Mimus Polyglottos)

J. Field Ornithol., 58(3):297-305

FLUCTUATIONS IN FALL AND WINTER TERRITORY SIZE IN THE NORTHERN MOCKINGBIRD (MIMUS POLYGLOTTOS)

CHERYL A. LOGAN Departmentof Psychology Universityof North Carolina,Greensboro Greensboro,North Carolina 27412 USA

Abstract.--Many speciesof birdstrack resourcescontained in their feedingterritory. Ob- servationsreported here illustratefluctuations in territory sizefrom fall to winter in peren- niallyterritorial mockingbirds in the piedmontof North Carolina.In coldweather, territory size decreasedfrom fall to winter. The decreasewas neither a competitiveloss of spaceto intrudingconspecifics, nor a directresponse to the competitivethreat posed by otherspecies relying heavily on winter fruit. During the unusuallywarm winter of 1984, territory size decreasedlittle if at all. The decreaseobserved in coldweather may, therefore, be a response to alteredenergetic demands imposed by cold weather in which decreasedterritory size minimizesenergy expenditure. These observationsstress the flexibility of the mockingbird's perennialterritoriality in the face of environmentalvariation.

FLUCTUACIONES EN EL TAMAlqlO DEL TERRITORIO DE SINSONTES ~ (MIMUS POLYGLOTTOS) DURANTE EL OTONO Y EL INVIERNO Sinopsis.--Muchasespecies de aveseva16an los recursosque hay en su territorio de ali- mentaci6n.Las observacionesque hice en sinsontesen Carolina del Norte, ilustran fluc- tuacionesen el tamafiodel territorioperemne de estasaves durante el otofioy el invierno. E1 territorio disminuy6en tamafiodel otofioal invierno.La disminuci6nno estuvoasociada a perdida de espaciopor competenciacon otros sinsonteso comorespuesta a la amenaza competitivapor otras especiesde avesque dependenen gran medida de frutas durante el invierno.Durante el "tibio" inviernode 1984, losterritorios apenas se redujeron en tama•o. Por endela disminuci6ndel territorioobservada durante inviernos fr•os muy bien pudiera ser una respuestaa la demandaenerg6tica impuesta por las bajastemperaturas en donde la reducci6ndel tamafio del territorio reduce tambi6n el gastoenerg6tico. Estas observaciones recalcan la flexibilidad del territorio peremne de los sinsontescon relaci6n a variables climato16gicas. The adaptivebenefit of territoriality dependsupon a net advantage that accruesto the residentthrough exclusiveaccess to resourcescontained in the defended area (Brown 1964, Davies and Houston 1984, Schoener 1983). Changes in the costsof defense,the value of resources,or the energeticdemands of maintenancemay alter patternsof territorialdefense over both the long and short term (e.g., Carpenter and MacMillen 1976, Hixon et al. 1983). Changesin patternsof territorial defenseshould be especiallypronounced in speciesexhibiting year-round territoriality in a variableenvironment. Differences in resourcequality or distribution,in weather, and in the nature of competitorsmay producealtered patterns of territoriality.The observationsreported here chart the courseof changes in the fall and winter territorial behaviorof the Northern Mockingbird (Mimus polyglottos). In the SouthernUnited States,mockingbirds exhibit year-rounddefense of an all-purposeterritory. Throughout the year most of the residents' activitiesare focusedon the defendedarea (Laskey 1936, Michener and

297 298] C.A. Logan j. FieldOrnithol. Summer 1987

Michener 1935). However, patternsof resourceuse may changeconsid- erably. For example,demands imposed by breedingare relaxed in the fall and winter, when benefits of territorial defenseaccrue primarily through food availability (Hailman 1960, Laskey 1936, Michener and Michener 1935). Competitivepressures change as well. In the fall, the number of conspecificintruders increases with the large influx of strange conspecificsattempting to establishterritories (Breitwisch et al. 1986, Laskey 1936, Logan et al. 1983, Michener and Michener 1935). In addition, competitionwith other speciesmay increasein the winter, as mockingbirdsvie for increasinglyvaluable fruit (Moore 1978). Such changesmay substantiallyalter patternsof resourcedefense.

METHODS I observedresident wild mockingbirdsin Guilford County, North Car- olina, in the fall and winter of 1981 and 1984-1985. All territories were locatedin residentialor urban areason or around the University of North Carolina at Greensboro.This area providesabundant resourcesfor resident mockingbirdsthat remain in the area year-round. With minor shifts in territorial boundaries,many residentsappear to remain in the same area for life. I engagedin systematicfocal animal sampling(Altmann 1974) of 12 males in three sampling periods in 1981 and 1984-1985. Six were observed in 1981 and 6 in 1984-1985. In 1981 observations were collected from 5-27 Sept., from 17 Oct.-7 Nov., and from 28 Nov.-19 Dec. All 6 were color-bandedand with one exception,each bird sharedhis territory with a female. In 1984-1985, observations were conducted from 2-20 Oct., from 30 Nov.-22 Dec., and from 14-31 Jan. Of the 6 birds monitored, 3 were mated and 3 were unmated. One of the mated males and 3 of the unmatedmales were color-banded.Conspecific aggression peaks during the early fall (Laskey 1936). Therefore, to distinguishconspecific aggressionfrom that directedtoward other speciesdifferent observation periods were used in 1984-1985. Each sampleconsisted of 30 min of continuousmonitoring of the bird's behavior.During this period, the locationsof all activitieswere recorded on a map drawn to scalerepresenting each bird's territory. I alsorecorded the temporalsequencing of all behaviorobserved. The behavioralpatterns monitored included conspecificfights, chasesand the ritualized dance (Hailman 1960), aggressiveinteractions with other species,all vocaliza- tions, and all major perch changesand foragingpositions. Because fights and chaseswith conspecificintruders occurredthroughout the occupied space,I defined the occupiedspace as a true territory rather than as a home range. Total numbers of samplesobtained in 1981 for each bird in each sampling period were 32, 32, and 18, respectively,totalling 41 h/bird. In 1984-1985, samplestotaled 12, 10, and 12/bird, respectively for the three samplingperiods. Total sampletime was 17 h/bird. As mockingbirdsare accustomedto the presenceof humans, no blinds were used. However, observersmaintained a distance of at least 5 m from the vol.58, •o. • Fall Territorialityin Mockingbirds [299 focalbird. Observationswere evenly distributed throughout each sampling period,and nearly equal numbersof sampleswere obtainedduring the morning(0800-1200) and afternoon(1300-1700) hours.Territory sizes were calculatedby determiningasymptotic observation-area curves as definedby Odum and Kuenzler (1955). This methodadds the observed area acrosssuccessive samples to determinethe point at which additional samplesadd no more than 1% to the total area observed.The 1% criterion was usedin 1981; in 1984, thoughfewer sampleswere used,the number of sampleswas sufficientto ensure that no more than a 5% increasein area was addedwith additionalsamples. Exceptions are notedbelow. An electronicplanimeter was usedto calculateterritory areas.

RESULTS Changesin territory size.--Observation area curvescalculated from September 1981 forward acrosssuccessive samples through December 1981 and from December1981 backwardacross successive samples in- dicatedstable changes in the sizeof the 1981territories. Figure 1 illustrates thesechanges in 2 of the 6 males. In bird C, two different asymptotic sizesare apparent, one in the fall (calculatedfrom Septemberthrough December)and onein the winter (calculatedfrom Decemberto Septem- ber). Bird X exhibitedtwo fall asymptotes,both considerablylarger than the asymptoticsize of the Decemberterritory. A winter decreasein territory size distinguishingthe larger fall space from the smaller winter spacewas observedin 5 of the 6 birds sampledin 1981. In these5 birds the December spacewas, on the average, over 50% smaller than the Septemberspace (Table 1, P < 0.06, two-tailed, Walsh Test [Seigel 1956]). The locationof the 6th bird's spaceshifted continuously throughout the fall and winter of 1981, and no asymptoticvalue could be obtained. The pair disappearedfrom the area in the springof 1982; no data from this pair are includedin the analysesbelow. The early fall representsthe period of greatestconspecific territorial intrusionin mockingbirdsinhabiting the SouthernUnited States(Laskey 1936, Logan et al. 1983, Michener and Michener 1935). It is, therefore, possiblethat the winter lossin territory sizereflects a competitiveloss of spaceto conspecifics.Two observationssuggest that this is not the case. First, the territoriesof three adjacentneighbors each shrank away from the other, and no bird occupiedthe interveningspaces. In the following spring,each bird's territory expandedto includethe vacatedwinter space. Secondly,in 1981, the fall influx of intruding conspecificspeaked before measurabledecreases in territory sizeswere observed.The number of conspecificintrusions was reducedby 50% during the Octobersampling period,though sizeable decreases in territory size were not evidentuntil mid-November. Becausefewer sampleswere conductedin 1984, direct comparisons cannotbe made betweenterritory sizesin the two years.However, sizes calculatedin 1984 were within the range of variation observedin 1981, and in all but two cases(see Table 1) in 1984 an averageof lessthan 300] C.A. Logan J.Field SummerOrnithol. 1987

bird C

.60 1

.30- .30

bird X

.60-- .60 2nd Ist J .30

September •) • December to criterion to criterion SUCCESSIVE 30 MIN. SAMPLES (1981) FIGURE1. Asymptoticallylarge and small territory sizes calculated in 1981 from September forwardthrough December and December backward across successive samples. Samples obtainedfrom bird G indicateone large and one small territory asymptote. Samples obtainedfrom bird X indicatetwo successive large asymptotes and one small asymptote.

5%of the asymptote was added to the total area in thelast three samples obtained.Comparison ofthe differences in territory size in Octoberversus December,1984 (Table 1), however,indicates considerable variability amongsubjects. Across birds, the size of theDecember territory ranged from23.5%-107.7% of the Octoberterritory size. The changein territory sizewas not highly reliable (0.06 < P < 0.094,two-tailed, Walsh Test). Observationscollected during the January sampling period did, however, reveala significantincrease in Januaryover December territory sizes (P < 0.03,two-tailed, Walsh Test). January territory sizes ranged from 117.4%-300.0%of Decemberterritory sizes (Table 1), and all birds remained stableresidents into the spring. Competitivepressure from other species.--The frequency of conspecific intrusionsdecreases during the late fall andwinter in mockingbirds(Las- key1936, Logan et al. 1983).However, Moore (1978) reports that in the wintermockingbirds selectively attack and drivefrom fruitingre- Vol. 58, No. 3 Fall Territorialityin Mockingbirds [301

Comparisonof changesin territory sizeacross fall and winter samplingperiods.

Territory size (ha)a in 1981 Bird Sept. Dec. Dec. as % of Sept.

B 0.47 0.23 48.9% E 0.62 0.25 40.0% X 0.37 0.25 67.6% C 0.72 0.14 19.4% S 0.49 0.34 69.4% Mean 0.53 0.24 49.1% SD 0.14 0.07 20.7%

Territory size (ha)a in 1984/1985 Bird Oct. Dec. Jan. Dec. as % Oct. Jan. as % Dec.

B2 0.34 0.09 0.11 26.5% 117.2% EF 0.32 0.24 b 0.35 75.0% 145.8% PB 0.57 0.56 0.89 98.2% 158.9% Z 0.26 0.28 0.40 107.7% 142.8% G 0.56 0.13 b 0.39 23.2% 300.0% L 0.29 0.16 0.19 55.2% 118.8% Mean 0.39 0.24 0.39 64.3% 163.9% SD 0.14 0.17 0.27 35.6% 68.6%

Conversionto hectareswas performedfollowing statistical analyses. Size criterion in last three samplesis <5% added area. sourcesother specieswhose diet consistsprimarily of fruit. Therefore the Decemberdecrease in territory size observedin 1981 may representa shiftfrom fall defenseof a largeall-purpose territory against conspecifics to winter defenseof a smallerfeeding territory againstthe threat posed by other speciesof frugivores.As the threat of intrusionby mockingbirds declines(Laskey 1936, Logan et al. 1983) and as other speciesuse increasinglyvaluable fruit (Moore 1978), territory defensemay be concen- trated on smaller areas containingfruit. To determinethe relationshipbetween variations in territory size and the frequencyof fightswith other species,I examinedthe occurrenceof fightsand chasesdirected at other species.The mean number of fights per sampledid not changereliably from fall to winter in either 1981 or 1984 (P > 0.06, one-tailed,Walsh Test). Combiningdata on birds observedin 1981 and 1984, I calculatedthe correlationbetween the change in a bird's winter territory size and the fall versuswinter differencein the averagenumber of fightsper samplewith other species.The correlation was not statisticallysignificant (Spearman rho [n = 11] = -0.349, P > 0.14). Birds with the greatestdecrease in territory size did not necessarilyengage in more fightswith other species(Table 2). However, there was a significant positive correlation between the mean number of fightswith other speciesobserved for eachbird in the fall and again in the winter (Spearmanrho [n = 11] = +0.534, P < 0.05). This suggests 302] c.A. Logan j. FieldOrnithol. Summer 1987

TABLE2. Relationshipbetween winter/fall ratios of mean intrusions/30 min and the winter/fall ratios of changein territory size.

Intrusionsa Territory size Bird (w/f) (w/f)

C 3.67 0.19 X 55.00 0.68 S 0.39 0.69 B 1.23 0.48 E 1.16 0.40 B2 1.21 0.26 EF 0.96 0.75 PB 1.25 0.98 Z 2.50 1.08 L 3.75 0.55 G 0.48 0.23

Includesfights and chaseswith other speciesof birds. that the number of fightswith other speciesis a stablecharacteristic of either individual mockingbirdsor their territories.

DISCUSSION The observationsreported here indicate that the mockingbird'sterritory sizemay fluctuateconsiderably from fall to winter. Moore's (1978) observationssuggest that mockingbirdscompete selectively with otherspecies for accessto winter fruit. My observationsindicate that in 1984, over 85% of aggressiveencounters with other specieswere directedtoward speciesranked by Moore as exhibitinga high percentageof frugivory. However, thoughmockingbirds compete with frugivoresin the fall and winter, becauseaggression toward other speciesdoes not changesystem- atically with changesin territory size, competitionwith frugivoresdoes not appearto producechanges in winter territorysize. Decreased territory size may facilitate efficientcompetition for fruit. If territoriesshrink, mockingbirds'activities may be focusedon the area of greatestfrugivore pressure.That thisneed not be the caseis indicatedby two birdsin which aggressiontoward other speciesincreased in December(1984) though territory size either remainedthe sameor increasedfrom fall to winter (see Table 1). In 1981, the territory sizesof all stableresidents decreased to an average of 48.6% of the fall space.In 1984, on the other hand, somedecreased, somedid not changeand someincreased. What accountsfor the variation in territory sizechange across seasons ? Variations in winter weather are known to affect foragingand spacingstrategies in severalavian species (Gassand Montgomerie1981, Grubb 1975,Jorde, Krapu and Hay 1984, Schmidtke and Brandl 1983). Both behavior and metabolism can be affectedby changesin temperatureand irradiation (DeJong 1979). In the piedmontof North Carolina, the mean temperaturedrops approxi- mately 16ø C from Septemberto December,and virtually all resident v,•l.•8, •o. • Fall Territorialityin Mockingbirds [303 mockingbirdsoverwinter. The 1981 shrinkagein winter territory size mayreflect increased thermoregulatory demands associated with dropping temperatures. I examined recordsof daily weather changesfor the months of Sep- tember through Decemberin 1981 and 1984. A correlatedt-test showed that daily temperaturesfor December 1984 were considerablyhigher than the 30-yr norm for that month, while daily temperaturesfor De- cember1981 were somewhatcolder than the 30-yr norm (Fig. 2, ? < 0.01). The averagedaily deviationfrom normal in 1981 was -1.64 ø C; in 1984, the averagedaily deviationfrom normal was +4.59 ø C. Direct comparisonof the average daily temperaturesfor only those days in December, 1981 versusDecember, 1984 on which behavioralsamples were taken indicatessignificantly warmer averagetemperatures in 1984 (6.03ø C versus2.42 ø C, P < 0.05, F-test). Neither month had appreciable amounts of snow or ice, and the months do not differ from one another in the averagenumber of minutesper day with sunshine. A consistentdecrease in the mockingbird'sterritory size was evident only in the colderyear. This suggeststhat mockingbirdsmay adaptwinter territorial behaviorto variation in ambient temperatureconditions. The temperature drop associatedwith colder winters may place a premium on decreasedactivity as thermoregulatorydemands increase. Behavioral adaptationto harsh winter weather has been describedin other avian species.During the coldestperiods of the Nebraskawinter, for example, mallardscongregate in warmer canal habitats,returning to riverine areas when temperaturesrise. Jorde et al. (1984) suggestthat mallards select a microclimatethat minimizesenergy expenditure for thermoregulation. In mockingbirds,smaller winter territoriesmay reflecta similar weather- driven thermoregulatorystrategy. Activity required to patrol the larger area is less and less necessaryas conspecificintruder pressuredeclines. When the temperaturedrops, as in December,1981, decreasedactivity acrossa smaller area may reduceenergetic demands, thereby decreasing the costsof thermoregulation. Observationscollected in the winter of 1984 suggestthat as constraints imposedby the drop in winter temperaturesare relaxed, individual differencesin age,experience and matingstatus may contributeto differences in territory size. Though they report no differencesin territory size, Breitwischet al. (l 986) observeddifferent patterns of fall territory defense in mated versusunmated male mockingbirdsin Florida. Comparisonsof the size changesin the 3 mated versusthe 3 unmated males I observed in 1984, reveal lesschange in the unmated males (29.8% versus41.1% decrease).Because unmated males may acquire mates in the fall (un- publishedobservations), warmer winters may extendthe period of mate attraction, possiblyentailing larger territories in unmated males. Inter- actionsbetween two neighboringunmated males, both new to the area, suggestthat experiencetoo, may be an important factor.As observations proceededfrom Octoberthrough December,one bird's (EF's) spaceshifted into that held by another(B) whoseterritory shrankconsiderably: the 304] C. A. Logan j. Field Ornithol. Summer 1987

30-Year •_._•/•X• • Nor•

1984: :

1981 ß l

30yr Norm: --

I I I I 10 20

SUCCESSIVE DAYS (DECEMBER) FIGURE2. Averagedaily temperaturesin the piedmontof North Carolina,for December 1981 (& &) and December 1984 (•). The third line (•) indicatesthe 30-yr temperaturenorm for each Decemberday in the piedmontof North Carolina. Data were providedby the National Weather Service,Greensboro/Winston-Salem/ High Point RegionalAirport.

Decemberloss of spaceby B was a result of spaceyielded to EF. Bird EF did not expand his spaceinto those occupiedby his three other neighbors,each a long-time establishedresident. Rather, he directedhis intrusionstoward his least experiencedneighbor. Together the above observationsemphasize the dynamic nature of mockingbirds'fall and winter territories.As might be expectedof year-roundterritoriality in a variableenvironment, territorial behaviorappears to be finely tuned to local changesin environmentaldemands.

ACKNOWLEDGMENTS

I thank Patti Budman,Laura Hyatt, Kimberly Johnson,and Linda Walker for their invaluablehelp in the collectionand analysisof data.Reed Hunt, TimothyJohnston, and RobertLickliter provided extremely helpful suggestions on an earlierdraft of themanuscript. I am especiallygrateful for the constructivecomments provided by Frank Moore and an anonymousreviewer. Use of the planimeterwas providedby the UNC GreensboroDe- partmentof Geography.This researchwas supportedby the Universityof North Carolina Research Council.

LITERATURE CITED

ALTMANN,J. 1974. Observationalstudy of behaviour:sampling methods. Behaviour 49: 227-265. BREITWISCH,R., M. DIAZ, N. GOTTLIEB,R. LEE, AND J. ZAIAS. 1986. Defense of fall territoriesby matedand unmatedNorthern Mockingbirdsin SouthernFlorida. J. Field Ornithol. 57:16-21. BROWN,J. L. 1964. The evolutionof diversityin arian territorial systems.Wilson Bull. 76:160-169. Vo•.•8, No. • Fall Territorialityin Mockingbirds [305

CARPENTER,F., ANDR. MACMILLEN. 1976. Thresholdmodel of feedingterritoriality and test with a Hawaiian honeycreeper.Science 194:639-642. DAVIES,N. B., ANDA. I. HOUSTON. 1984. Territory economics.Pp. 148-169, in J. Krebs and N. Davies, eds. Behaviouralecology: an evolutionaryapproach, second edition. Blackwell Scientific, Oxford. DF.JONG, A. 1979. Thermal irradiationpreferences of White-crownedSparrows in relation to ambient temperature.Physiol. Behav. 22:157-171. GASS,C. L., ANDR. D. MONTGOMERIE.1981. Hummingbird foragingbehaviour: decision- making and energy regulation.Pp. 159-194, in A. C. Kamil and T. Sargent,eds. Foragingbehavior, ecological, ethological and psychologicalapproaches. Garland Press, New York. G•tUBB,T. C. 1975. Weather-dependentforaging behavior of somebirds wintering in a deciduous woodland. Condor 77:175-182. HAILMAN, J. 1960. Hostile dancing and fall territory of a color-bandedmockingbird. Condor 62:464-468. HIXON, M. A., L. CARPENTER,AND O. C. PATON. 1983. Territory area, flower density, and time budgetingin hummingbirds:an experimentaland theoreticalanalysis. Am. Nat. 122:366-391. JORDE,D.G., G. L. KRAPU,AND m. HAY. 1984. Effectsof weather on habitat selection and behavior of mallards wintering in Nebraska. Condor 86:258-265. LASKE¾,A. 1936. Fall and winter behavior of mockingbirds.Wilson Bull. 48:241-255. LOGAN,C.A. 1983. Reproductivelydependent song cyclicity in matedmale mockingbirds. Auk 100:404-413. LOGAN,C. A., P. BUDMAN,AND K. R. FULK. 1983. The role of chatburstversus song in the defenseof fall territory in mockingbirds(Mimus polyglottos). J. Comp. Psychol.97: 292-301. MICHENE•t, H., AND J. MmHENg•t. 1935. Mockingbirds, their territories, and individu- alities. Condor 37:97-140. Moo•tg, F. 1978. Interspecificaggression: toward whom should a mockingbirdbe ag- gressive?Behav. Ecol. Sociobiol.3:173-176. MYERS,J.P., P. G. CONNER,AND F. A. PITELKA. 1981. Optimal territory size and the sanderling:compromises in a variable environment.Pp. 135-158, in A. C. Kamil and T. Sargent,eds. Foraging behavior, ecological, ethological and psychologicalapproaches. Garland Press, New York. ODUM, E., AND E. KUENZLER. 1955. Measurement of territory and home range size in birds. Auk 72:128-137. SCHMIDTKE,K., ANDR. BRANDL. 1983. Winterrevier und ansitzstratgiedes Raubwurgers Laniusexcubitor. Verh. orn. Ges. Bayern 23:443-458. SCHOgNE•t,T.W. 1983. Simplemodels of optimal feeding-territorysize: a reconciliation. Am. Nat. 121:608-629. SgIGgI•,S. 1956. Non-parametric statisticsfor the behavioralsciences. McGraw-Hill, New York. Received21 Aug. 1986; accepted4 Feb. 1987.