJ the
Linnean Society, 74, 339-3 50.
Gans, C. (1988)Adaptation and the form-functionrelation.
American Zoo logist, 28, 681- 697.
Godley, B.J., Broderick, A.C., Downie, J.R., Glen, F.,
Houghton, J.D., Kirkwood, L, Reece,S. & Hays, G.C.
(2001)Thermal conditionsin nestsof loggerheadturtles:
further evidence suggesting female skewed sex ratios of
hatchling production in the Mediterranean. Journal oJ
ExperimentaI Mar ine B io Iogy and E cology, 263, 45- 63.
Graves,B.M. & Duvall, D. (1995)Aggregationof squamate
reptilesassociated
with gestation,oviposition,andparturition.
HerpetologicaI M onographs, 9, 102-1 19.
Hayes,L.D. (2000)To nest communallyor not to nestcommunally: a review of rodent communal nestingand nursing.
Animal Behayiour,59,677-688.
James,A. & Henderson,R.Vl (2004)Communal nestingsite
inthe snakeLiophisju liae. Domirnca, West Indies. Car ibbean
JournaloJ Science,40,263-265.
Leimar, O., Van Dooren, T.J.M. & Hammerstein,P. (2004)
Adaptation and constraintin the evolution of environmental
sex determination. Journal of Theoretical Biology, 227,
561-570.
Marco, A., Diaz-Paniagua, C. & Hidalgo-Vila, J. (2004)
Influenceof eggaggregationand soil moistureon incubation
of flexible-shelled
lacertideggs.CanadianJournalof Zoology,
8 2 ,6 0 - 6 5 .
McRae,S.B.(l 996)Familyvalues:costsandbenefitsof communal
nestingin the moorhen.Animal Behaviour,52,225-245.
Muth, A. (1980) Physiological ecology of desert iguana
(Dipsosaurusdorsclr's)eggs:temperatureand water relations.
Erclogy,6l,1335-1343.
Naulleau,G. & Bonnet,X. (1996)Body condition threshold
for breedingin aviviparoussnake.Oecologia,l0T,301-306.
Oda,WY (2004)Communal egglaying by Gonatodes
hwneralis
(Sauria, Gekkonidae) in Manaus primary and secondary
forestareas.Acta Amazonica,34. 331-332.
P r z z a t t o ,L . , A l m e i d a - S a n t o s ,S . M . & S h i n e , R . ( 2 0 0 7 )
Habitat use, morphology and life-history: morphological
and reproductive adaptations to arboreality in snakes.
Etology,88, 359-366.
Rand, A.S. (l 972)The temperatures
of iguananestsand their
relation to incubation optima and to nesting sites and
season.Herpet ologica, 28, 252, 253.
Rhind, S.G.(2003)Communal nestingin the usuallysolitary
rrrarstpial, Phascogaletapoatafa.Journal oJ Zoology, London,
261,345-351.
Roff, D.A. (2002) Life History Evolution. Sinauer Associates,
Sunderland.MA.
Rose, M.R. & Lauder, G.V (1996) Adaptation.Academic
Press,New York.
Shine, R. (1992) Relativeclutch massand body shapein lizards
and snakes:is reproductive investment constrained or
optimized? Eto lut ion, 46, 828- 833.
Shine,R. (1999)Egg-laying reptiles in cold climates:determinants and consequencesof nesttemperaturesin montane
lizar ds. JournaI of E voIut ionary B iology, 12, 9 1I -9 26.
887
Advantages of
communal
ovipo.tition
@ 2007 The Authors
Journal compilation
@ 2007British
Ecological Society,
Journal of Animul
Ecology,76,
8 81- 8 8 7
Shine, R. (2002) An empirical test of the 'predictability'
hypothesisfor the evolution of viviparity in reptiles. Journal
of Evolutionary Biology, 15, 553- 560.
Shine, R. & Elphick, M.J. (2001) The effect of short-term
weather fluctuationson temperaturesinside lizard nests,
and on the phenotypic traits of hatchling lizards. Biological
Journalof the LinneanSociety,72,555-565.
Shine, R. & Harlow, PS. (1996)Maternal manipulation of
offspring phenotypesvia nest-siteselectionin an oviparous
rcpt|le.Ecology,77,1808 1817.
Shine,R., Elphick, M.J. & Harlow, PS. (1995)Sisterslike it
hot. Nature, 378, 451 452.
Shine,R., Elphick, M.J. & Harlow, P.S.(1997)The influence
of natural incubation environments on the phenotypic
traits of hatchlinglizards.E cology,'78, 2559-2568.
Shine,R., Barrott, E.G. & Elphick, M.J. (2002)Somelike it
hot: effects of forest clearing on nest temperaturesof
montane reptlles.Ecology,83, 2808-2815.
Sinervo,B. & Licht, P. (1991)Proximateconstrainrson the
evolution of egg size, number, and total clutch mass in
lizards.Stience,252. 1300-1302.
Uller, T. (2003)Viviparity as a constrainton sex-ratioevolution.
Evolution. 57. 927-931.
Vitt, L.J. & Congdon, J.D. (1978)Body shape,reproductive
effort, and relative clutch mass in lizards: resolution of a
paradox.AmericanNaturalist,lf2,595 608.
Waldman, B. (1982)Adaptive significance of communal
oviposition in wood frogs (Rana sylvatica). Behavioral
Ecology and Sociobiology, 10, 169-174.
Wilkinson,L.R. &Gbbons, J.W (2005)Patternsof reproductive
allocation: clutch and egg sizevariation in three freshwater
ttxtles. Copeia, 2005, 868-879.
Williams, G.C. (1966) Adaptation and Natural Selection.
Princeton University Press,Princeton, NJ.
Zbinden, J.A., Margaritoulis, D. & Arlettaz, R. (2006)
Metabolic heating in Mediterranean loggerheadsea turtle
clutches. Journal of Experimental Marine Biology and
Ecology,334,I 51-l 57.
Received25 May 2007, accepted29 May 2007
Handlins Editor; Jean Clobert