FACTORS AFFECTING NEST AND SITE FIDELITY

IN ADIRONDACK BARN SWALLOWS

(HIRUNDO RUSTICA)
WILLI^M M. SHmLDS
Department of Environmental
andForestBiology,
StateUniversityof New York,
Collegeof Environmental
Science
andForestry,Syracuse,
New York 13210USA

AI•STRACT.--During 1979-1983,I documented the nataland breedingdispersalof the Barn

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

Swallow(Hirundorustica)in the CranberryLake,New York, regionof the Adirondacksby
followingthe lifetimemovementsandbreedinghistoriesof individuallymarkedbirds.Dur-
ing the study,847 birdswere marked,and the movementsof mostwere followedduring
the seasonthey were marked.About 40%of the adultsand 2% of the nestlingsand juveniles
were resightedor recapturedin two or moreyears.All recoverednestlingshad dispersed
from their natal colonies(mediandistance= 6.37kin). Most breedingadultsremainedfaith-
ful to previouslyusedcolonies,aswell asto clusterswithin colonies,nests,and mates,when
the latter were alive and available.Someindividuals,however,did disperse.More females
than malesdispersedand usuallymovedfarther both within and betweenbreedingseasons.
Malesand, especially,femalesthat had nestedunsuccessfully with a particularmate in a
particularlocationhad a higher probabilityof desertingand dispersingthan did successful
breeders,both within and between breeding seasons.Old nestswere reusedat high fre-
quenciesfor consecutivebreedingattemptsacrossbreedingseasons. After successfulfirst
nests,breederstendedto moveto secondarynestswithin 25 m of their first nests.Received
8 December
1983,accepted
28 March 1984.

POPULATION structure is considered to be a pyrrhonota;Mayhew 1958), and Barn Swallow

major influence on evolution and socialbehav- (Hirundo rustica;Davis and Davis 1936, Mason
ior. Dispersaland demographyare the prime 1953,and in someEuropeanpopulationscited
determinants of population structure. They in Farnet 1945). The general swallow pattern
have important influenceson mating systems is that most adults manifest varying but low
(for review, see Murray 1984), on local pat- levelsof breedingdispersal(definedasthe dis-
terns of relatednessand, hence,on the degree tance moved between consecutive breeding
of cooperationand competition characterizing sites).First-time breedersmanifest greater but
differentkinds of socialinteractions(e.g.Ham- still varying degreesof natal dispersal(defined
ilton 1964;Sherman 1980, 1981;Hoogland 1981, as distancebetween hatching and first breed-
1983), on whether particular casesof apparent ing site). (For more detailed dispersaldefini-
altruism are more likely to be the result of in- tions and rationales, see Greenwood et al. 1979a,
dividual selection,kin selection,reciprocity,or Shields 1983.)
group selection(e.g. Zahavi 1974, Brown 1978, As Freer(1979)noted,althoughreportsabout
Emlen 1978,Woolfendenand Fitzpatrick1978, the degreeof adult site tenacityand juvenile
Ligon and Ligon 1978, Wilson 1980), and on philopatty (i.e. the proportionof birds remain-
any group'stempo and mode of evolution (for ing in a previously used "area" and the actual
reviews, see Wright 1978, Shields 1982). distribution of dispersal distances)abound,
Dispersal has been studied extensively in there are few studiesconcerningthe potential
birds and many other taxa (for reviews, see causesof the observedpatterns in swallows.
Baker 1978, Greenwood 1980, Shields 1982). Studiesof other bird groupsindicate that age
Among swallows, dispersal has been investi- (e.g. Austin 1949), sex (e.g. Greenwood 1980,
gated in the Common House-Martin (Delichon 1983),local habitat or nest-sitestability (e.g.
urbica;see Rheinwald 1975, Bryant 1979), Pur- McNicholl 1975, Freer 1979), and prior breed-
ple Martin (Prognesubis;Allen and Nice 1952), ing experienceat a particularsite or with a par-
Tree Swallow (Tachycinetabicolor;Chapman ticularindividual (e.g.Darley et al. 1977,Nolan
1955),BankSwallow(Ripariariparia;Stoner1936, 1978, Freer 1979, Coulson and Thomas, 1983,
Mead 1979, Freer 1979), Cliff Swallow (Hirundo Dow and Fredga 1983, and, for review, see
780 The Auk 101: 780-789. October 1984
October1984] BarnSwallow
SiteFidelity 781

Rowley 1983) could affectthe degreeand mag- how much this recent changein breeding dispersion
nitude of dispersal. affectsthe behavior we observe.By direct count and
Since 1979, my field associatesand I have estimation, Cranberry Lake itself hosts 300-500
been studyingthe behaviorand ecologyof in- breeding pairs from year to year. We have concen-
trated on four major and many smaller coloniesover
dividually marked populations of the Barn
the years. In the main study area at the biological
Swallow. Because previous studies on this
station (hereafter CLBS), there were 17 pairs in 1979,
species(e.g. Davis and Davis 1936 and Mason 18 in 1980, 11 in 1981, 10 in 1982, and 15 in 1983,
1953 versus the studies cited in Farher 1945) scatteredaround the campus.The secondaryarea at
reported wide variation in dispersal patterns the biological station marina (hereafter Bay City or
among their populations,we wished to docu- BC) hosted20 pairs in 1980, 16 in 1981,22 in 1982,

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

ment dispersalin our area in order to charac- and 16 in 1983, all under the roof of a single boat
terize sufficiently its local population structure slip. We also worked in other large and small colo-
nies around the lake.
as a basisfor other ecologicaland behavioral
studies(e.g.Shields1984,Crookand Shieldsin We defined as coloniesany group of nesting swal-
lows that was likely to interact socially,especiallyat
press).In addition, we planned to explore the
commongrounds (e.g. sunning areas,mud puddles
effectsof sex,age, status,prior experience,and during nest building, common foraging grounds),
local environmental conditionson dispersaland more than occasionallyduring the breeding season.
mate and site fidelity in our population in or- During incubationand nestling stages,we never ob-
der to test some of the generalizationsabout served breeders visiting other colonies. Within the
the proximatecontrolof avian site fidelity (e.g. spatially larger colonies, nests were significantly
Baker 1978, Freer 1979, Greenwood 1980). clumped(for CLBSthe coefficientof dispersionfor
nests = 2.09; X2 = 125.7, P < 0.001; Pielou 1969, see
STUDY AREA AND METHODS also Ball 1983). From data indicating that mobbing
groupswere recruitedfrom nestswithin 25 m of a
We studied the Barn Swallow in the vicinity of the focal nest (Shields 1984), we defined as a cluster all
Cranberry Lake BiologicalStation of the State Uni- nests within a circle 25 m in diameter around each
versity of New York, College of Environmental Sci- focal nest. Birds within clustersinteracted on a daily
ence and Forestry.The stationand other study sites basisrather than occasionally.As defined, a nest could
are locatedon and around Cranberry Lake, the third belong to more than one cluster,and clustersranged
largest lake within the Adirondack State Park. The in size from single isolated nests to as many as 25
lake is large (>30 km2 of surfacearea and >80 km of nests on or in single buildings. From these defini-
shoreline), relatively undeveloped (>80% of the tions, we identified CLBS as a multicluster colony,
shoreline is undeveloped),and surroundedby typi- whereas the BC colony consistedof a single cluster
cal second-growthnorthern hardwood forest and as- in every year.
sociatedhabitat. Historically, Barn Swallows nested Through 1983,we banded 264 adult, 130 juvenile,
in and on cliffs and caves (Bent 1942) and may still and 453 nestling swallows. Since 1980, this has in-
be found in suchnatural sites(Speichpets. comm.). cluded all unmated birds and all but one breeder at
Today, the speciestypically nestson human artifacts CLBS,and about 50%of the Bay City breeders(about
like bridgesand buildings(Bent1942).Our regionis 10% in 1980-1981, 85% in 1982, and 100% in 1983).
analogousto an archipelago of suitable habitat is- We banded all CLBS and BC nestlings,a sample of
lands surroundedby large expansesof unsuitable nestlingsfrom other coloniesaround the lake, and
forest, not unlike the traditional distribution of suit- many free-flying juveniles (known to have hatched
able nest sites.Cranberry Lake and its inhabited en- elsewhere, but captured in our colonies in July and
virons are isolated from other inhabited areas (e.g. August). All adults, since 1980, were given unique
Star Lake 20 km to the west, Massawepie17 km to combinationsof permanent plastic color bands and
the northeast, and Sabattis 12 km to the southeast) temporary tail colors, in addition to U.S. Fish and
by even larger tractsof forest, designatedas wilder- Wildlife servicealuminum bands,in every year. The
ness. Many, but not all, of the suitable inhabitated birds' white tail spotswere painted with Testorsair-
areas harbor swallow colonies. plane dope,one color on eachside, after captureor
As in previousstudies(e.g.Bent 1942,Snapp1976), recaptureeachyear. The paint faded over the breed-
our studies indicated that the "colonies" varied in ing seasonbut could still be read 3-4 months after
size from a single nest to as many as 25 nests on or application.Nestlingswere given singlecolorbands
in single structures,usually boathousesor boatslips. indicative of hatching year and clutch membership
Thesecolonysizesare probablysimilarto thosefound (CLBSnestlingsonly), membershipin other colonies
in traditional sites and are the same as those found (other nestlings), or "foreigner" status (origin un-
in such sites today (nest densities ranging from 1 to known, including all juveniles ). We captured most
15;Speichpets.comm.).We have no way of knowing birdswith mist netsat commongroundsand flyways,
782 W•LLIA• M. SHIELDS [Auk, Vol. 101

50-

• BREEDING
[-'•]RENEST
'"'""•
NATAL

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

t•t 20-

ß lO-
.{3
E
z

0 100 200 300 1000 10,000 20,000

Dispersal Distance (m)

Fig. 1. Distributionof BarnSwallownatal,breeding,and renestdispersaldistances
in meters,1979-1983.
In our studyarea,individualsmovingmorethan 300 m changedcolonies,morethan 25 m changedclusters,
and more than 0 m changed nests.

and there was no evidence that our operationsin- persal).Distanceswere measuredwith a 30-m tape
duced any birds to desertthe study area. within colonies and estimated from U.S.G.S. topo-
Birdswere sexedby tail lengths(Samuel1971a)or, graphicmapsbetweencolonies.From 1980through
aftermarking,by behavior(e.g.only malessingcom- 1983, 111 nests attempts including 74 known first
plete songs;femalesdo the majority of incubating; nestsand 37 known renests(including thosefollow-
Bent 1942, Ball 1983). Unmarked adults occasionally ing successful
andfailedfirstnests)providedthe dis-
could be classedas yearlings or older by skulling. persal data base. Becausewe could not document
When skulling was equivocal,they were classedas every variablefor every attempt,all analysesreport
yearlings.Throughoutthe study,completelife his- on some subset of this total.
tories were taken of as many nestsas our resources
allowed (all nestsat CLBS in 1980-1983 and at BC in RESULTS
1982-1983). Data included the identity of breeders,
the location of the nest with respectto colony and Natal dispersal.--Ofthe 331 nestlings banded
clusterwithin colony,the fate of the nest,the causes through 1982that couldhave returnedto breed
of nest failure, and whether it was a first attempt in through1983,we haverecapturedor resighted
a season or a tenest.
7 (2%),but only 5 of thesewere documentedto
Dispersaldatawere obtainedas a resultof recap- be breeding.The other two (both males)were
ture or resightingof breedingbirds that had been recapturedat CLBSin May but disappearedbe-
identified as individuals or as membersof specific
fore nest building had begun. None of our
clutches,clusters,or colonies during previous years.
Dispersaldistanceswere measuredbetweennatal
nestlingsbred in their natal colonies(Fig. 1).
nestsand the first known breeding site of a bird (na- Owing to the wide spacingof colonies,their
tal dispersal)or betweenconsecutive nests,clusters, median dispersal was 6,375 m, with one un-
and coloniesof experiencedadultsboth within (re- sexed bird being recovereddead during the
nest dispersal)and betweenseasons(breedingdis- breeding seasonby a camperabout 15,000m
October1984] BarnSwallowSiteFidelity 783

TABLE 1. Returns of adult Barn Swallows across

same building, for their first attempts in con-
breeding seasons,1979-1983. secutivebreeding seasons.Three of the four
Per- colony switches between seasonswere due to
Newly Prior Total Total centage femalesand two of these were due to a single
Year banded returns marked returns a returns female that bred at CLBS in 1980 and 1981, then
1979 24 -- 24 10 42.0 switched to an island colony 600 m away in
1980 48 10 58 22 38.0 1982,only to return to CLBSin 1983.
1981 31 22 53 21 39.6 We gathereddispersaldata on 1-, 2-, 3-, and
1982 60 21 81 37 45.6
4-year-oldbreeders.Fidelity to mate,nest,clus-
Total -- -- 216 90 41.6 ter, and colonywas not influencedsignificant-

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

• Returns are from the year after marking, i.e. 1980 ly by adult age either acrossor within (e.g. re-
returns are counted in 1981 and include birds marked nestfidelity, seebelow) breedingseasons (Table
before or during 1980. 2). Acrossbreeding seasons,fidelity to mate,
nest, cluster,and colony and the distancedis-
from its hatchingsite.Three of the four known- persersmovedwere influencedsignificantlyby
sex returns were males, and of the sexed birds, sexand prior breeding experience.More males
the lone femaledispersedthe farthest(8,125m, (48%) than females (24%) manifested nest fi-
Fig. 1). delity (i.e. reusedthe samenest in consecutive
BreedingdispersaL--About40% of the marked breeding seasons),although the sexual differ-
adultsbreedingin our coloniesin any year re- encewasonly marginallysignificant(X2= 3.12,
turned to breed in the study area in the next 1 df, P = 0.077). When the sexeswere pooled,
year (Table 1). Given the 50-70% annual adult previousexperience(i.e. whether or not the in-
mortalityexpectedof migratorypasserines (e.g. dividual successfullyreproduced with that
Lack 1954) and the Barn Swallow (e.g. Lack mate,at that nest,in that clusterand colonyin
1949,Mason1953),this is consistent with strong the previous season)had no effect on nest fi-
adult site tenacity. Only 7% of the adults were delity (37% of the successfuland 33% of the
ever documentedchanging coloniesbetween failuresstayed,X2= 0.06, 1 df, P = 0.80, Table
breeding seasons:36% of the colony-faithful 3). Failure, however,was associated with a sig-
reused the same nest, and an additional 29% nificant reductionin mate fidelity (42% of the
remained in the same cluster, usually on the successful pairs,but none of the failuresmain-

TABLE
2. Probabilityof nest,cluster,and colonyfidelityawithin and betweenbreedingseasons
asa function
of age and sex in the Barn Swallow?

Sex and
type
of Ageinyears
c
fidelity 1 2 3 4
Male

Nest 0.69 (13) 0.42 (24) 0.53 (18) 0.75 (8)

Cluster 1.00 (13) 0.75 (24) 0.67 (18) 1.00 (8)
Colony 1.00 (13) 0.96 (24) 1.00 (18) 1.00 (8)
Female

Nest 0.56 (16) 0.32 (22) 0.22 (9) 0.33 (6)

Cluster 0.88 (16) 0.64 (22) 0.44 (9) 0.67 (6)
Colony 0.94 (16) 0.91 (22) 0.89 (9) 0.83 (6)
aThe proportionof birds that returnedto colonyor to a clusterwithin a colonyor reuseda nestthey had
used in previous seasons.
bThe differencesbetween nest, cluster,and colony fidelity within seasonswere significant(X2 = 34.0, 2 df,
P < 0.001);the age differencesin fidelity were not.
• Breedingdata reflectthe proportion of individuals that are faithful between first and second(2), or second
and third (3), etc., breedingseasons. The renestdata reflectthe proportionof individualsthat remained
faithful to nests,etc.,during their first (1), second(2), etc., breedingseason.Breedingand renestdata were
pooled for older birds, but only renestdata were available for 1 yr olds.
784 WILLIAMM. SHIEnDS [Auk, Vol. 101

TABLE3. Probability of mate, nest, cluster,and colony fidelity between seasonsand the breeding dispersal
distance[mean +_ SE (n)] in meters as a function of sex and previous breeding experiencein the Barn
Swallow.

Prior Fidelity• Dispersal

Sex success Mate (n) Nest (n) Cluster (n) Colony (n) distance(m)
Male Success 0.41 (17) 0.58 (19) 0.74 (19) 1.00 (19) 25 _+8.4 (19)
Failed 0.00 (9) 0.67 (9) 0.67 (9) 0.89 (9) 820 _+798.2 (9)
Female Success 0.41 (17) 0.35 (17) 0.53 (17) 0.94 (17) 95 _+55.1 (17)
Failed 0.00 (8) 0.00 (8) 0.38 (8) 0.63 (8) 1,155 _+777.5 (8)

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

• The proportionof birds that returnedto samemate,to reusednest,or to their previouslyusedclusteror
colony acrossseasons.

tained their bonds, X2 = 6.48, ! df, P = 0.0!; for percentageof nestsreuseddid not differ with-
more on divorce, see Crook and Shields in in and between seasons, but the statistical
press).The latter effect probably was due to equalityresultedfrom a reversalof the effects
femalestending significantlyto change nests of prior experienceat the two stagesof the
and mates after a failure [100% of the female breeding cycle. Within seasons,successful
failureschangednestsand mates,whereassuc- nesterswere more likely to changenests(22%
cessfulfemales only changed nests 65% (X2= vs. 68% fidelity for unsuccessfulbirds, X2=
3.7!, 1 df, P = 0.053), and mates 57% of the time !0.22, ! df, P < 0.001, Table 4), whereas the re-
(x2 = 4.94, ! df, P = 0.03); Table 3]. In contrast, verse was true between seasons(47% of suc-
39% of the successfuland only 29% of the un- cessfulvs. 33%of failures reusedpreviousyear's
successful maleschangednests(X2= 2.!3, ! df, nests, Table 3). The difference between the
P = 0.!4), althoughthe trend in mate fidelity stages wassignificant
(X2= 5.07,1 df, P = 0.024)
was necessarily the samefor both sexes(Table and appearedto be due to a decreasedproba-
3). bility that unsuccessful
femaleswould change
Cluster and colony fidelity were similar to nestsand an increasedprobabilitythat success-
nest fidelity, but, as the distance scale in- ful breedersof both sexeswould changenests
creased,there was a greatertendencyfor birds beforeattemptinga secondbrood(Tables3 and
to remain faithful to previouslyusedsites(44% 4). Finally, both sex and prior experienceap-
nest, 60% cluster,and93% colony fidelity, X• = peared to affect mate and site fidelity during
28.7, 2 df, P < 0.001). The few birds that renesting.Unsuccessful birds,and particularly
changedcolonieshad failedto raiseyoungdur- females,had a higher probability of changing
ing the previousbreedingseason(Table3). In mates,clusters,and colonies,althoughthe dif-
sum, successful breeders of both sexes tended ferenceswere not as great as between seasons
to return to their previous colonies, clusters and only one approached statistical signifi-
within colonies, nests, and even mates, if the cance(27% of unsuccessfuland 7% of the suc-
latter were still available. Unsuccessful birds of cessfuldivorced, X2 = 2.77, ! df, P = 0.09, Table
both sexes,and femalesin general, tended to 4).
dispersemore frequently and moved the far- Casehistories.--Thequantitative analysispre-
thestwhen dispersing(Table3). sentedaboveprovidesinformationon how the
Renestdispersal.--BarnSwallows were more "average"swallowis likely to behave.Consid-
sedentary and more faithful to mates for the eration of the finer behavioral details and the
purposeof renestingthan they were across lifetime movements of individual birds would
breedingseasons (nestfidelity:53%reusednests providemore informationand perhapsneces-
within vs. 44% acrossseasons,X2= 0.93, 1 df, sitate different conclusions.To conserve space,
P = 0.33;cluster:88%vs. 60%fidelity, X• = 11.1, I will briefly discussrepresentativehistoriesof
! df, P < 0.00!; colony:98% vs. 93% fidelity, individual birds and nests.I hope that they will
X2 = 2.9!, ! df, P = 0.087; and mate: 78% vs. 27% illustratehow the generalpattern hasbeen de-
fidelity, X• = 27.4, ! df, P < 0.00!). Those that rived from the behavior of those individuals.
did dispersetended to move shorter distances Becausethe tables include only recovered
before renesting(compareTables3 and 4). The birds,they provide no data on disappearances.
October1984] BarnSwallow
SiteFidelity 785

TABLE4. Probabilityof mate, nest, cluster,and colony fidelity within seasonsand the renestingdispersal
distance[mean _+ SE (n)] in meters as a function of sex and previous breeding experiencein the Barn
Swallow.

Prior Fidelitya Dispersal

Sex success Mate (n) Nest (n) Cluster(n) Colony (n) distance(m)
Male Success 0.88 (8) 0.22 (9) 0.89 (9) 1.00 (9) 7.7 + 6.2 (7)
Failed 0.79 (19) 0.74 (19) 0.95 (19) 1.00 (19) 8.8 + 8.3 (19)
Female Success 0.88 (8) 0.22 (9) 0.89 (9) 1.0 (9) 7.8 + 5.4 (8)
Failed 0.67 (21) 0.62 (21) 0.81 (21) 0.95 (21) 105 + 86.2 (21)

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

aProportion of birds that remained with same mate, reused nest, or remained in their previously used
cluster or colony for renest.

More individuals (2 males and 6 females) dis- historiesbegin courting (e.g. becauseboth of
appearedduring the breeding seasonafter an their previous mates have died or disap-
unsuccessful first nest than after successful nests peared),there is at leastthe possibility of a con-
(1 male and 2 females).Although the sampleis flict over where to nest. We have observedpos-
too small for rigorousstatisticalanalysis,it ap- sible nest-siteconflictsin four pairs. In those
pearsto be likely that someof the within-sea- pairs,malesattemptedto defend two nest sites,
son disappearances(and by implication some the one that their female "preferred" (as evi-
between-seasonas well) were colony deser- denceby her nest-buildingactivities)and a nest
tions rather than deaths. This implies that, al- that the male had used previously. In every
thoughsitefidelity may be strong,it is lessper- casethe pair ended up nesting at the site cho-
fect than the documentedreturnsalone imply. sen (and in 3 of the 4 cases,previously used)
Histories of individual birds indicated that by the female.
they were not "innate" or continuouswander-
ersand that after changingcoloniesthey would DISCUSSION
settlepermanentlyafter nestingsuccessfully (2
cases)or return to a previouslyusedareaif they Natal dispersal.--In the Adirondacks, as in
had failed in a new colony (1 case).It appears other North American populations(e.g. Mason
that swallows in our population were seden- 1953, Ball 1983), juvenile Barn Swallows usu-
tary unless poor or changing conditions ally disperse from their natal colonies before
"forced"them to move.This was supportedby breeding for the first time. Large expansesof
the fact that all of the colony shifts (n = 6) and unsuitablehabitat and long distancesdo sepa-
most of the cluster shifts (80%, n = 20) we ob- rate suitable colony sites in our region. Like
servedwere associatedwith previousnest fail- Mason's (1953), however, our data are consis-
ures or the disappearanceor death of previous tent with the generalization that juvenile fe-
mateswith whom the dispersershad bred suc- male passetinesare more likely to disperseand
cessfully. move farther than males (Greenwood 1980).
Individual nest historiessuggestedthat fa- Whether the sexual bias functions to avoid too
vored nestsmight be used in every year, even intense inbreeding (Greenwood 1980) or re-
if owner identities changed from year to year suits from differences in intrasexual competi-
(2 nestsusedfor 4 yr, 4 nestsfor 3 yr, and 6 for tion for mating resources(Greenwood 1980,
2 yr). In addition, someindividuals appearto Moore and Ali 1984) remains problematic. Un-
have nest traditions (5 birds), with favored nests til more natal returns are available, answers to
for first attempts in a seasonand different fa- such questionswould be speculative.
vored nests for renest attempts.We also have Unlessjuveniles from the samenestsor col-
documented individuals using a nest in one onies disperse and immigrate into breeding
year, using a different one the next, only to colonies together (for which there is no evi-
return to the original in the third year (4 cases). dence), it is likely that the breeding adults
Over time, individuals appear to develop within a colony in our population are not par-
personal preferences for favored nest sites. ticularly close kin. This implies that, unlike
When two individuals with different nesting many colonial or even territorial bird species,
786 WILLIAM
M. SHIELDS [Auk,Vol. 101

in which closekin are probably socialinterac- petition for a place in the breeding population.
tants [e.g. some cooperative breeders, Emlen Once femaleshave begun breeding, they ap-
1978;the Great Tit (Parusmajor),Greenwoodet pear to have a higher mortality than males
al. 1979b], the primary forms of social interac- (Mason 1953,Shields unpubl. data). The result
tion among breeding Barn Swallows should is a male-biasedsexratio amongbreeding birds
tend to be more selfish, competitive, or even (Emlen and Oring 1977),with somemale Barn
disruptive than cooperative or altruistic Swallowsremaining unmated in every year in
(Greenwoodet al. 1979b,Gadgil et al. 1983). eachmajorcolony(Crook and Shieldsin press).
Breeding dispersaL--Oncea Barn Swallow has In such circumstances,males are expected to
bred, it tendsto return to the samecolony,clus- competeintensely for mating opportunitiesde-

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

ter within a colony, building, nest (or set of spite the monogamousmating systemcharac-
nests), and even mate for as long as it lives terizing the species(for discussionof greater
(Tables 3 and 4). This fidelity may be favored male competition in another monogamous
by a number of environmental factors (re- swallow, the Purple Martin, seeBrown and Bit-
viewed in Baker 1978,Shields 1982).Returning terbaum 1980).
to the same colony assuresthat a bird will be As Greenwood (1980) noted, because males
familiar with local resource distribution, in- of mostbird speciescompetefor territories(and
cluding foraging areas and predator refuges, nest sites) needed by females, it is males that
and so need not waste time in exploration early would be expectedto show greatersite fidelity,
in every breeding season. For example, our and in the Barn Swallow they do (Table 3). Once
swallowsoften forageat shelteredbeaverponds a male has competedsuccessfullywith a spe-
during cool windy weather. These ponds are cific set of male neighborsand gained control
surroundedby forestand occurat varying dis- of a site that attractsa female, it may pay him
tances from the breeding areas (l-6 km). It to remain faithful to that cluster and its famil-
probably savestime and energy to learn the iar social environment. On a speculative note,
locationof suchalternativeforagingsitesonce the value of male tradition could stem from a
in a lifetime rather than to learn it anew every number of factors.Moving to a new cluster
year, and the savingsare likely to benefitboth would require another round of competitive
sexes. interactionswith relative strangers.This could
A second resource affected by familiarity is entail a greaterlikelihood of failure than would
potential mates.When both membersof a pre- settling in a familiar cluster with familiar in-
viously mated pair of Barn Swallowsreturn to dividualswith which a male hasalreadycom-
the samecolony, they can and often do remate peted successfully.It is even conceivable that
(Tables 3 and 4; Crook and Shields in press). the familiarity and repeated interactions of
Colony fidelity is likely to facilitate mate fidel- clustermembers,both within and acrossyears,
ity, and the latter might generate additional might facilitatelessintenseaggressionamong
savings in time and energy. An experienced neighborsasa form of reciprocity(Trivets 1971).
pair may haveto courtlessand shouldbe more Reducedaggressionamongfamiliar neighbors
compatible than experiencedbut newly paired relativeto that displayedtowardsstrangershas
birds (older birds whose previous mates have frequently been documented in territorial
died or disappeared)or, especially,than newly songbirds(e.g. Weeden and Falls 1959, Emlen
paired yearlings. Such familiarity could, as it 1971).
has in other monogamousspecies,result in in- Regardlessof the value of cluster fidelity,
creasedreproductive success(for reviews, see each cluster contains a variable number of suit-
Rowley 1983,Coulsonand Thomas 1983). ablenestsitesand old nests.Assumingthat nest
Male fidelity.--Becausethe primary benefits sites vary in intrinsic quality, it would pay a
of colony and mate fidelity would accrue re- male to use and advertise the highest quality
gardlessof whether a bird usedthe samenest nest site available. If a male does return to a
or returned to the same cluster, the less in- colony,and to a specificclusterwithin that col-
tense, but still significant, fidelity observedat ony, it may be that it contains one best nest
these levels (Table 3) requires additional expla- site, in terms of safety,shelter,or defensibility.
nation. Such precise tradition is significantly If a site is better, it is also likely to have been
strongerin malesthan in females.This sexbias used in previous years by the same or other
could result from more intense intramale com- males.In fact,the presenceof an old nestmay
October1984] BarnSwallow
SiteFidelity 787

actuallyadd to a site'squality, asit takesmuch sure that she desertedhim as well, regardless
less time and energy to refurbish an old nest of his intrinsic suitability.In either case,a fe-
than to build one from scratch (Bent 1942, Sam- male would have a greater probability of im-
uel 1971b).Once he hasreturned to a particular proving her reproductivesuccess by dispersing
colony, a male might be forced to return to a awayfrom her previouslyusednestand cluster
previousclusterby socialcompetitionand then than would a male. An unsuccessful male could
reusesthe same nest becauseit is repairable alsobenefit by changingmatesand nest sites,
and is located in the best site. In addition to and some few actually do disperse(Table 3).
offering potential explanationsfor an individ- More males than females, however, are appar-
ual male'snestfidelity, suchfactorsmight help ently forced by intrasexualsocial competition

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

explain the observationsof traditional nest sites to maintain the status quo with respect to
and nestsused by different individuals in dif- breeding sites, even after breeding unsuccess-
ferent years. fully (Table 3).
Femalefidelity.--Once femaleshave returned RenestdispersaL--All factors favoring mate
to a colony, probably gaining the sameadvan- and site fidelity acrossbreeding seasonsare
tagesas males, their pattern of within-colony likely to operate within breeding seasonsas
fidelity is similar but not nearly as intense as well. There are also additional reasons for ex-
that of the males (Table 3). If a female wished pecting the even stronger fidelity observed
to remate, she might be "forced" to return to (Table 4). In many passerines,including the
the same cluster because of her mate's cluster Barn Swallow, first nests are likely to fledge
fidelity. She would then be likely to use the more young than are renests(for the swallow,
previouslyusednestat the highest-qualitynest see Mason 1953; for reviews, see Lack 1954,
site for the samereasonsas her mate (e.g. ear- 1966). In addition, whether an attempt is the
lier breeding, safer or more shelterednest). In first of a seasonor a renest,for many passetines
contrast,if for any reasonshe wished to move the earlier a pair breeds,the higher the repro-
(e.g. availability of a better mate, nest site, or ductive successit can expect (Lack 1966). This
territory), she would not have to competewith implies that for many speciesbreeding condi-
unfamiliar local malesbefore settling in a new tions might deteriorate during the breeding
cluster,even if shewere forcedto competewith season. If this were true, time would be a lim-
local females for a nest site or male. If female iting resourceand could constrain an individ-
competition were less intense than male, the uals'sdecisionsabout renest dispersal.Even if
costsof moving would probably be lessfor fe- a nest has failed and a bird could "blame" its
males, and they might chooseto move more mate or its nest site, it might not have the time
readily. to desert, search out a new mate and area, and
Most female dispersal follows nest failure have ashigh a probability of successasif it had
with a particular male at a particularsite (Table remained faithful. The increasedtendency to
3). If nest failure can be attributed to site- or change nestsafter successfulfirst nests(Table
mate-specificfactors,then a female might gain 4) might be a responseto those factors (e.g.
by deserting her mate and his breeding terri- ectoparasites)that can reduce the quality of an
tory and trying elsewhere. For example, if her active nest over the courseof a breeding sea-
current nest failed because her mate failed to son.

defendit adequatelyor becausehe wasin some

way behaviorally incompatible with her (e.g. CONCLUSIONS
see Coulson and Thomas 1983), then she would
probablydo better with a new mate.If a female The Barn Swallow, like most of the other
chose to desert her mate, his cluster and nest swallows (e.g. Freer 1979) and passerinebirds
fidelity would force her to deserthis territory in general(for reviews,seeBaker1978,Shields
and nest as well, regardlessof their intrinsic 1982),manifestshigh levels of adult site tenac-
suitability.Alternatively, if the nest or site it- ity coupledwith significantnataldispersal.The
self were unsuitable (e.g. it were heavily par- greater natal dispersalcould be an adaptation
asitized or vulnerable to rain or wind), then limiting the possibilityof intense inbreeding
she would probably do better by choosinga (e.g. Greenwood1980)or couldbe a fortuitous
new nest site. If a female choseto changenest result of the greater social subordinanceof
sites,her mate'ssite fidelity would usually in- younger birds (e.g. Murray 1967, Gauthreaux
788 WILLIAM
M. SHIELDS [Auk,Vol.101

1978).The sexualbias in dispersal,with greater BROWN, C. R., & E. J.BITTERBAUM.

1980. Implications
male than female site fidelity (Table 3), is sim- of juvenile harassmentin Purple Martins. Wil-
son Bull. 92: 452-457.
ilar to the pattern observed in a majority of
passerines(Greenwood 1980, 1983).Becausethe BROWN,
J. 1978. Avian communalbreedingsystems.
Barn Swallow's natal dispersaleliminates the Annu. Rev. Ecol.Syst.9: 123-155.
BRYANT,
D. M. 1979. Reproductivecostsin the House
possibilityof closeinbreeding,the sexualbias Martin (Delichonurbica).J. Anita. Ecol. 48: 655-
in breeding dispersalcannot be an adaptation 675.
to prevent inbreeding. Rather, it appearsto re- CHAPMAN, L. B. 1955. Studies of a Tree Swallow
suit from sexual differences in the kinds and
colony. Bird-Banding26: 45-70.
levels of intrasexual competition characteriz- COULSON,J. C., • C. $. THOMAS. 1983. Mate choice

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

ing the swallow. Greenwood (1980, 1983) also in the Kittiwake gull. Pp. 361-376 in Mate choice
suggestedand Moore and All (1984) empha- (P. Bateson,Ed.).London,CambridgeUniv. Press.
sized such competitive explanations of sex- CROOK, J. R., & W. M. SHIELDS.In press. Sexually
biaseddispersal.Finally, like many other bird selected infanticide by adult male Barn Swal-
lows. Anita. Behav.
species(e.g.Darley et al. 1977,MacDonald1977, DARLEY,J. A., D. M. SCOTT,& N. K. TAYLOR. 1977.
Nolan 1978, Coulson and Thomas 1983, Dot
Effectsof age, sex,and breedingsuccesson site
and Fredga 1983, reviewed in Rowicy 1983), fidelity of GrayCatbirds.Bird-Banding48: 145-
the breeding dispersalof adult swallowsthat 151.
are usually site tenaciousis associatedprimar- DAVIS,E. M., & W. M. DAVIS. 1936. BandingBarn
ily with nest failure. The observationof a re- Swallows(Hirundoerythrogaster).Bird-Banding
versal in renest dispersal, nest desertion more 7: 149-156.

often following successfulattempts, is appar- DOT, H., & $. FREDGA.1983. Breedingand natal
ently unique to the Barn Swallow and remains dispersalof the Goldeneye,Bucephala
clangula.
J.
Anita. Ecol. 52: 681-695.
to be explained.
EMLEN,$. T. 1971. The role of song in individual
recognitionin the Indigo Bunting.Z. Tierpsy-
ACKNOWLEDGMENTS chol. 28: 241-246.
1978. The evolution of cooperativebreed-
I appreciatethe logisticsupportprovidedby the
College of EnvironmentalScienceand Forestryand,
ing in birds.Pp. 245-281in Behaviouralecology
(J. R. Krebs and N. B. Davies, Eds.). Sunderland,
especially,the maintenanceand operationsstaff,led Massachusetts, Sinauer Assoc.
by Larry Rathman,at the CranberryLake Biological
Station. For assistancein the field, I thank S. Balzano, --, & L. W. ORING. 1977. Ecology,sexualselec-
L. Birr, H. Broce,S. Cogswell,T. Dale, M. Fargione, tion, and the evolutionof mating systems.Sci-
ence 197: 215-223.
S. Gaynor, M. Grogan, J. Herter, W. Kappelman, P.
Mason, M. Matiz, M. Somerville, S. Wiles, and, es- FARNER, D. $. 1945. The return of robins to their
pecially,M. Ehlersand J. Crook.I am alsogratefulto birthplaces.Bird-Banding16: 81-99.
C. Brown,J. Crook, B. Murray, E. Waltz, L. Wolf, and FREER,
V.M. 1979. Factors
affectingsitetenacityin
two anonymousreviewers for commenting on the New York BankSwallows.Bird-Banding50:349-
manuscript. 357.

GADGIL,M., N. V. JOSHI,& S. GADGIL. 1983. On the

LITERATURE CITED mouldingof populationviscosityby naturalse-
lection. J. Theor. Biol. 104: 21-42.
ALLEN,R. W., & M. M. NICE. 1952. A studyof the (aAUTHREAUX, S. A., JR. 1978. The ecologicalsignif-
breedingbiologyof the Purple Martin (Progne icanceof behavioral dominance.Pp. 17-54 in
subis).Amer. Midi. Natur. 47: 606-645.
Perspectives in ethology,vol. 3 (P. P. G. Bateson
AUSTIN,O. L. 1949. Site tenacity,a behaviortrait of and P. H. Klopfer, Eds.).London, Plenum Press.
the CommonTern. Bird-Banding.20: 1-39.
BALL,G. F. 1983. Evolutionaryand ecologicalas- GREENWOOD, P.J. 1980. Matingsystems, philopatry,
pectsof the sexualdivision of parental care in. and dispersal in birds and mammals. Anita. Be-
hav. 28: 1140-1162.
Barn Swallows. Unpublished Ph.D. dissertation.
Newark, New Jersey,RutgersUniv. Inst. Anita. 1983. Mating systemsand the evolutionary
Behav. consequencesof dispersal.Pp. 116-131 in The
BAKER,
R. R. 1978. The evolutionaryecologyof an- ecology of animal movement (I. R. Swingland
imal migration.London,HodderandStoughton. and P. J. Greenwood,Eds.)Oxford, England,Ox-
BENT, A.C. 1942. Life histories of North American ford Clarendon Press.

flycatchers,larks,swallows,and their allies.New --, P. H. HARVEY, & C. M. PERRINS. 1979a. The

York, Dover Press. role of dispersalin the Great Tit (Parusmajor):
October1984] BarnSwallow
SiteFidelity 789

the causes,consequences and heritability of na- ROWLE¾,

I. 1983. Remating in birds. Pp. 331-360 in
tal dispersal.J. Anita. Ecol. 48: 123-142. Mate choice (P. Bateson, Ed.). London, Cam-
, , & - 1979b. Kin selection and bridge Univ. Press.
territoriality in birds?A test. Anim. Behav.27: SAMUEL,D.E. 1971a. Field methodsfor determining
645-651. the sex of Barn Swallows (Hirundo rustica). Ohio
HAMILTON,
W. D. 1964. The geneticalevolutionof J. Sci. 71: 125-128.
social behavior. I, II. J. Theor. Biol. 7: 1-52. 1971b. The breeding biology of Barn and
HOOGLAND, J. L. 1981. Nepotism and cooperative Cliff swallowsin West Virginia. Wilson Bull. 83:
breeding in the black-tailedprairie dog (Sciuri- 284-301.
dae: Cynonmys ludovicianus).
Pp. 283-310 in Nat- SHERMAN,P.W. 1980. The limits of ground squirrel
ural selection and social behavior (R. D. Alex- nepotism.Pp. 505-544 in Sociobiology:beyond

Downloaded from https://academic.oup.com/auk/article/101/4/780/5191362 by guest on 13 October 2022

ander and D. W. Tinkle, Eds.).New York, Chiron nature/nurture? (G. W. Barlow and J. Silverberg,
Press. Eds.). Boulder, Colorado, Westview Press.
1983. Nepotism and alarm calling in the --. 1981. Intraspecifickilling and nepotisticde-
black-tailed prairie dog (Cynomysludovicianus. fense.Pp. 311-331in Natural selectionand social
Anim. Behav. 31: 472-479. behavior (R. D. Alexander and D. W. Tinkle, Eds.).
LACK,D. 1949. Vital statisticsfrom ringed swallows. New York, Chiron Press.
Brit. Birds 42: 147-150. SHIEEr•S,
W. M. 1982. Philopatry, inbreeding, and
1954. The natural regulation of animal the evolution of sex. Albany, New York, State
numbers. Oxford, England, Oxford Clarendon Univ. New York Press.
Press. 1983. Optimal inbreeding and the evolu-
ß 1966. Population studiesof birds. Oxford, tion of philopatry. Pp. 132-159 in The ecology
England, Oxford Clarendon Press. of animal movement (I. R. Swingland and P. J.
LIGON,J. D., & $. H. LIGON. 1978. Communal breed- Greenwood, Eds.). Oxford, Oxford Clarendon
ing in Green Woodhoopoesas a case for reci- Press.

procity. Nature 276: 496-498. 1984. Barn Swallow mobbing:self-defence,

MACDONALD,M. A. 1977. Adult mortality and fi- collateralkin defence,group defence,or paren-
delity to mate and nestsite in a group of marked tal care? Anita. Behav. 32: 132-148.
Fulmars.Bird Study24: 165-168. $NAPP,B. D. 1976. Colonial breeding in the Barn
MASON,E.A. 1953. Barn Swallow life history data Swallow (Hirundorustica)and its adaptivesignif-
basedon banding records.Bird-Banding24: 91- icance. Condor 48: 471-480.
100. STONER,D. 1936. Studies on the Bank Swallow, Ri-
MAYHEW,
W.W. 1958. The biologyof the Cliff Swal- pariaripariariparia(Linneaus).RooseveltWildl.
low in California. Condor 60: 7-37. Bull. 4: 126-233.
McNIcHoI•L, M. K. 1975. Larid site tenacity and TRIVETS,
R. L. 1971. The evolution of reciprocal al-
group adherancein relation to habitat. Auk 92: truism. Quart. Rev. Biol. 46: 35-57.
98-104. WEEDEN,J. $., •r J. B. FALLS. 1959. Differential re-
MEAD,C.J. 1979. Colony fidelity and interchange sponses
of maleOvenbirdsto recordedsongsof
in the SandMartin. Bird-Study26: 99-106. neighboring and more distant individuals. Auk
MOORF,J., & R. ALI. 1984. Are dispersal and in- 76: 343-351.
breedingavoidancerelated?Anita. Behav.32:94- WILSON,D.S. 1980. The natural selectionof popu-
112. lations and communities. Menlo Park, Califor-
MURRAY,B. G., JR. 1967. Dispersal in vertebrates. nia, Benjamin-Cummings.
Ecology48: 975-978. WOOLFENDEN, G. E., •r J. W. FITZPATRICK.1978. The
1984. A demographictheory on the evolu- inheritanceof territory in group-breedingbirds.
tion of mating systemsas exemplifiedby birds. BioScience 28: 104-108.
Evol. Biol. 18: 70-140. W•IC•T, S. 1978. Evolution and the geneticsof pop-
NOLAN,V., JR. 1978. Ecologyand behavior of the ulations, vol. 4. Variability within and among
Prairie Warbler (Dendroica discolor). Ornithol. natural populations. Chicago, Univ. Chicago
Monogr. No. 26. Press.
PIELOU,E.C. 1969. An introduction to mathematical ZAHAVI,A. 1974. Communalnestingby the Arabian
ecology.New York, Wiley-Interscience. Babbler: a case of individual selection. Ibis 116:
RHEINWALD, G. 1975. The pattern of settling dis- 84-87.
tances in a population of House Martins (Deli-
chonurbica).Ardea 63: 136-145.