The Auk
A Quarterly
Journalof Ornithology
Vol. 118 No. 1 January2001
The Auk 118(1):1-10, 2001
PERSPECTIVES
DELAYED
IN ORNITHOLOGY
DISPERSAL:
LIVING
NEPOTISTIC
UNDER
THE
REIGN
OF
PARENTS
JAN EKMAN,• VITTORIOBAGLIONE,S•SNKEEGGERS,
AND MICHAEL GRIESSER
Evolutionary
BiologyCentre,Department
of Population
Biology,UppsalaUniversity,
Norbyviigen18D, SE-75236 Uppsala,Sweden
THE INCLUSIVEFITNESSconcept (Hamilton
1964)formulatedconsequences
of socialbehavior in geneequivalents.In doing so,it enabled
fitnessconsequences
of socialbehaviorto be
understood
within theframeworkof geneticinheritanceof traits. Delayeddispersalof birds
wasonesystemwheretheinclusivefitnessconceptwasput to test.The key issuewas to understandhow delayeddispersalcouldbe reconciled with evolution through natural
selection,when retainedoffspringforegopersonal reproductionwhile they remain in the
natal territory (e.g.Skutch1961).
Cooperativebreeding seemsto have a secondary role for the maintenanceof delayed
dispersal,although96% of bird specieswhere
the offspringremain with their parentsinto
adulthoodto form family groupsalso breed
cooperatively(Em!en 1995). Although that
associationbetweendelayeddispersaland cooperativebreedingindicatesthat delayeddispersal is a permissivefactor for the maintenance of cooperativebreeding,there is not
necessarilya causationgoingin the opposite
observationthat dispersal can be delayed
withoutthe retainedoffspringengagingin reproduction.Even if someof the retainedoffspring in a speciesparticipatein cooperatively
breedingunits,thereareusuallya substantial
fractionof themthat do not engagein help-atthe-nest, and only a few specieslike the
White-wingedChough(Corcorax
melanorhamphus)can be classifiedas an obligatecooperative breeder (Brown 1987). Even stronger
support for the fact that delayed dispersal
does not require any involvement of retained
offspringin cooperativebreedingcomesfrom
a numberof specieswhereretainedoffspring,
as a rule, never help (Gayou 1986, Veltman
1989, Birkhead 1991, Ekman et al. 1994, Newton et al. 1994, Walls and Kenward 1996, Green
and Cockburn 1999, Robinson2000). Therefore,it seemsthat the maintenanceof delayed
dispersalrequires an explanationthat does
not haveto resortto fitnessgainsof cooperative breeding (see also Hatchwell and Komdeur 2000).
direction.Cooperativebreeding canbe seenas
an independentdecision,and as suchit is a
consequence
rather than a causeof delayed
UNRESOLVED
ISSUES
Thefactthatcooperative
breedingshouldnot
dispersal(Brown 1987,Staceyand Ligon 1987, be essentialfor delayeddispersalis consistent
Koeniget al. 1992,Emlen 1994,Hatchwelland with theviewthatthebehaviorof remainingin
Komdeur 2000), which is consistentwith the the natal territory is maintainedasa productof
ecological constraintson dispersal options.
Constraints
E-mail: [email protected]
can come in different forms such as
lack of mates (Rowley 1981, Pruett-Jonesand
2
Perspectives
in Ornithology
[Auk,Vol. 118
Lewis 1990),or high risksinvolvedwith dis- persal(e.g.Brown1987,Emlen1997),whereas
persal (Emlen 1982). Most attentionhas been a criticaltestby Popperianphilosophy
should
focused on a constraint in limited
access to
aim for systemsthat falsify the hypothesisto
habitat in territorial species(Selander1964, resolvewhy thehypothesisfails(Popper1968).
Brown 1969, Koenigand Pitelka 1981, Emlen It is now well documented that the habitat sat1982, Emlen and Vehrencamp1983). Popula- urationmodelfailsto predictthe absence
of detionshaverestrictedranges,and the question layed dispersalin severalcasesincludingtemof an ecologicalconstraintin lack of vacant perate regionsparids (genusParus).Removal
habitat (habitatsaturation)thereforeis reduced experimentsnot only provided compellingevto a matterof thequalityof unoccupied
habitat idencefor habitat saturationin severalparid
(Brown 1969,Koenigand Pitelka 1981,Stacey species(Ekman et al. 1981, Ekman 1989), but
and Ligon 1991). The offspring should gain additionally,their socialsystemsharesa suite
from dispersingonly as long as unoccupied of featureswith speciesthat delay dispersal
habitatis of sufficientlyhigh qualityto be suit- and are cooperativebreeders.Severalof temable and offer better conditions than what can
perate-regiontits live in small, coherent,and
be gainedin the natal territory (Brown1969). sedentarygroups occupyingexclusiveterritoThereis now alsoa growinginsightin therole ries (Goodbody1952, Dixon 1956, 1963, 1965;
of variationin habitatqualityin shapingdis- Weiseand Meyer1979,Ekman1979,1989;Nilspersal strategies.There is tangible evidence sonandSmith1985,Matthysen1990)justasthe
showingthat retainedoffspringpostponedis- overwhelmingmajorityof cooperative
breedpersalwhile theywait for territorialvacancies ers(Emlen1995).The failureto providean exof high habitatquality (Zackand Ligon 1985, planationfor the lack of delayeddispersalin
Komdeur 1992).Still that evidencefor the role the genusParusis a challenge
to theecological
of habitatquality doesnot meanthat delayed constraintsmodel that has rarely been acdispersalis well understood(Heinsohnet al. knowledged(but see Staceyand Ligon 1991,
1990).Despite its apparentexplanatorypower, Koeniget al. 1992),let aloneresolved.
the conceptof an ecologicalconstraintin habitat saturationleavesseveralquestions
awaiting
WHERE TO WAIT
to be resolved,including (1) How can one accountfor the absenceof delayeddispersalin
Roleof parents.--Thetrade-offnatureof disspeciesliving in saturatedhabitats?(2) Where persaldecisions,
whichwasimplicitin Brown's
to wait for a vacancy,that is, why foregodis- (1969) criterionthat dispersaloptionshave to
persalratherthanto disperseandqueuefor va- be "suitable,"is now widely recognized(e.g.
cancies
in a territoryof higherquality?(3)How Staceyand Ligon 1991,Emlen 1994).Dispersal
canoneexplainthat delayeddispersalis found is governednot onlyby constraintson accessto
predominantlyamongspecieswith a life-his- resourceselsewhere,but shouldbe delayed
tory characterized
by low adult mortality,low whenthe natalterritoryoffersbenefitsthat eifecundity,and deferredmaturity?
ther arehigherthan,or cannotbe gained,elsewhere(benefits
ofphilopatry;
Stacey
andLigon
DISPERSAL IN A SATURATED HABITAT
1991). Althoughidentificationof the role of
habitatquality for the timing of dispersalrecIt is not clear how habitat saturation can acognizesthe trade-offnatureof dispersaldecicountfor the lack of delayeddispersalin spe- sions,its focuson habitatqualityalonemaybe
cies,whichapparentlyexperience
anecological toosimplistic.Variationin habitatqualityis not
constraintin accessing
habitatthat is assevere a featurethatis uniqueto species
with delayed
asin species
with cooperative
breeding(Brown dispersaland cooperativebreeding.Further1969, Staceyand Ligon 1991, Koenig et al. more, a constrainton accessto high quality
1992).That lack of generalityin the ecological habitatelsewheredoesnot precludetherealso
constraints
approachwasrecognizedbyBrown being other reasonsfor postponingdispersal
(1969).Still, studiesof the role of habitat satu- where the benefits cannot be coined in terms of
ration with respectto the timing of dispersal habitatquality.
havebeenconfinedalmostexclusively
to studRemainingin thenatalterritory,asa rule,enies of cooperativebreederswith delayeddis- tailsthat the offspringassociate
with relatives,
January2001]
Perspectives
in Ornithology
and parentsin particular(Skutch1961,Brown
1987,Emlen1995).Offspringaretheevolutionary currencyof theirparents,whichhaveaninterestin promotingthereproductivesuccess
of
their offspring.Therefore,the consequences
of
dispersalare not only an issue for the offspring,but resolutionof the cost-benefittradeoff concerningthe timing of dispersalshould
includethe parentsand their responseto the
presenceof offspringinto adulthood(Ekman
and Rosander 1992, Cockburn 1996). Almost
three decadesafter Trivers (1974) recognised
the parent-offspringconflict,its role for how
familiesresolvethe dispersaldecisionremains
to be explored.Field observation
confirmthat
parentsdohavea rolein the timingof dispersal, mostobviouslywhen parentsenforcedispersal.In somespecieswith delayeddispersal,
suchas the Gray Jay(Perisoreus
canadensis)
and
the GreenJay(Cyanocorax
yncas),the offspring
are evictedby their parentsat the onsetof the
breedingseasontheyearafterhatching(Gayou
1986,Stricklandand Ouellet 1993).Although
retainedoffspringfrom the previousbreeding
seasonare then evicted,it is implied that they
have been tolerated
so far.
Parentalnepotism.--Involvement
of parentsin
promotingdelayeddispersalmustby necessity
bemoresubtlethanenforcingdispersal.It isbiologicallyunrealisticthat the parentsenforce
3
the choiceof evolutionarycurrency.Evolutionary consequences
of living in family groups
have conventionallybeen evaluatedin reproductiveequivalents.However,the potentialof
survival prospectsto accountfor delayed dispersal is apparentwith a multiseasonalapproach,where reproductionis not the solefitness component(Kokko and Johnstone1999).
The model of Kokko and Johnstoneshowsthat
survival prospectsweigh heavily in the dispersaldecision,and that it is theoreticallyfully
conceivable
for theoffspringto benefitfromdelayeddispersalevenif theyshouldnot contribute to reproductionat all. Thatmodelprovides
a theoreticalconfirmationof the suggestion
that delayeddispersaldoesnot require fitness
gains of cooperativebreeding (Brown 1987,
Emlen 1994).
A parentalbehaviorthatenhances
theaccess
to foodfor retainedoffspringshouldhavea direct bearing on starvationrisk, and it is now
knownfor severalspeciesthat parentsdo concedefood resourcesto retainedoffspringin the
nonbreedingseason.Parentsallow their offspringaccessto food that is deniedunrelated
individuals (Verbeek and Butler 1981, Barkan
et al. 1986, Ekman et al. 1994, Pravosudova
1999),but theyalsoprotecttheiroffspringfrom
aggressionby competitorson feedinggrounds
(Scott1980).The parental behaviorof conced-
delayeddispersal.Rather,parentshaveto pro- ing resources
to the offspringqualifiesasneprelamotepostponeddispersalby encouraging
their otism in that it is "favoritism shown...
offspringto remain. Parentsare unlikely to be
able to affect conditions elsewhere,and the en-
tives"
(Webster's International Dictionary,
1976).Nepotismis a behavioralmechanismof
couragement
hasto be basedonprovidingben- kin selection,and all favoritismof kin is nepefitsin thenatal territoryby offeringconditions otism (Sherman1980). What is specialwith
thatarebetterthanwhat theoffspringcangain nepotismis the emphasison the differencein
elsewhere.
Therearedifferentoptionsavailable how kin are treated relative to unrelated indito parentswith respectto how to promotethe viduals, and a nepotisticrelationshipdefines
prospects
of independentbreedingof theiroff- how the parent-offspringrelationshipgoverns
spring.Brown and Brown (1984) emphasized theaccess
to foodin foragingbird flocks.Howthat parentscouldassisttheir offspringin the ever,nepotismmustnot be confinedto food reacquisitionof a breeding territory and repro- sources, but so far there are no studies of
ductive status. They called that mechanism whetheralarm-callingin response
to predators
"parental facilitation."
is nepotisticin birds, like it is known to be in
To becomebreeders,the offspringhave to some mammals (Sherman1977, Cheney and
survivewhile they are waiting for a suitable Seyfarth1990). That opportunityfor the parbreedingopportunity to emerge,and the off- entsto offer benefitsto offspringremainingin
springshoulddelay dispersalif they survive the natal territory remainsto be exploredin
better in the natal territory than elsewhere birds.
(Brown 1978). The potential role of survival
Any evolutionary consequencesof parental
prospects
while queuingfor the timing of dis- nepotism must be derived from fitnessconsepersalhasnot beenfully recognizedbecause
of quences,and there is now evidencein the Si-
4
Perspectives
in Ornithology
[Auk,Vol. 118
berianJay(Perisoreus
infaustus)
suggesting
that is of higherquality.Givenvariationin habitat
an association
with nepotisticparentsenhances quality,why do the offspringnot disperseto
first winter survival (Ekman et al. 2000). Inas- wait for a vacancyin a territoryofbetterhabitat
muchasnepotisticparentalbehaviorpromotes quality?One reasoncouldbe that the parents,
delayeddispersalby enhancingoffspringfit- by definition,canbe foundonlyin thenatalterness,delayeddispersalrepresentsan "extend- ritory. The opportunityfor offspringto assoed parentalinvestment"(Zahavi 1974,Ligon ciatewith nepotisticparentsis a true "benefit
1981, Brown and Brown 1984, Fitzpatrick and of philopatry,"whichcanbe gainedonlyin the
Wolfenden 1986, Ekman and Rosander 1992).
natal territory. If it were not for parentalnepDecouplingqueuingand associating
with the otism,offspringshoulddobetterby shiftingto
parents.--Benefitsof parental nepotism do not a territoryof higherquality,in linewith anideexcludeecological
constraints
suchasa limited al free distribution (Fretwell and Lucas 1970,
access
to high qualityhabitatelsewhere.
Rath- Sutherland1996).Giventhat the qualityof an
er, ecological constraintspromoting delayed alternativeis sufficientlyhigh,dispersalwould
dispersal are reinforced by benefits gained notbe hinderedevenby crowding.Thatpoints
from remainingin the natal territory.Equally to the importanceof the natureof the parentimportant,the conceptof "extendedparental offspringrelationshipfor howthedispersaldeinvestment"canprovidea solutionto why off- cisionis resolved.The presenceof nepotistic
springchooseto remainin thenatalterritory,a parentswouldmakethe differencebetweenthe
questionthat is not resolvedby constraints
on natal territory and other territoriesof higher
independentreproduction.An offspringthat quality as a placeto queue.
respondsto ecologicalconstraintson indepenWithout the benefitof associating
with the
dent reproductionby waiting for a better op- parents,the offspringcouldqueueanywhere,
tion than the ones currently availableoften or float,andthatis exactlywhattheydoin sevdoesso in association
with the parents,like in eral species.Decouplingthe decisionof waitcooperativebreeders.An association
with the ing for a breedingopportunityof high quality
parents could be seen as a by-productof de- from associating
with the parentsallowsthe
layed dispersalbut there is nothing that ne- roleof habitatqualityandnepotismto be evalcessitates
that the offspringwait in the natal uatedindependently.
Sucha decoupling
of the
territory.Evenif the offspringshouldhaveto two decisionscouldresolvethe long-standing
wait for a breedingopportunityto emerge,this problemof accountingfor dispersalof species,
doesnot providean answerto the questionof although their habitat is saturated (Brown
whereto wait. Rather,the responseto a lackof 1969).Thekeydifferencein thesocialbehavior
breeding opportunities is composed of two of parids, in contrastto specieswith delayed
separatedecisions:(1) whether to postpone dispersaland living in family groups,is not
personal reproduction and wait for a "suit- foundin theresponse
to habitatsaturation
per
able"breedingopportunityof sufficientlyhigh se,but in the association
to theparents.Wherequality,and (2) whetherto associate
with the as temperateregionparids do wait for a vacanparentswhile waitingfor a breedingopportu- cy as subordinateflock members(Ekman et al.
nity. Delayeddispersalthereforesuggests
that 1981,Ekman 1990,Hogstad1987),they do not
an association
with the parentshas a valuein form flockswith their parents(Ekman 1989,
itselfto theoffspring.Thefactthat"home"has Matthysen1990),unlike s ,eciesthat delaydisa specialstatushas not traditionallybeenin- -persal(Brown1987,Emlen1995).It is therefore
corporatedin modelsof the evolutionof dis- possiblethat the explanationfor why parids
persal. Recently, Kokko and Lundberg (in neither delay dispersal, nor form family
press)haveshownthatthisfactorstronglyen- groups, is to be found in parent-offspringrehancesthe prospectsfor delayeddispersal.
lationships.Then why shouldparidsnot have
Where an ecologicalconstraint(e.g.lack of an incentiveto associate
with their parents?Is
high quality habitat) fails to accountfor de- it so that parentscouldbe nepotisticin some
layed dispersal,parental nepotism is able to speciesbut not in others?
providean answerto why theoffspringshould
Thelogicof parentalconcession.--The
notion
wait in the natalterritoryratherthan disperse that delayeddispersalhasan elementof parento queuein anotherterritorywherethehabitat tal careemphasizes
a cooperative
relationship,
January2001]
Perspectives
in Ornithology
5
in contrastto a competition-orientedview of
To explorehow diminishingreturnsfrom
parent-offspring relationships(Emlen 1982, food resources
exactedin survivalprospects
EmlenandVehrencamp
1983).Parents
provide couldpromoteparentalnepotismin concession
EkmanandRosander(1992)
benefits,
whichpromptthe offspringto post- of foodresources,
ponedispersal,and the decisionto delaydispersalwould be decoupledfrom any participationof the offspringin reproduction.
That
viewof delayeddispersalasa formof parental
theoretically
analyzed
a situation
wherea parent and offspringhaveto survivefrom a commonandlimitedresource.
An extendedparental
investment
in
concession of resources
(called"prolongedbroodcare"by Ekmanet al.
operativebreedingis not a necessary
condition 1994)is expressedasparentalnepotism,which
that
for delayeddispersal.Furthermore,it is consis- allow the offspringto consumeresources
tentwith observations
thatdispersal
canbede- the parentswould have been able to control.
The asymmetryin fitnesscostsand benefits
layedalthough
theoffspring
donothelp.
betweenparents
An extendedparentalinvestmentisbasedon fromsucha resource-transfer
and
offspring
can
be
illustrated
by thebiologthefactthatit isin theinterest
oftheoffspring
ically
realistic
assumption
that
parents,
being
to remainin thenatalterritoryandtheparents
foragers,canachievea lower
permit it. However, the concessionof food moreexperienced
shows
thatparentsdomorethansimplypermit risk of starvation(highersurvival)than inexit. Theyaugment
conditions
fortheiroffspring periencedoffspring from the same resource
An alternativeapproachwith simandactivelypromotepostponement
of dispers- abundance.
ilar
consequences
wouldbe thatparentsarein
al. What is good for the survivalof the offcontrol of a major share of resources.Given
care is consistent with the conclusion that co-
springis alsoin the interestof the parents,
in survivalprospects,
thereis
because surviving offspring can produce that difference
scopefor parents to make an inclusivefitness
grand-offspring.The parents will therefore
gainjust from that retainedoffspringremain gain from concedingresourcesthey controlto
the offspring(Fig.1A), andthe enhanced
offalive. Still, a nepotisticbehaviorhas to be recspring survival would promote delayed
onciledwith the fact that parentsvaluetheir dispersal.
own survival and reproductionhigher than
thatof theiroffspring(Trivers1974).
LIFE-HISTORY AND DELAYED DISPERSAL
Tradingparentandoffspring
survival.--Toresolvewhatis seemingly
a conflictofinterestbetweenhowparents
valuetheirownwell-being Delayeddispersalis moreprevalentin spe-
cieswithlowadultmortality,
lowreproductive
against
thatoftheiroffspring,
it isimportant
to rates,
and deferred maturation. That link be-
realizetheprobabilistic
natureof survival.Surtodelay
vival is nevercertain,but the probabilityof es- tweenthelifehistoryanda disposition
dispersal
hasbeensuggested
(Brown1987)and
capingstarvationincreaseswith accessto food
the empiricalsupportfor thathypothesis
has
and energyreserves(McNamara and Houston
by phylogenetic
analyses(Ar1990). Therefore, behavior will be related to beenreinforced
noldandOwens1998).Hence,it mustbe possurvivalprospectsin a quantitativeway. The
sibleto reconcileany explanation
for delayed
upperlimit to survivalprospectsentailsa non- dispersalwith that correlation.However,delinearitywherethemagnitude
of fitnessgains scriptivepatternsof comparativeanalyseswill
in enhancedprobabilityof survivalfrom re- not in themselves establish causation. Furthersources
leveloff asit approaches
unity.Theop- more,it is not necessarilyobviousthat life-hisportunityfor parentsto gain from conceding torycharacteristics,
likelowreproductive
rates,
doesnot dependon the levelof resourceabun- should result in a habitat saturation that is
danceor survivalassuch,but it is a productof moresevere
thanfor species
with highreprothechangein survivalprospects
with resourc- ductiveratesandwithoutdelayeddispersalas
es. The slope of this function is, however, with,forinstance,
several
temperate-region
Palinkedto absolute
valuesfor survivalprobabil- russpecies
with clutchsizesin therangeof8 to
ities(Fig.1). Therefore
it is parentswith high 10 eggs(Bent 1946,Cramp and Perrins1993).
survival prospectsthat can gain from con- Still, any explanatorymodelmust producea
ceding.
distributionof delayeddispersalamongtaxa,
6
Perspectives
in Ornithology
[Auk,Vol. 118
Parent
1.0
Average
per
capita
abundance----[
A
Parent
survival
Offspring
0.9
cost from concession
0.8
0.7
Offspringsurvival
0.6
gain from concession
0.4
0.3
0.2
0.1
0.0
.,•
.•,.
conceded
received
by parent
by offspring
Resource
FIG.1. The asymmetryin survivalconsequences
from parentalnepotismin concession
of foodresources
to an inexperiencedor subordinateoffspring.The fitnessfunctionsfor parentsand offspringas a function
of resourceabundanceis expressedasthe probabilityof escapingstarvationby havingmoreenergyreserves
than a critical survival cutoff, assumingthat food encounteris a Poissonprocess(=survival). The amount
of resourcesthat are concededby parentsequalsamountreceivedby the offspring.Noticethat offspring
benefitshave to be devaluedby degreeof relatedness(r; here 0.5) to give the inclusivefitnessgain from
concession
to parents.The effectof resourceabundanceis illustratedby two examples:(A) high abundance,
and (B) low abundance.(A) Concessionof foodpaysin inclusivefitness;high food abundancewith potential
for high parental survival prospectsproducesa survival asymmetrywith a gain to offspring(recipient)
whichexceedscostthe parent(donor).(B) No inclusivefitnessgain from concession
of food;low foodabundanceand low parentalsurvivalproducesa survivalasymmetrywherecostto the parent(donor)outweighs
benefitto offspring(recipient).(Modified after Ekman and Rosander1992).
an inclusivefitnessgain: (1) The asymmetry
with survivalgains(to offspring)that exceed
coststo parentswith high survivalprospects
is
reversed at low parental survival prospects
(see Fig. lB). (2) Parentsshouldconcederesourcesonly when their own survival prospectsare sufficientlyhigherthan thoseof the
fromnepotismis availableonlyto parentswith offspring,whereasoffspringsurvivalhasto be
(here0.5)
high survivalprospects.Theorypredictsthat devaluedby thedegreeof relatedness
parentsshouldbe willing to provideprolonged to give the marginal inclusivefitnessgain to
brood care,promotingdelayeddispersal,only parents.
when their own survival prospectsare high
Hence, the theoreticalpredictionfor when
(Fig. 1A). Two factorscurtail the potentialfor parentsshouldbe nepotisticis consistent
with
either directly causedby life-history traits or
not,whichis consistent
with the observedpattern in life-historycorrelates.
Parentalnepotismin concession
of food may
not only providethe offspringwith a survival
benefitthat promotesdelayeddispersal.It is
apparentthat this potential to make a gain
January2001]
Perspectives
in Ornithology
7
Parent
1.0
B
Average per capita
abundance •
Offspring
0.9
0.8
0.7
0.6
0.5
Parent
0.4
0.3
survival
cost from
concession
0.2
Offspringsurvival
gain from concession
0.1
0.0
conceded
received
by parent
by offspring
F•c. 1.
Resource
Continued.
life-history correlates.Furthermore,a causal
relationshipbetweenlife-historyand parental
nepotismwould be able to accountfor why
specieslike parids do not delay dispersal,althoughtheyexperience
a saturatedhabitat.Accessto food is critical to survival prospectsof
parids (Janssonet al. 1981,Brittinghamand
Temple1988),whichfurthermorearelow compared to specieswith delayeddispersal(Arnold and Owens 1998).Therefore,parid parents may lack the incentive for nepotistic
sharingof food, which would entail that the
offspringhavenothingto gainfrom waitingin
association
with their parents.
Theoryshowsthata costin reducedpersonal
survival from concedingfood resourcesis critical to the willingnessof parentsto concederesources(Ekmanand Rosander1992).That cost
canbe outweighed
by theinclusivefitnessgain
to their offspringonly for parentswith high
survivalprospects
(Fig.1).Thesameprediction
has been made with different modelingtechniques (Taylor 1988, McNamara et al. 1994),
suggesting
that the predictionis robustto assumptions.Apart from that theoreticalrobustness,the inherentstrengthof the "prolonged
brood care" model can be summarized as (1)
the predictedspeciesdifference,in the timing
of dispersalin responseto parentbehavior,is
consistentwith how delayeddispersalis correlatedto certainlife-historytraits.(2) Parental
nepotism can be accommodatedin a trade-off
approachto dispersaldecisionsas an intrinsic
benefitwhendispersalrequiresa suitableoption. (3) There is a behavioral mechanismfor
how the parentsare nepotistic.(4) The occurrenceof nepotismcan be testedby comparing
if parents concederesourcesto offspringbut
not to non-related
flock members.
So far, there are few examplesof species
wherenepotisticparentsshareresources
with
retainedoffspringin the nonbreedingseason
(Barkan et al. 1986, Ekman et al. 1994, Pravo-
sudovaet al. 1999).Still, thoseexamplesdemonstratethat the behavioris a reality.Furthermore,thescarcityof examples
hastobe seenin
8
Perspectives
in Ornithology
the light of a bias in interesttowards reproductiveconsequences
of specieswith delayed
dispersaland cooperativebreeding.A nepotistic behaviorin that parentsconcederesources
is a "nonbehavior"characterizedby a lack of
aggressiveness
when accessto food is controlled through despoticbehavior It is a less
conspicuous
behaviorthan the morespectacular cooperative
breeding,andit couldtherefore
easily be overlooked,despiteits potential importanceto the decisionto postponedispersal
and maintain an association
with the parents.
Concession
is basedon the factthat parentsrefrain from claimingtheir priority to resources.
Confirmationof suchnepotismbecomesevident only in a systematiccomparisonof social
relationswithin nonbreedingflocks.
ACKNOWLEDGMENTS
[Auk,Vol. 118
BROWN,J. L., AND E. R. BROWN.1984. Parental facil-
itation:Parent-offspringrelationsin communally breedingbirds. BehavioralEcologyand Sociobiology14:203-209.
CRAMP,S., AND C. M. PERRINS.1993. The Birds of the
WesternPalearctic.
Vol.8, Flycatchers
to Shrikes.
Oxford University Press,Oxford.
CHENEY, D. L., AND R. M. SEYFARTH.1990. How Mon-
keysSeetheWorld.Universityof ChicagoPress,
Chicago.
COCKBURN,
A. 1996. Why do so many Australian
birdscooperate:
Socialevolutionin theCorvida?
Pages451-472 in Frontiersof PopulationEcolo-
gy (R. B. Floyd,A. W. Sheppard,and P. J. DeBarro, Eds.). CSIRO, East Melbourne, Australia.
DIXON,K. L. 1956.Territorialityand survivalin the
Plain Titmouse.
Condor
58:169-182.
DIXON, K. L. 1963. Someaspectsof socialorganiza-
tionin the CarolinaChickadee.Pages240-258in
ProceedingsXIII InternationalOrnithological
CongressVol. 1. (C. G. Sibley,Ed.). AmericanOrnithologists'Union,Washington,D.C.
JacobH6glund, Hanna Kukko, and Steffan Ulf- DIXON, K. L. 1965. Dominance-subordination relastrandread the manuscriptand suggestedmany imtionshipsin Mountain Chickadees.Condor67:
291-299.
provements.The study was supportedby grants
from the Swedish Natural Science Research Council
EKMAN,J. 1979. Coherence,compositionand terri(to J.E.)
toriesof winter socialgroupsof the Willow Tit
Parus montanus and the Crested Tit P. cristatus.
LITERATURE
Ornis
CITED
Scandinavica
10:56-68.
EKMAN,J. 1989.Ecologyof non-breedingsocialsysARNOLD,K. E., ANDI. P. E OWENS.1998. Cooperative
breedingin birds:A comparativetestof the life
historyhypothesis.Proceedings
of theRoyalSociety of LondonSeriesB 265:739-745.
tems of Parus. Wilson Bulletin
101:263-288.
EKMAN,J. 1990. Alliances in winter flocks of Willow
Tits; effects of rank on survival and reproductive success in male-female associations. Behav-
ioural Ecologyand Sociobiology
26:239-245.
BARKAN, C. P. L., J. L. CRAIG, S. D. STRAHL,A.M.
STEWART,AND J. L. BROWN.1986. Social domi-
EKMAN, J., J. CEDERHOLM,AND C. ASKENMO. 1981.
nancein communalMexican JaysAphelocoma
ul-
Spacingand survivalin winter groupsof Willow
tramarina. Animal
Tit Parus montanus and Crested
Behaviour 34:175-187.
BENT,A. C. 1946. Life Histories of American Jays,
Crows and Titmice.
Part 2. Smithsonian
Insti-
tute, United States National Museum Bulletin
191.
BIRKHEAD,
T. R. 1991.TheMagpies.TheEcologyand
Behaviour
of
Black-Billed
and
Yellow-Billed
Magpies.Poyser,London.
BRITTINGHAM, M. C., AND S. A. TEMPLE. 1988. Im-
pactsof supplementalfeedingon survivalrates
of Black-cappedChickadees.Ecology69:581589.
BROWN,J. L. 1969.Territorial behaviorand population regulationin birds.WilsonBulletin81:293329.
Tit Parus crista-
tus--A removal study. Journalof Animal Ecology 50:1-9.
EKMAN, J., AND g. ROSANDER.1992. Survival en-
hancementthroughfood sharing:A meansfor
parentalcontrolof natal dispersal.Theoretical
PopulationBiology42:117-129.
EKMAN, J., B. SKLEPKOVYCH,AND H. TEGELSTR•)M.
1994. Offspring retention in the SiberianJay
(Perisoreus
infaustus):
The prolongedbroodcare
hypothesis.
BehavioralEcology5:245-253.
EKMAN, J., A. BYLIN, AND H. TEGELSTR•)M.2000. Pa-
rental nepotismenhancessurvival of retained
offspringin the SiberianJay.BehavioralEcology
11:416-420.
BROWN,J. L. 1978. Avian communalbreeding sys- EMLEN,S. t. 1982. The evolutionof helping:I. An
ecologicalconstraintsmodel. American Natutems.Annual Review of Ecologyand Systematics 9:123-155.
ralist
119:29-39.
BROWN,
J.L. 1987.Helpingand CommunalBreeding EMLEN,S. t. 1994. Benefits, constraints and the evolutionof the family. Trendsin Ecologyand Evoin Birds. PrincetonUniversity Press,Princeton,
New Yersey.
lution
9:282-285.
January
2001]
Perspectives
in Ornithology
EMLEN,
S.t. 1995.An evolutionary
theoryof thefamily. Proceedings
of the NationalAcademyof Sciences USA
92:8092-8099.
EMLEN,S. T. 1997.Predictingfamily dynamicsin social vertebrates.Pages228-253 in Behavioural
9
productiveskew.Proceedings
of the RoyalSociety in London SeriesB 266:571-578.
KOKKO,H., AND P. LUNDBERG.2001. Dispersal, migration, and offspring retention in saturated
habitats.AmericanNaturalist(in press).
J. 1992.Importanceof habitatsaturation
Ecology,4th ed. (J. R. Krebsand N. B. Davies, KOMOEUR,
Eds.). Blackwell ScientificPublications,Oxford.
EMLEN,S. t., AND S. L. VEHRENCAMP.
1983. Cooperative breeding strategiesamong birds. Pages
93-120 in Perspectives
in Ornithology(A. H.
Bushand G. A. Clark, Eds.).CambridgeUniver-
and territory quality for evolutionof cooperative
breedingin the Seychelles
Warbler.Nature 358:
492-495.
LIGON,J. D. 1981.Demographicpatternsand communal breeding in the Green Woodhoopoe,
Phoeniculus
purpureus.Pages261-280 in Natural
sity Press,Cambridge,United Kingdom.
FITZPATRICK,J. W., AND G. E. WOOLFENDEN. 1986.
Selection and Social Behavior: Recent Research
Demographicroutesto cooperativebreedingin
andNew Theory(R. D. AlexanderandD. W.Tinkle, Eds.).Chiron,New York.
somenew world jays. Pages137-160 in Evolution of Animal Behavior(M. H. Nitecki and J.A. MATTHYSEN,E. 1990. Non-breeding social organiKit&ell, Eds.). Oxford University Press,
zationin Parus.CurrentOrnithology7:209-249.
MCNAMARA, J. M., AND A. I. HOUSTON.1990. The
Oxford..
FRETWELL, S. D., AND n. L. LUCAS. 1970. On territo-
value of fat reserves and the trade-off
between
rial behaviorand otherfactorsinfluencinghab-
starvationand predation.ActaBiotheoretica
38:
itat distribution
37-61.
in birds.
I. Theoretical
Devel-
MCNAMARA, J. M., A. I. HOUSTON,AND J. W. WEBB.
opment.ActaBiotheoretica
19:16-36.
GAYOU,D.C. 1986.The socialsystemof the Texas
1994.Dynamickin selection.
Proceedings
of the
Green Jay.Auk 103:540-547.
RoyalSocietyof LondonSeriesB 258:23-28.
GOODBODY,
I. M. 1952.The post-fiedgingdispersal NEWTON, I., P. E. DAVIES, AND D. MOSS. 1994. Philof juveniletitmice.BritishBirds45:279-285.
oparryand populationgrowthof RedKitesMilGREEN,D. J., AND A. COCKBURN.
1999.Life history
vusmilvusin Wales.Proceedings
of theRoyalSociety of London SeriesB 257:313-323.
and demographyof an uncooperativeAustralian passerine,
the BrownThornbill.Australian NILSSON,
J.•., ANDH. SMITH.
1985.Earlyfledgling
Journalof Zoology 47:633-649.
mortalityand the timingof juveniledispersalin
Marsh Tit Paruspalustris.
OrnisScandinavica
16:
HAMILTON,
W. D. 1964.Thegeneticalevolutionof so293-298.
cial behaviour.I and II. Journalof TheoreticalBiPRAVOSUDOVA,
E. V. 1999. Forestfragmentationand
ology 7:1-52.
the socialand geneticstructureof a permanentHATCHWELL,
B. J.,ANDJ.KOMDEUR.
2000. Ecological
constraints,
life historytraits and the evolution
of cooperativebreeding.Animal Behaviour59:
1079-1086.
HEINSOHN, R. G., A. COCKBURN,AND R. A. MULDER.
resident bird. Ph.D. dissertation, Ohio State Uni-
versity,Columbus.
POPPER,
K. 1968.Conjecturesand Refutations.Harper and Row, New York.
S. G., AND M. J. LEWIS. 1990. Sex ratio
1990.Aviancooperative
breeding:Old hypoth- PRUETT-JONES,
esesand new directions.Trends in Ecologyand
and habitatlimitationpromotedelayeddispersEvolution
5:403-407.
HOGSTAD, O. 1987. Social rank in winter flocks of
Willow
Tit Parus montanus. Ibis 129:1-9.
JANSSON,C., J. EKMAN, AND A. VON BRC)MSSEN.
1981.
Winter mortality and food supply in Parusspp.
Oikos
37:313-322.
al in SuperbFairy-wrens.Nature 348:541-542.
ROBINSON,
T. R. 2000. Factorsaffectingnatal dispersal by Song Wrens (Cyphorhinus
phaeocephalus):
Ecologicalconstraintsand demography.Ph.D.
dissertation,University of Illinois, Urbana.
ROWLEY,
I. 1981.The communalway of life in the
KOENIG,W. D., AND E A. PITELKA.1981. Ecological
SplendidWren,Malurussplendens.
Zeitschriftffir
factors and kin selection in the evolution of coTierpsychologie
55:228-267.
operative breeding in birds. Pages261-280 in SCOTT,D. K. 1980. Functional aspectsof prolonged
Natural Selection and Social Behavior: Recent
parentalcarein Bewick'sSwans.Animal Behaviour 28:938-952.
Research
and New Theory(R. D. Alexanderand
D. W. Tinkle, Eds.).Chiron, New York.
SELANDER,
R. K. 1964.Speciationin wrensof the geKOENIG, W. D., E A. PITELKA, W. J. CARMEN, R. L.
nus Carnpylorhynchus.
University of California
MUMME, AND M. T. STANBACK.1992. The evoPublicationsin Zoology,no. 74.
lution of delayeddispersalin cooperativebreed- SHERMAN,
P.W. 1977.Nepotismand the evolutionof
alarm calls. Science 197:1246-1253.
ers.Quarterly Review of Biology67:111-150.
KOKKO,H., AND R. A. JOHNSTONE.
1999. Socialqueu- SHERMAN,P. W. 1980. The meaning of nepotism.
American
Naturalist
116:604-606.
ing in animal societies:A dynamicmodel of re-
10
Perspectives
in Ornithology
SKUTCH,A. E 1961.Helpersamongbirds. Condor63:
198-226.
[Auk,Vol. 118
TRIVERS,
R. 1974.Parent-offspringconflict.American
Zoologist14:249-264.
STACEY,
P.B., ANDJ.D. LIGON.1987.Territoryquality WALLS, S.S., AND R. E. KENWARD, 1996. Movements
and dispersaloptionsin theAcornWoodpecker,
of radio-taggedBuzzards Buteobuteoin early
life. Ibis 140:561-568.
and a challengeto the habitat-saturation
model
of cooperative breeding. American Naturalist WEISE, C. M., AND J. R. MEYER. 1979. Juvenile dis130:654-676.
persal and developmentof site-fidelity in the
STACEY,P. B., AND J. D. LIGON. 1991. The benefit of
Black-CappedChickadee.Auk 96:40-55.
philopatryhypothesisfor the evolutionof co- MELTMAN,C. J. 1989. Flock, pair and group living
operativebreeding:Variationin territory quality
lifestyleswithoutcooperativebreedingby Ausand group size effects.American Naturalist 137:
tralian Magpies (Gymnorhina
tibicen).Ibis 131:
831-846.
601-608.
STRICKLAND,
D., AND H. OUELLET.1993. Gray Jay
(Perisoreus
canadensis).
In The Birds of North
America,no. 40 (A. Poole,P.Stettenheim,and E
Gill, Eds.).Academyof Natural Sciences,
Philadelphia, and American OrnithologistsUnion,
Washington,D.C.
SUTHERLAND,
W. 1996. From Individuals to Population Ecology.Oxford UniversityPress,Oxford.
VERBEEK,
N. g. M., AND R. W. BUTLER.1981. Cooperativebreedingin the NorthwesternCrow Cor-
TAYLOR, P. D. 1988. An inclusive fitness model for
ZAI4Av•,A. 1974.Communalnestingby the Arabian
dispersalof offspring.Journalof TheoreticalBiology 130:363-378.
vus caurinus in British
Columbia.
Ibis 123:183-
189.
ZACK,S., ANDJ. D. LIGON.1985. Cooperativebreed-
ing in Laniusshrikes.II Maintenanceof groupliving in a non-saturatedhabitat.Auk 102:766773.
Babbler. A case of individual
84-87.
selection.
Ibis 116: