Nestling Growth And Thermoregulatory

TheAuk 118(1):116-136, 2001

NESTLING

GROWTH AND THERMOREGULATORY

DEVELOPMENT

IN SUBALPINE DUSKY FLYCATCHERS

MARIA ELENA PEREYRA • AND MARTIN L. MORTON DepartmenotfBiologyO, ccidentaCl ollegeL,osAngelesC, alifornia90041,USA

ABSTR^CT.--ThDeuskyFlycatche(rEmpidonaoxberholserai)s, mall,open-nestingNeotropicalmigrant,is a relativelycommonsummerinhabitantof subalpineenvironmentisn the westernandsouthwesterUnnitedStatesN. estlinggrowthanddevelopmenotf thermoregulationwerestudiedin a populationatTiogaPass(3,000m) in theeasternSierraNevadaof California.Despitetheselectiveadvantagetshatmightbeexpectedfromacceleratinrgates of growthandthermoregulatordyevelopmenitn, anenvironmenwt heretheseasonwasrelativelyshortand coldtemperatureasndinclemenwt eatherwerecommont,herewaslittle evidenceof adaptivemodificationin eitherof thoseparametersg;rowthrateswerecomparableto thoseof othertyrantflycatchersF.eathereruptionbeganaroundday5, brushingin dorsalandventraltractsbegantwo dayslater,andbrushingof theflightfeathersaround day 9. Asymptoticbodymasseswereattainedaroundday 12,but nestlingsdid notfledge foranother4 to5 days.Thephysiologicadlevelopmenotfendothermwy ascloselycorrelated with increasesin body massand accompanyindgecreaseisn surface-area-to-volumraetio andage,but wasnot significantlycorrelatedwith changesin plumagedevelopmenwt,hen massand age were statisticallycontrolled.On exposureto ambienttemperaturesbetween -3 and5øCfor 10min (roughlytwiceaslongastheaveragefemaleboutof inattentiveness), individualnestlingswereunableto preventdeephypothermiauntil after day 7. Homeothermy,at 80%of adultlevels,wasnotattaineduntil afterday12,althoughbroodsofnest!ingsmaintainerdelativelystablebodytemperatureassearlyasday5.Whereacslutchsizes in thispopulationgenerallyrangedfromtwoto foureggst,hemoststablethermalenvironmentswereprovidedbybroodsof threeor fourchicksA. lthoughbroodsof nestlingswere generallyableto maintaintemperaturews ithin5øCof adultlevelst,emperaturperofilesduringstormsindicateda remarkablecapacityfornestlingstotoleratetemperatureisn thenest aslow as15øCfor periodsaslongas4 h withoutapparentill effect.Receive1d9May1999, accepte8dAugust2000.

IN MANYVERTEBRATaEcSc,eleratedgrowthis positivelycorrelatedwith vulnerabilityduring the juvenilestage(Williams1966,Case1978). In birdswhoseyoungdevelopin relativelyexposednestsites,thisis oftenreflectedin early thermoregulatorydevelopmenat nd early mobility (Morton et al. 1972, Austin and Ricklefs 1977,Finch1984).Altricial birds are particularly vulnerableduringthenestlingperiod,depending completelyupon parental care for warmth, food, and protectionfrom the elements.To escapepredatorsor weather,chicks of open-nestingspeciesmustquicklydevelop in size and degreeof neuromusculardevelopmentandmaturephysiologicaful nctionthatis necessaryfor survival outsidethe nest. The ability to maintainelevatedbody temperatures is an importantcorrelateof this,enablingefficientphysiologicarlesponseasndtherapid,co-
E-mail: [email protected]

ordinated movements needed for locomotion
andfeeding.Homeothermytendsto occurat a lower percentageof asymptoticbody mass thanin cavitynestersof similaradultbodysize (Visser 1998). Emerging adult capabilitiesin
those areas do not come without costs, how-
ever.The energyinvestedin activitiessuchas thermoregulatioinsunavailableforgrowthand tissuedevelopmenta,sindicatedby theslower growthratesthatoccurin tissueswith a high degreeof maturefunction(Ricklefs1979,Vis-
ser and Ricklefs 1993, Hohtola and Visser 1998,
Ricklefs et al. 1998). The geographicatlrendsfor rapid growthin
temperate-zonebirds,ascomparedwith tropical and subtropicalspecies(Nice 1954, Lack 1968, Ricklefs 1968, Klaasen and Drent 1991) andin speciebsreedingin areaswheretheyare relatively exposedto predatorsor weather, form the basisof the view advocatedfirstby Lack (1948,1954,1968)and otherssince(Ricklefs 1968), that environmentalfactorsare the

116

January2001]

GrowthandThermoregulatiinoFnlycatchers

117

driving selectiveforcebehindthosepatterns. Further,thoseecologicahl ypothesepsostulate thatgrowthratesareshapedby environmental pressuresuchaspredationw, eather,andfood supply,largelythroughtheireffectsonnestling mortality (O'Connor 1977,1978;Ricklefs1983).
Testsof thoseideasare often complicated, however,by the taxonomicbiasesthat accompany geographicacl omparisons(see,for example,Ricklefset al. 1998).Regionalcomparisonsbetweengrowth ratesin temperateand tropical zonescan be biasedby the zoogeographicaldominanceof somefamiliesin certain regions(Vuilleumier 1975).Sharedattributeswithin familiesoftenextendto typesof nestsites,foragingmodes.andanatomicasl pecializationsfor feedingand locomotion(see,for example,Traylor and Fitzpatrick1982,for the Tyrannidae)further obscuringeffectsof environmentalfactorson developmentP. erhapsin part due to thosefactors,attemptsto testthe validity of the ecologicalhypotheseshave sometimesyielded equivocalresults.In some studies,thereis evidenceto supportthe idea thatratesof mortalityanddevelopmenatrerelated (Bosqueand Bosque1995,Ricklefset al. 1998);in others,fledgingagebut not growth ratesappearto be relatedto nestlingmortality rates(Maher1964;Ricklefs1969a,b; King and Hubbard 1981); and in still others,there is no
supportat all (Ricklefs1969a,c, 1976,1982).As a result,alternativehypotheseshavebeendevelopedto explain interspecificvariability in growth.Of those,onesetrevolvesaroundconstraintson developmentimposedby energetics of parental effort (Ricklefs 1968, Drent and Daan 1980),and anotherfocuseson potential effectsof anatomicaland physiologicalconstraints upon development (Ricklefs 1969c, 1983, 1984;Ricklefset al. 1998).
Becaustehenestlingperiodandageof fledging dependona numberof growth-relatedparameters, including overall rates of develop-
ment and tissue maturation and the levels of
functionalmaturity attainedat eachstageof growth,studiesof nataldevelopmenitn those environmentscontinueto provideinformation relevantto understandingselectivefactorsthat haveshapedpatternsof aviangrowth.Dataon growth and thermoregulatorydevelopment from a greaterdiversityof speciesp, articularly high-latitudeorhigh-altitudemembersoftropical groupsand tropicalmembersof predomi-

nantly temperate-zonefamilies, enableinterspecificcomparisonsthat can be helpful in differentiatingbetweenthosehypotheses.
Thisstudyfocuseson growthanddevelopment of thermoregulationin the Dusky Flycatcher(Empidonaoxberholserai),small (ca. 10 g), insectivorousNeotropical migrant that breedsin montaneenvironmentsthroughout westernNorth America(Johnson1963,Sedgwick 1993).Thisspeciesbuildsopen-cupnests in a variety of habitatsranging in elevation fromaround1,000m in thenorthernmosptarts of its range,to over3,000m in thesouth(Johnson1963).In the easternSierraNevadaof California,it is theonlytyrantflycatchetrhatroutinely breeds above 3,000 m (Gaines 1988). Thus, an interestingopportunityis provided for investigatingtherelationshipbetweenecologicalsettingandpatternsof growthandthermoregulatory development.Assuming that risk of mortalityduringthe nestlingperiodis an importantselectivefactorin theevolutionof growthpatterns,severalpredictionsmightbe madewith regardto thepatternofgrowthand thermoregulatorydevelopmenthat shouldbe mostadvantageouisn a high-altitudeenvironment. If growth ratesare sensitiveto suchselection,theyshouldbe morerapid in thisspe-
cies than in low-latitude or low-altitude
congenersa,ndthenestlingperiodshouldalso be shortenedS. imilarly,becauseambienttemperaturesdecreasewith altitude, the age at whichhomeothermyis attainedmightbe expectedto occurearlier in this speciesthan in
those of the same size in more favorable envi-
ronments.If, however,growthis not sensitive to thosefactorsbut is constrainedby internal factorsas suggestedby Ricklefset al. (1998), the pattern of developmentin E. oberholseri shouldbe moresimilarto that in othertyrannids,irrespectiveof environmentG. rowthand thermoregulatordyevelopmenitn thisspecies ispresentedin thecontextsofindividualphysiologicaldevelopmentt,hebroodenvironment,
and environmental factors such as inclement
weather.
MATERIALS AND METHODS
Growthandthermoregulationwerestudiedin 192 nestlingE.oberholsefroi mMay throughAugustover a 5 year period (1981through1985)at TiogaPass, Mono County,in the easternSierraNevadaof California.Thestudysiteconsistedoftwoglaciatedc, on-

118

PEREYRAANDMORTON

[Auk, Vol. 118

TABLE 1. List of abbreviations used in text.

Abbrev.
t b Ta ATb Ts tbrood-1 Zbrood-2 /•'Tbrodo
Tegg
P6T P6pln P6• Md• Mdpm Md• Vt, Vte•n VtB RI• R1p•n RIB

Definition
Corebody temperature,øC Ambient temperature,øC Temperaturedifferencebetweenbodyand ambient(Tb-Ta),øC Skintemperaturem, easuredbeneathwing, øC Broodtemperatureat startof experimenta, fterflushingfemale,øC Broodtemperatureat end of 10 min of coldexposure,øC Changein broodtemperatureover10 min testingperiod(Tbrood_2--Tbrood_l) Nesttemperaturea, smeasuredby thermocouple-implanteedgg,øC Totallengthof left 6thprimary(countedfromoutermost1,0thprimaryin) Lengthof ensheathedportionof left 6th primary Brushlengthof left 6thprimary Averagetotallengthof mid-dorsalfeathers Averagelengthof ensheathedportion,mid-dorsalfeathers Averagebrushlengthof mid-dorsalfeathers Averagetotallengthof mid-ventralfeathers Averagelengthof ensheathedportion,mid-ventralfeathers Averagebrushlengthof mid-ventralfeathers Totallengthof left centralrectrix Lengthof ensheathedportionof left centralrectrix Brushlengthof left centralrectrix

nectingvalleysthat run roughlynorth-southand east-west.Thearearangesin elevationfrom2,857m at its lowestboundary(ElleryLake)to about3,150m partwayup theslopesoftheridgessurroundingthe two valleys,with an abundanceof habitatsuitable for E. oberholsetrhiroughoutthe area.Nestsweresituatedin a varietyof plant speciesi,ncludingwillow (Salixspp.),aspen(Populutsremuloidesw),ild currant (Ribesspp.)andpine(PinuscontortaP,.albicaulisa),nd variedconsiderablyin height,position,and degree of exposureto the elements.
Growthwasmonitoredin nestlingsduringeachof thefiveyearsof thestudyandthermoregulatordyevelopmentwasdocumentedfrom 1981to 1984.Nests were locatedduringconstructiona,ndaftercompletion were checkeddaily for clutchstarts.Hatching generallyoccurredon thefifteenthor sixteenthday of incubationa, ndwasprecededa dayaheadbypipping,whichbeganwith smallstar-shapedextrusions
on the surface of the shell. In most cases,the first vis-
it on the day of hatching(designatedday 0) was madewithin 3 h ofemergenceIn. neststhatalready containedday 0 chickswhen visited,the probable time of hatchingcouldbe establishedwithin a few hoursbycomparisown ith descriptionosf a reference group of nestlingsfor which hatchingtimeswere
known for certain.
Eachhatchlingwasmarkedwith a waterproofelttippedmarkerandthecenteror reartoenailofeither the right or left footclippedto correspondwith its hatchingorder.Toenailswerereclippedlaterif necessary,but in most instances,were clearly distinguishableuntil fledgingatday16or 17.Chickswere weighedto 0.01g on an Ohaustriple-beambalance atthetimeof discoveryandonalternatedaysforthe

restof the nestlingperiod.Measurementosf plumageweretakento the nearest0.5 mm, fromthebase of thefeatherto thetip, usinga thin, flexiblep, lastic
millimeter ruler. In 1983 and 1985, chicks were sexed
by laparotomyand bandedwith numberedalumi-
num bands obtained from the U.S. Fish and Wildlife
Service.Weightsobtainedafter sexingwere not includedin analysesof growth.
Nestlingswerecoldstressedfor 10min at ambient temperatures(Ta)assolitaryindividuals (1981,1982) or asbroods(1983,1984).All abbreviationasndsym-
bols for measurements are defined in Table 1. Broods
used as controlsfor effectsof temperaturestresson growthrateswereweighedandmeasureddailyorat 2-3 day intervals.No differencesin growth were seenbetweenneststhatwerevisiteddaily andthose visitedat longerintervals.However,in orderto minimize any possibilitythat growthmightbe affected, nestlingsusedin coldstressexperimentsweretested onalternatedays.Becausme eanweightsandfeather lengthsforeachdayof agewerenotsignificantlydifferent between cold-stressedand control chicks(ttests,P > 0.05),the two groupswere combinedfor analysisV. isitstoeachnestweremadebetween0530 and 0800 PDT when Ta were between -3 and 5øC. Chicksbeingtestedindividuallyfor physiological developmenot f endothermywereremovedfromthe nestand placedin separatecardboardboxes(7 x 11 x 4 cm).Behaviorandposturewerenotedandatthe endof 10 min, deepbody (Tb)andskintemperatures (T•) were recordedalongwith body weightsand featherlengths.In 1981,temperaturesof nestlings were taken orally with a telethermomete(rYellow Springs)and thermistorprobecoatedwith vaseline and threadedinto the proventriculusI.n 1982and

January2001]

GrowthandThermoregulaitnioFnlycatchers

119

1983,temperaturesweretakenwith a quick-registering, smalldiameterthermocouple(BaileyRET-3balltipped probe)insertedinto the cloaca.In addition, beginning in 1982, skin temperatureswere taken with a BaileySST-2skin probe(surfacediameterof 0.5 cm) pressedflat againstthe ilium until a stable temperature registered.
Brood temperatures (Tbrood-w•)ere taken (19821984)immediatelyafter thefemalewasflushedfrom the nest.A quick-registeringball-tippedprobewas insertedinto the centerof the clumpof chicksand pressedagainstthe ventralsurfaceor sideof each one,beneaththe wing and abovethe lung. In 1982, cloacal temperaturesmeasuredbefore the chicks were removedfor coldstresswere alsousedto representTbroodB-•y. either method,temperaturesdid notvary amongchicksaslongasmeasurementws ere takenquickly.In 1983and 1984,the abilityof whole broodsto thermoregulatewastestedasa functionof brood size,nestenvironment,and brood age.Brood age was defined as mean age of all chicksin the brood,roundedto thenearestinteger,andexpressed as "brood age x" or "age x" to distinguishit from the actualageof individual nestlings(dayx). Nests were visited between0530 and 0800 PDT as in previous work, but as soon as the female was flushed,
themaximumtemperatureof thechickswasrecorded ventrally and beneaththe wings. Following10 min of exposureto Ta,their temperaturewas again recorded(T•rooa-T2h).e temperaturedifferentialthat occurredduring that time period(ZXT•roowaa)sused to assessthe thermoregulatorydevelopmentof
chicks in the context of their natural nest environ-
ment.Tenmin of coldexposurewaslongerthanaveragematernalabsenceb,ut stillremainedwellwithin the rangeof observedoff bouts;the averagetime off the nestfor a femalewith chicksbeing about6 min (calculatedfrom nestswhereeggtemperatures in nestswith chickswerebeingrecorded).Only femaleswereobservedto brood,althoughboth sexes
fed chicks.
Measurementsof primary 6 (P6), rectrix 1 (R1), and averagemeasurementsof mid-dorsal (Md) and mid-ventral (Vt) contour featherswere used to describeplumagedevelopmentP. 6 wascountedfrom the outermost(10th)primary in, andR1 represented the centralpair of rectrices.Measurementsof middorsalandmid-ventralcontourfeathersrepresenat n averagelengthof feathersin thoseregionsbecauseit wasdifficultto getthesameonesrepeatedlyF. eather lengthsfor all tractsweredividedintobrushlength (P6B,MdB,Vt•, RI•) and pin length (P6p,Md•,,Vtp, RI•,). Brushlengthwasdefinedasthe lengthof the unfurledportionof thefeatherterminust,o thenearestmillimeter,takenwith a flat ruler.Pin lengthwas defined as that portion of the growing featherstill ensheathedT. otallength(P6T,MdT,Vt•, RI•) wasthe sum of both. Analysesof body temperaturewere performedon both absolutebody temperature(T0

and the temperaturedifferencebetweennestlingT• and Taafter 10 min of exposure(ZITb).
Partialcorrelationanalysesof T• andzIT•with age, bodymass,andfeatherdevelopmenwt ereconducted on nestlingsgroupedby eachof threestagesof thermoregulatorydevelopment,in order to determine the amount of variance in T• that was attributableto eachof thosefactors.The first (phaseI: inertial stage)consistedof nestlingsfrom 0.9 to 5.5 g, the second(phaseII: early regulationstage)consisted of nestlingsfrom5.5to 8.5g, andthethird (phase III: late regulation)encompassednestlingsheavier than 8.5 g. Eachnestlingwas generallyrepresented only onceduring eachphaseof growth. Statistical analyseswere performed using SPSS,BMDP, and SYSTATstatisticalsoftwarepackages.Means are reported as ___SiDn text, or as __+S2Ein figures,unless otherwisenoted.Two-segmentedlinear regressions (Yeagerand Ultsch 1989) were used to determine transition points from thermal conformity to regulation.
RESULTS
GROWTH
Accumulatioonfbodymass.--Growthwassigmoidalin youngE. oberholseirni,creasingfrom a meanmass(+SD) of 1.19 __0_.14 g on day 0, roughly10%of themeanadultmassof 11.5+ 0.9 g (n = 15females,2 males),to a meanmass of 11.1 + 0.8 g on day 12 (Fig. 1A). Because
most birds had achieved over 90% of their as-
ymptoticmassby thisage,day12wasusedas a cutoff date in the calculationof logistic growth rate constants(see below). Between days12 and 16, furtherincreaseswereslight, and weight profilesfor individualsindicated that massesusuallystabilizedby day 14, undergoingsmalldailyfluctuationsuntil fledging two or threedayslater.Fledginggenerallytook placeon day 16 or 17,but couldoccurasearly as day 14 if broods were disturbed. In one broodof three,theyoungeswt asday12when fledging occurred, but it was not known
whether that chick survived.
Using the method of Ricklefs(1967), the growth curvein Figure 1A was transformed into a straightline using logisticconversion factorsbasedon an asymptoticestimateof 11.7 g, whichyieldedthe linearregressionof best fit. Thelogisticgrowthrateconstan(tK) deter-
mined for E. oberholseri from those data was
0.425,with ratescalculatedfor separateyears rangingfrom0.387to 0.448.Usingthesameas-

120

PEREYARNADMORTON

[AukV, ol.118

or thepreviouseveningwereup to 8 h older

lOO thanthosehatchingthenextmorning.

Eruption in primariesand contourfeathers

occurredsimultaneouslyre, achingmaximum 60 '• growthratesbyday11,althoughonlythecon-

40 tourfeathershadcompletedevelopmenbty

20 fledgingfiveor sixdayslaterAftereruption, the 6th primary(P6)lengthenedat a rateof

about1mmperdaybetweendays4 and6,in-

creasingto roughly3.5mmper dayfor therest

40 B Pnmar6y
3O

i-• BrushLength
I SheatLhength

m • -I-

of thenestlingperiodB. yfledgingi,tslength wasstillonly58%ofyearlinglengths(1963),al-

....._..,.iiii,,,,. 2O

thoughmorethan70%of theprimarywasun-

10

sheathe(dFig.lB). Contoufreather(sFig.1C,

0

D) lengthenedat around2 mm per day,but

• 2o C Mid-Dorsal
E
.c lo ......
g 2o D Mid-Ventral

were fully grown and brushed over 75% of
theirlengthsb,ydays13and14.At thisage, overallbodycoveragbey brushingwasalmost
complete.Becausethe mid-ventral and mid-
dorsalcontourseruptedat the sametime and

u. o

closelyparalleledeachotherin ratesofgrowth

and brushing,the mid-dorsalcontourswere

20

lo E.R..e. ct ''iii•

0

,

O

2

4

6

8 10 12 14 16 18

usedto represenot verallcontourdevelopment in subsequenatnalysesT. herectriceswerethe
lastmajorfeathersto erupt(Fig.1E).Thecen-

Nestling Age (days)

tralpair(RI-1) begantoemergebetweendays

6 and 8 and grew at a rateof between2 and 3 prFimIOar.y1.6 M(Be),amnibdo-ddoyrmsaal sans(dAm),ifde-avethnetrraglrcoowntthoiunr mmperdayfromday9 on,attainingroughlya tracts(C,D),andrectrixI(E) fornestlingE.oberholseriquarterof yearlinglengthsby day15. by dayof age.Uncappedverticalines(A) indicate Relationshbipetweebnodymassandplumagdeerangeofvaluesobserveadteachage,cappedvertical velopment.--Twqouestionswere investigated linesrepresen+t 2 SEaboveandbelowthemean(A) with regardto massandfeatherdevelopment

orabovethemean(B-E)andvaluesin parenthesesat a given age.The first was whether heavier

indicatethesamplesizeof nestlingws eighedand measuredat eachage.Blackbars(B-E) indicatethe
lengthof theensheathepdortionof feathersli,ght shadingindicatetshelengthof thebrush.

nestlingasta givenageweredevelopmentally
more advanced,leadingto acceleratedfeather growth. The seconddealt with the extent to
whichpin-feathedr evelopmenmtightinflu-

ence variation in age-specificbody mass

througheffectson fluid balance.

ymptote(11.7 g), the ratio of asymptoticto Thefirstquestiownasaddressebdyanalyz-

adult mass was 1.02.

ingthepartialcorrelatiobnetweenbodymass

Featherdevelopment.--Externianldicationsof andtotalfeatherlengthwhile at thesametime

plumagedevelopmenbteganwith light pig- controllingforage.Weexpectedthatif plummentatiounndertheskininsometractsasearly agedevelopmentwasmoreadvancedin heavi-

as late day 2, followedby the appearanceof er nestlingst,hentotalfeatherlengthshould

beadinganddarkeningpigmentatioansfeath- positivelycorrelatewith masswhenagewas

ersapproacheedruptionE. ruptionof thepri- held constantT. otalfeatherlengthin each
maries and feathers of the mid-dorsal and mid- tract was indeed positively correlated with

ventralregionsdidnotbeginuntillateday4, bodymassfor all threephasesof growth, buthadoccurreidnnearlyallday5chick(sFig. whenagewascontrolled(Table2). During 1B-D).Variationin timeof featheremergencephaseII (5.5to 8.5g), between22 and30%of mayhavebeenpartlya functionof imprecise variationin featherlengthwasassociatedwith

aging;nestlingsthathatchedduringthenight variationin body mass(ventral contours,r =

January2001]

GrowthandThermoregulatinioFnlycatchers

121

TABLE2. Zero-order,markedwith (--) for controlcolumn,and partial correlations(r) for bodymasswith
totalfeatherlength(MdT,Vt•, P6v R1r) andensheathedpin featherlength(Mdp,nV, tp,nP, 6p,,,Rlp•n)w, hile controllingfortheeffectsofnestlingageand/ ortotalfeatherlength.Separateanalysesconductedforeach of the threephasesof growthare discussedin text.Tractsrepresentedare the mid-dorsal,mid-ventral,
primariesandrectricesD. uringphaseI (0.5-5.5g), featherswerecompletelyensheathedh,encepinlength
was the same as total.

Feather

variable

Control var.

Md T VtT P6T Rlr Mdmm
Vtp,.
P6p.•
Ripln

-Age --
Age -Age --
Age -Age Age, MdT --
Age Age, VtT --
Age Age, P6pn, --
Age Age, R1T

0.5-5.g5

r

P

0.58 0.22 0.49
0.14 0.62 0.27 --
-0.58 0.22 -0.48
0.14 -0.62
0.27 ---
-....

<0.001 <0.001 <0.001
<0.001 <0.001 <0.001
--
-<0.001 <0,001
-<0.001
<0.001 --
<0.001
<0.001 ---
--

Bodymasswith featherlength

5.5-8.g5

r

P

8.5g+

r

P

0.78 0.49 0.77
0.47 0.80 0.55 0.34
0.02 0.69 0.44 0.04 0.56
0.31 -0.11
0.81
0.57 -0.36
0.02

<0.001 <0.001 <0.001
<0.001 <0.001 <0.001 <0.001
NS <0.001 <0.001
NS <0.001
<0.001 0.030
<0.001
<0.001 --
0.004
NS

0.75 0.38 0.75
0.38 0.75 0.40 0.69
0.21 -0.06
0.21 0.09 -0.03
0.19 0.05 0.17
0.29 0.20 0.72
0.35 0.29

<0.001 <0.001 <0.001
<0.001 <0.001 <0.001 <0.001
<0.001 NS
<0.001 0.025 NS
<0.001 NS
<0.001
<0.001 <0.001 <0.001
<0.001 <0.001

0.47;primary 6, r = 0.55).However,by phase III (8.5+ g), the amountof variationin plumage length that continuedto correlatewith body masswas only about4-16%, depending upon the feathertract. The declinein partial correlationsbetweenphasesI! andIII indicated that body massexplainedlessof the variationin plumagedevelopmenitn olderchicks and that developmentaldifferencesdid not persist over the nestling period. They also suggestedthat someproportionof that correlationmay havebeeninfluencedby increasesin thefluid contentof developingfeathers.
The extentto whichage-specificbody mass mightbe influencedby fluid contentof thepin featherswasaddressedby examiningthepartial correlationsbetween body massand ensheathedfeatherlength,while controllingfor ageandtotal featherlength(Table2). Weanticipated that if the correlationbetweenbody massandfeatherdevelopmenwt aslargelybe-
cause of the increased fluid content of the
growingplumaget,henbodymassshouldcontinue to positivelycorrelatewith ensheathed feather length when age and total feather

lengthwere controlled;furthermore,that correlation should declineas feathersbegan to brush.During phaseI, wheneruptionin most tractshadyettobeginandfeatherdevelopment wasstill largelyconfinedto areasbeneaththe skin,only2-7% ofvariationin bodymasscould be attributedto variationin lengthof the pinfeathers(mid-dorsalcontours,r = 0.22;midventralcontours,r = 0.14;primary 6, r = 0.27; P < 0.0001for all). During phaseII, however, when all tractswere activelyeruptingor had alreadyemergedt,heproportionof variationin massthat correlatedwith pin developmenitncreased to 10% in mid-dorsal contours (r =
0.31, P < 0.0001) and 32% in primary 6 (r = 0.57,P < 0.0001).By phaseIII, partial correlationsbetweenbody massand pin lengthhad declinedin all threetractsasfeathersbeganto harden (P6pi•r,: = 0.20,P < 0.0001;Mdp•: r = 0.09,P < 0.05;Vt•,•,:r = 0.05, P > 0.1). The rec-
trices,in which developmenwt as mostrapid and the shaftsstill softduringphaseIII, were the only tract for which there was an increase in the partial correlationfor mass and pin lengthbetweenphaseII (r = 0.02,P > 0.1)and

122

PEREYRAANDMORTON

[Auk,Vol.118

phaseIII (r = 0.29,P < 0.0001)T. ogethert,hose data indicatethat not only was featherdevelopmentslightlyaccelerateidn largernestlings, buttheextentof pin-featheringduringphases I andII in particularappearedto influencevariation in age-specificbody mass.
THERMOREGULATION

40

(19) • m

(1•.6•.)--• 35

30

(22)z

25

20(1(31)õ•'"'•

20

(1(52)1)•

15

10 (•

Developmeonfpt hysiologiceanldothermy.--Dur- 5

ingthefirstninedays,meanTbin cold-stressed 0

nestlingsincreasedatarateofslightlylessthan

2øC per day,from 10.6_ 1.0øCon day0 to 26.4 __2_.1øCon day9 (linearregressionT: b= 1.836

0

2

4

6

8

10

12

14

Age(days)

(Age) + 10.218,r = 0.92,F = 980.97,df = 1 and 159, P < 0.0005,Fig. 2A). Daily improvements in Tbbetweenconsecutivedayswere greatest betweendays9 and 11, after whichtemperatureschangedmoreslowly(Tb= 1.442(Age)+
16.701, r = 0.61, F = 40.51, df = 1 and 66, P <
0.0005),peakingat 36.0+ 2.0øCon day 14,the latestage for which thesedata couldbe ob-

B 4O 35 3O 25 2O '15 '10

(16} (19)
I [ (13)(9)

loo
90
8o •
70 •
6O .•
5O •
•o •
•o •.

tained; this was approximately86% of the 5

mean adult Tbof 42.6 ---+0.8øC(n = 15).

Theabilityfor nestlingsto maintainelevated

0

Tbat low Ta (Z•Tb,)followedthe samepattern

0

2

4

6

8

10

12

14

(Z•Tb,= 1.904 (Age) + 5.800, r = 0.89, F =

Age (days)

604.491,df = 1 and159,P < 0.0005),increasing from 5.9 __0_.86øCat day 0 to 29.4 __3_.56øCby day11(Fig.2B).Significanitncreaseisn Z•Tbetweenadjacentdaysoccurredondays2, 6, and 8, andthendailythroughday11.Roughly20

FIc. 2. Meanbodytemperaturesfor individualE. oberholsecrhi icksin nest,at startof testing(Tb•oodq), bodytemperature(Tb),andskintemperature(T,) after 10 min of cold exposurein isolation,(A), and meandifferencebetweenbodyandambientemper-

to 30%of adult z•Tbwasattainedbetweenday ature(ATb)by dayof age(B).Meanambienttemper-

1 and day 2, with relativelysmall improve- atures(Ta)fortestingperiodsoneachdayofageare

mentsoccurringovertherestof thefirstweek. The greatestdaily changesoccurredbetween days9 and 11, with ATblevelingoff around 31øC(z•Tb = 0.987(Age) + 18.162,r = 0.37,F
= 10.49,df = 1 and 66, P < 0.005).Relativeto totalmassaccumulateduringthenestlingperiod, the steepestincreasesin Z•Tboccurred

indicatedonbottomof (A). Cappedverticallines(A) represent2 SE aboveand below the mean. (B) Uncappedverticallinesrepresentthe rangeof values observedfor eachday of age,boxesindicate2 SE aboveand below the mean, shownby a horizontal dash.Samplessizesfor nestlingswhosetemperaturesweremeasuredon eachdayof ageareindicated in parentheses.

whennestlingswerebetween10-20%and80-

100%of their asymptoticweights(Figs.1A,

2B). By day 14, the age of earliestfledging 6;changeasfterthiscontinuedmoreslowlyun-

whendisturbedZ, •Tbwasstillonly83%ofadult til asymptoticmasseswere attained(transition

levels.

pointdeterminedby themethodofYeagerand

In additionto its relationshipwith age,Tb Ultsch,[1989]).Untilthispoint(8.9g), Tbwas was highly correlatedwith body mass(r = linearly related to mass(Tb = 1.788(Mass) +

0.93,P < 0.0005;Fig.3) andtheratioof surface 9.929, F = 813.37, df = 1 and 140, P < 0.0005,

areato volume(r = -0.91, P < 0.0005).Thera- Fig. 3), and ratio of surfaceareato volume(Tb

tio of surfacearea to volumedeclinedsteeply = --3.28 (SA:Volume)+ 34.86,F = 358.24,df =

(SA:Vol= -0.583 (Age) + 7.537,r = -0.953, F 1 and105,P < 0.0005,Fig.4). Asbodymasses

= 849.86,df = 1 and 86, P < 0.0005)until day exceeded8.9g (approximatelyday9) however,

Januar2y001]
35
• 30
•=
•_ •o

GrowtahndThermoregulaitnioFnlycatchers

123

o o
o •o o
o• 08 o •'oø oo o o O•o oe
ß o •_o .
ß •.•e o
ß •, _.• o
ß :,;• '.o-

Approximate Age Class (days)

0

1

2 3 4 5 6 ? 89110+

30

,• !'::

•5

10
0.94
Transition mass = 8. 9 g

i

i

[

i

i

i

l

0

2

4

6

8

10

12

14

,0::a:t": 'i

i 'i

Body Mass (g)

FIG.3. Individual body temperaturesas a func-

tion of nestlingbody mass.Nestlingsfrom day 0

0

throughday9 are indicatedin black,and nestlings

8

day 10 and older are indicatedin gray.

7

6

5

4

Surface Area: Volume

FiG.4. Bodytemperatureasa functionof thera-

the divergencein T0betweenolderandyoungernestlingsfallingwithinthesamemassintervalsbecameincreasinglyvisible(Figs.3, 4).
In therelationshipbetweenT0andmass,potential sourcesof unexplainedvariation in T0 includedchangesin regulatorycapacitythat were relatedto age,and in physicalfeatures

tio of surfacearea(SA, cm2= 8.11Massø-•7W; alsberg
and King 1978) to volume (estimatedby mass,in grams)in nestlingE. oberholseRrie. lationshipbetween decliningratio of surface-area-to-volumaend advancingageis indicatedby dashedboundarylines indicatingtheapproximateboundariesfor eachage class.Regressionline for To(in results)is calculated for nestlingswhoseratiosof SA:Volumewere be-

suchas featherdevelopmen(tFig. 2B). To in- tween7.8and4.2(4.2 = thetransitionpointbetween

vestigatethelatter,partialcorrelationanalyses largelypassivecoolingandwhennestlingsbeganto

were performedon Tb with mass,age, and activelyregulateT0. PhaseII andphaseIII nestlings

plumagecomponentfsor eachof the three are indicatedin gray.

phasesof thermoregulatordyevelopmenNt. es-

tlingsweighing<5.5 g (phaseI), werelargely

passivein responsetocoldexposureS. hivering phaseof growth,to 0.72in thesecondphase,

firstbecamevisibleduringphaseII, whennest- and0.45for the lastphaseof thenestlingpe-

lingswerebetween5.5and8.5g,andbyphase riod (Table3: P < 0.0001for all threephases).

III (8.5+ g), nestlingswereactivelyattempting Wheneffectsof ageandTawereheldconstant,

to maintainhomeothermy(seeFig.3).

partialcorrelationfsorT• withbodymasswere

Tables3 and4 summarizethe partial corre- statisticallysignificanot nlyduringthefirsttwo

lationsfor T0with mass,age,andplumagede- phasesof growth,whenmasswasincreasing

velopmendt uring eachof thosephasesB. e- mostrapidly (phaseI, r = 0.48,P < 0.0001;

causethe partial correlationsfor Tb and AT• phaseII, r = 0.53,P < 0.01).After phaseII,

with mass,age, and plumage development body massincreasedmore slowly,eventually

closelyresembledeachother,onlythe values reachinga plateaua; t thatpoint,otherdevel-

for T0 are reportedhere.The correlationbe- opmentalfeaturesclearlysupercededmassas

tweenT0and massdeclinedasnestlingdevel- explanatoryfactors(phaseIII, r = 0.09,P >

opmentprogressedfr,om0.83duringthefirst 0.05).

124

PERMYRAANDMORTON

[Auk,Vol. 118

TABLE3. Partialcorrelationcoefficientsfor bodytemperature(Tb)with massandwith age,duringeachof threephasesof growth,while controllingfor the effectsof ambienttemperatureandplumagevariables. Zero orderpartials (simplecorrelation)are provided under "no control" category.

No control
Age Ta Age, T• Age, T•, MdT Age, T•, VtT Age, T•, P6T Age, T•, Md•,.n Age, Ta,Vtp•n Age, T•, P6p.• Age, T•, MdT-P6T Age, T•, Md-P6p,,
No control
Mass
Ta Mass, Ta Mass, Ta, MdT Mass, Ta, VtT Mass, T•, P6t Mass, T•, Mdpi• Mass, T•, Vtpi• Mass, T•, P6pl• Mass, T•, MdT-P6T Mass, T•, Md-P6p,n

Phase I

Phase II

r

P

r

P

0.83
0.55 0.84 0.48 0.49 0.51 0.54 0.49 0.51 0.54 0.56 0.56
0.76
-0.16
0.79 -0.06 -0.06 --0.06 -0.14 -0.06 -0.06 -0.14 -0.15 -0.05

Tb with mass

<0.001

0.72

<0.001 <0.001
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
Tb with
<0.001

0.57 0.72
0.53 0.39 0.42 0.29 0.47 0.53 0.29 0.30 0.35 age
0.54

NS

-0.06

<0.001 NS NS NS NS NS NS NS NS NS

0.57 0.05 -0.15 -0.09 -0.27 -0.09 -0.03 0.29 -0.39 -0.30

<0.001
0.001 <0.001
0.002 0.023 0.014 NS 0.007 0.002 NS NS 0.048
0.001
NS
0.001 NS NS NS NS NS NS NS 0.028 NS

Phase III

r

P

0.45
0.02 0.50 0.09 0.12 0.06 0.09 0.13 0.11 0.10 0.12 0.14
0.70
0.60
0.70 0.57 0.41 0.26 0.26 0.59 0.57 0.58 0.24 0.34

<0.001
NS <0.001
NS NS NS NS NS NS NS NS NS
<0.001
<0.001
<0.001 <0.001
0.002 0.033 0.038 <0.001 <0.001 <0.001 0.052 0.011

In contrasto thestrongcorrelationbetween heaviernestlingsweremoreadvancedin feath-

bodymassandTb(orATb)whenagewascon- er developmenta, ndearlyfeatherdevelopment

trolled, the correlationbetweenage and Tb explainedsomeofthevariabilityin age-specific

(whilecontrollingforeffectsofbodymass)was body mass, we addressedthe questionof

not statisticallydifferentfrom zero until the whetherearlypin developmenint fluencetsher-

lastphaseof growth(phaseIII, r = 0.57,P < molysisby analyzingtherelationshipbetween

0.0001). That suggested that incremental Tb, ATband ensheathedfeatherlength,while

changesin body massand surface-to-volumecontrollingfor age, mass,and total feather

relationshipweremoreimportantduringthe length.

earlierphasesof growth than simplechrono- With the start of feathereruptionin late

logical age, but that maturationof control phaseI nestlingsT, bwasnegativelycorrelated

mechanismspredominated during the last with pin lengthsin P6, mid-dorsal,and mid-

phaseof development.

ventralcontoursa, lthoughthe correlationwas

Most improvementsin thermoregulatory

abilitymightbe expectedto dependon mass- significantonly for the primaries(Table4).

relatedchangeisn thermolysiasndthermogen- Roughly8% of the variationin Tbwasattrib-

esis.However,changesin plumagedevelop- utableto eruptingpin lengthwhentheeffects

mentmight alsoaffectTbthrougheffectson of body mass, age, and T• were controlled

heat lossor retention.Plumagedevelopment (P6pinr:= --0.28,P = 0.018).Duringthesecond

wasstronglycorrelatedwith TbandATb,when phaseofgrowth,Tbcontinuedtocorrelatewith

nestlingsfrom the entirenestlingperiodwere primaryandrectrixlengths(Table4), although

examined (zero order correlations: Tb, ATb, thesignofthepartialwasnowpositiveD. uring

with P6T:r = 0.91;Mdt, Vtw:r = 0.92;RT:r = that phase,terminalportionsof thosefeathers

0.81; n = 149, P < 0.0005 in all cases).Because begantohardenwithinthesheathsandthecor-

January2001]

GrowthandThermoregulatinioFnlycatchers

125

TABLE4. Partial correlationcoefficientsfor body temperaturewith pin length,while controllingfor the effectsof age,Ta,andmass.Dashedline indicateszero orderpartial. During PhaseI (0.5 g to 5.5 g), the entirefeatherwas ensheathedh, encepin lengthwas the equivalentof the entirefeatherlength.During PhasesII (5.5g to 8.5g) andIII (8.5g+), the"pin" lengthwasa measureof theensheathedlength,which becameshorterasthe endof the featherbeganbrushing.

T b with
Mdpm T b with Vtpn, Tb with
P6pm T b with Rlp,•

Control Vars.
-Age, Ta,Mass -Age, Ta,Mass -Age, Ta,Mass -Age, Ta,Mass

Phase I

r

P

0.50
-0.14 0.56
-0.20 0.58
-0.28 ---

<0.001
NS <0.001
NS <0.001
0.018 ---

Phase II

r

P

0.61
0.11 0.42 -0.07 0.72
0.36 0.63 0.50

<0.001
NS 0.010 NS <0.001
0.032 <0.001
0.004

Phase III

r

P

0.04
-0.25 0.10
-0.14 0.07
-0.07 0.60
-0.10

NS
0.037 NS NS NS
NS <0.001
NS

relationwas essentiallyequivalentto that betweenTband totalfeatherlength.
During phaseIII, featherswere mostlyun-
sheathed in dorsal and ventral contour tracts,
but were still largely unbrushedin primary tracts.As the lengthof theensheathedportion declinedwith continuedbrushing,partial correlationsbetweenTband ATband pin lengths ceasedto differ significantlyfrom zero (Table 4). Although brush length increasedduring thisperiodandsimplecorrelationwsith feather brush length were significant,neither Tbnor ATbwere correlatedwith brushlengthwhen age,body mass,and total featherlengthwere controlled(P > 0.1 for all partials).That suggeststhatalthoughthermoregulatorimy provementscoincidewith plumagematurationand brush development,they are not contingent uponthosechangesb, ut dependinsteadupon the maturationof other physicaland physiologicalfactorsthat also coincidewith feather development.
Developmenotfeffectivendothermy.--Inisolated nestlings,developmenot f endothermywas relativelyslow,with little ability to maintain stableTbat adult levelsuntil day 10or 11.However,wenotedvisibleshiveringandthusactive attemptsto regulate temperatureas early as day 5. Becausemost nestlings occurred in broodslargerthanone,theextentto whichdevelopmentof homeothermywasinfluencedby presenceand number of brood mateswas addressedby measuringtemperaturesof chicks within thenest,immediatelyuponflushingthe female(Tbrooaa_n1)d, thenagainafter 10min of cold-exposur(eT•, Tbfor chickscold-stresseidn

isolation, Tbroo2a_for chicks cold-stressed in
broods). Meantemperaturesfor chickstestedindivid-
ually in thenest(Zbrooaa_nl)dthensubsequently cold-stressedin isolationare shownat the top of Figure 2A, and again after 10 min of exposurein isolation(TsandTb).Althoughobtained immediatelyafter flushingthe female,means for Tbroowaqerebelowthe averagefemalebody temperatureof 42øC,andincreasedsignificantly overthe nestlingperiod,from 31.1 _+5.5øC in day0 nestlingsto 41.2 + 0.7øCby day 13(linear regression,Zbroo1a=_ 0.427 (Age) + 35.587,
r = 0.71, F = 157.508, df = I and 157, P <
0.0005).By days5 and 6, mean Tbroo1ao_f 38.2 + 1.0øCdid not differ significantlyfrom those measured on subsequentdays (ANOVA, Tukey'stest, P > 0.05).
The reliability of Tsas an approximationof Tbwastestedin isolatednestlingsaftercoldexposure,when the differencesshouldbe magnified. Althoughskin temperaturesafter cold exposurewere significantlylower than core temperatureswhen isolatednestlingsof all ages were pooled (mean difference: 0.7 _ 1.5øC,Wilcoxon-signedranks test, P < 0.005), the differencesfor mostageswerenot statisticallysignificant(Wilcoxon-signerdankstest,P > 0.05in 8of 14agegroups)andwerelessthan 1øCin four of the agegroupsin whichsmall differenceswerefound(Wilcoxon-signerdanks test, P < 0.05). When age,mass,and Tawere controlled,these differenceswere negatively correlatedwith the extentof dorsalbrushing (Mdb:R = -0.26, df = 116, P = 0.004; P6B:r = -0.21, df = 116, P = 0.020).The closecorrela-