Thermoregulatory Role of the Unfeathered Head and Neck in Male Wild Turkeys

The Auk 113(2):310-318, 1996

THERMOREGULATORY ROLE OF THE UNFEATHERED HEAD AND NECK IN MALE WILD TURKEYS

RICHARD BUCHHOLZ • Departmentof Zoology,University of Florida Gainesville,Florida 32611, USA

AI•STRACT.--Thebrightly colored,unfeathered heads and necks of male Wild Turkeys (Meleagrisgallopavo) are generallythought to functionin sexualselection. However, studies in other bird specieshave suggestedthat uninsulatedbody regionsmay serve an important role in heat dissipation.I test the heat-dissipationhypothesis in Wild Turkeysby experi- mentally reinsulatingthe headsand necksof Wild Turkeysas though they were feathered. The oxygenconsumption, thermal conductance,cooling capacity, surface temperatures, and core temperatureof control and reinsulatedWild Turkeyswere comparedat 0ø, 22 ø and 35ø(2. Head insulationresulted in significantlyincreased rates of oxygenconsumption, higher body temperatures,and decreasedcooling capacitiesat 35øC,but had no significanteffect at the other temperaturestested. It appearsthat behavioral changesat low temperatures,such as tucking the head under the back feathers,effectively prevent the heat lossthat would oth- erwise be causedby the absenceof feathers.However, if the head were feathere& turkeys at high temperatureswould be unable to dissipatesufficient heat to maintain thermeostasis. Thus,given this finding for Wild Turkeys,it canno longerbe saidthat in all casesbare heads in birds have evolved by sexualselection alone. Lossof head and neck featbering in Wild Turkeysand other birdsmay have allowed thesespecies to take advantageof regionsin time and spacethat previouslywere unexploitabledue to the dangersof hyperthermia.Received 22 June1994, accepted 27 January1995.

ENDOTHERMSUSUALLY MAINTAIN body tem- ported by correlativestudies showing that un- peraturesabove environmental temperatureat featheredhead and neck skin is maximally ex- considerableenergetic cost. To saveenergy they posed at high temperaturesand that in some reduceheat lossto the environmentby insu- taxa the size of unfeatheredareas is greater at lating themselvescompletely with fur or feath- low latitudeswhere heat dissipationmay be of ers. Birds that have areas of unfeathered skin greater importance (Crowe 1979, Buchholz on their headsand necksare an unexplained 1994). Highly vascularizedfleshy ornamenta- exceptionto the pattern of completeinsulation tion presentsa functional puzzle when species seen in other endotherms.In carrion-feeding are distributedover a large latitudinal range in birds, unfeathered heads often are assumed to which they are exposed to both temperature be a hygienic adaptation (Welty 1975: 100). extremes.Although these speciesmay benefit However, in specieswhere the unfeatheredar- by usingtheir fleshystructures to dissipateheat easare alsobrightly colored,sexually selected under hot conditions, the uninsulated nature functionsare usuallysuspected (Zuk 1991). of thesestructures subjects the birds to extreme Despite some studies suggestingthat these heat loss under cold conditionsand heat gain areas of bare skin maintain sublethal brain tem- in the presenceof solar radiation. In this study, peraturesby dissipatingheat via cephalo-cer- ! testthe possiblethermoregulatory function of vical retes (Crowe and Crowe 1979, Crowe and unfeathered head ornamentation in a species Withers 1979, LaRochelie et al. 1982), thermo- that commonlyfaces extremes of cold and heat, regulatory hypothesesfor the evolution and the Wild Turkey (Meleagrisgallopavo). maintenanceof unfeatheredskin rarely are con- Wild Turkeys occur over a broad range of sidered.The heat-dissipationhypothesis is sup- temperatureextremes from their southernlimit in southern Mexico to their northern limits along the border of the United Statesand Can- • Present address:Department of Biology, North- ada (Dickson 1992). Males are twice as large as eastLouisiana University, Monroe, Louisiana71209, females,and have brightly coloredunfeathered USA. heads and necks (Buchholz 1995). In addition,

310 April 1996] BareHeads and Thermoregulation 311 this bare skin is coveredwith polyp like elab- was drawn through the metabolicchamber, pumped orationsof the integument called caruncles.A into glasscolumns filled with soda lime (to remove thin dewlap extendsfrom the mandible down CO2)and silicagel (to removeH20), after which flow to the neck. Perhapsmost distinctive is the bare, rates (g = 20.6 L/min) were measuredby a Brooks distensiblefrontal processor snoodthat pro- $ho-Rateflowmeter. Subsequently, the airstreamwas sampledwith an Applied ElectrochemistryS-3A oxy- jectsfrom the foreheadat the baseof the upper bill. gen analyzer. The temperatureand humidity of the room air varied little (23.5 + SE of 0.0øCand 61.2 _+ Fleshyhead ornamentation in Wild Turkeys 0.2%,respectively). Humidity in the chamberwas not and other galliforms often is thought to be controlled. The bird's evaporativewater losswas mea- maintained by sexual selection (i.e. the struc- suredgravimetrically (i.e. by weighing the silicagel). tures function in mate choice and male-male Core body temperaturewas measuredby insertinga competition).Ample empirical evidence sup- copper-constantanthermocouple, tipped with a thin portsthis contention (Brodsky 1988, Boyce 1990, layer of silicone, into the bird's intestine to a depth Hillgarth 1990,Ligon et al. 1990,Zuk et al. 1990a, of 20 cm from the cloacalopening. This measurement b, Spurrier et al. 1991, Zuk et al. 1992, Buchholz was taken immediatelybefore the subjectwas placed 1995). A role in sexual selection, however, does in the metabolic chamber and immediately after it not rule out concurrent functions for these was removedfrom the chamber.Six surface-temperature measurementswere taken: feather, leg, body structuresin thermoregulation.Although both skin, head skin, frontal caruncle,and dewlap. Surface speciesof present-dayturkeys (Meleagridinae) temperatureswere measuredwith a bare-tippedther- have unfeathered heads and necks, the common mocouple held against the appropriate spot, while ancestorsof modern turkeys presumablyhad the subjectwas still in its holding box before the trial featheredheads, as do mostgalliforms. Under- and, again, while it was in the metabolicchamber at standingwhy the unfeatheredareas of modern the end of a trial. Skin and feather surfacetempera- turkeysare maintained, despite the possiblecosts tureswere measuredon the chestapproximately 3 cm in termsof heatloss, may explainwhy ancestral ventral to the carpal joint of the wing at rest. Leg temperaturewas measuredimmediately posteriorto turkeys lost their head feathering over evolu- the third scaledistal to the tarsaljoint on the left or tionarytime. In this studyI experimentallyin- right leg, depending on which was accessible.Head sulatethe headsand necksof Wild Turkeysto skin temperature was measuredon the back of the assessthe thermoregulatory trade-offs that head at a point posteriorto the lower mandible. Sur- maintainunfeathered heads in this species. facetemperature of the frontal carunclesand dewlap were measured at the centers of these structures.

METHODS Different rates of physical activity acrosssubjects and trials can make it difficult to detect the effect of Subjectsand equipment.--Eighttwo-year-old, male experimentaltreatments on metabolicrate. Therefore, Wild Turkeysobtained as chicksfrom a game farm I minimized the bird's activity by conductingtrials (L&L Pheasantry,Hegins, Pennsylvania)were used at night in the dark. Metabolictrials lasting2.5 h were in the metabolictrials. Rearing conditionsare de- conductedbetween 1900 and 0200 EST. All subjects scribed in detail elsewhere (Buchholz 1994). There is were given at leastone day betweentrials. Individual no differencein the metabolicrates of Wild Turkeys turkeys were tested at the same time of day (either from game-farmor free-living sources(Gray and 1900 or 1100) acrossall treatments to minimize cir- Prince 1988). The averagebody massof the study cadian effects on matched comparisonsof metabolic individualswas 7.1 kg (range6.4-8.1 kg). During the rate. The first 30 min of eachtrial servedas an equil- study period (29 June-27 September1993) the birds ibration period during which the bird calmed down were provided with feed (Purina Gamebird Mainte- after handling. The lowest rate of oxygen consump- nance,12% protein) and waterad libiturn. Subjects were tion (correctedto standardpressure and temperature) deniedfood for 26 to 29 h immediatelyprior to each measuredduring eachof the four subsequent30-min metabolictrial to insure that they were postabsorp- periodswas used to calculatean averagemetabolic tive. Postabsorptiveconditions are necessaryto mea- rate for the entire trial. All individuals were given surethe basalor minimum rate of metabolism(McNab two 2.5-h habituationtrials prior to the experimental 1988a).Water was still availableduring the pre-trial trials.Usually, the subjectsrested quietly during the period. experimentaltrials. The following threebehavior pat- Oxygen consumption and total water loss were ternswere recordedas presentor absentthrough in- measuredin an open system(described by McNab stantaneoussampling (Martin and Bateson1986) ev- 1988b).The temperatureof the 329-L metaboliccham- ery 30 min: standing; head tucked under feathers; ber was regulatedby pumpingwater from a water panting. Observationswere made with the aid of a bath through the chamber'shollow wails. Room air flashlight through a small window in the chamber. 312 RICHARDBUCHHOLZ [Auk,Vol. 113

Experimentaldesign.--To evaluate the potentialther- of eachturkey in the experimentaltreatment to the tooregulatoryimpact that headfeathering would have values obtained from the same bird in the control on Wild Turkeys,I determinedthe thermal balance treatment, which serves to minimize the effects of when the turkey's head was "bare" (seebelow), and interindividual variation on the effect of the experi- when it was insulated as though feathered. To ap- mental treatment.Due to schedulingconflicts in the proximatethe insulatoryproperties of head and neck laboratory,every turkey was testedat 0øCbefore it feathering,the barehead and neckof the turkeywere was exposedto the other temperatures.The presen- coveredwith a doublelayer (0.6 cm on head, 0.9 cm tation order of the trials at 22 ø and 35øC, and treat- on neck)of acrylicsock (Adler "CasualAcrylic Crew"; mentsat 0%22 ø, and 35øCwas randomized.Repeated- 75% hi-bulk acrylic, 25% stretch nylon) with large measuresANOVAs (AbacusConcepts, Inc. 1989)were holes for the eyes and the entire bill. The nostrils used to test the effectsof body size (œ= 6.7 kg for were never covered.Any irritational effectsof the smalland 7.5 kg for large), chambertemperature, and insulatoryhead covering on metabolicrate were con- head insulation on: oxygen consumption(cm • 02' trolledby placinga hoodmade of thin, nylon netting g-'. h-•); coolingcapacity; total and dry thermalcon- with little insulatoryvalue on the headsof the "bare" ductances(mW.[cm•]-•.[øC]-'); and changesin body individuals.The controlhead net and insulatoryhead and surfacetemperatures (øC). The effect of each 30- sockingwere held in place with small alligator clips rain samplingperiod was also included when oxygen that attached to the back and chest feathers at the base consumptionwas the dependentvariable. Treatment of the neck. The efficacyof using head sockingto groups exhibited similar variances (F• tests; all approximatethe insulationprovided by normalfeath- P>0.05; Sokaland Rohlf 1981).Reported P-values were eringwas determined by studyingthe warmingcurves adjusted using Greenhouse-Geisserepsilon values of the feathered and unfeathered/reinsulated heads (AbacusConcepts, Inc. 1989). This techniqueconser- of domesticroosters (Gallus gallus; after Morrisonand vatively compensatesfor the useof repeatedmeasures Tietz 1957). by adjustingthe degreesof freedom. Thermoregu- Six dependentthermoregulatory variables may be latoryvariables are reportedas f + SE,as appropriate. affectedby head insulation.Metabolic rate, as indi- catedby oxygenconsumption (cm 3 O2.g-'.h-*), is a measure of the work the animal does to maintain RESULTS thermeostasis.The rate of evaporativewater loss(g/ h) is a measureof the heat lost via evaporation. Met- The massspecific rate of oxygenconsumption abolicheat production(Hm)and evaporative heat loss was significantly lower for large individuals (H,) can be converted to common units (roW/g) to comparethe coolingcapacity of the animalin differ- acrossall temperatures(œ = 0.4140 -+ 0.0070 vs. ent treatments.Cooling capacityis the bird's ability 0.4730 + 0.0130 cm•O2.g-•-h-'; Table 1). The to dissipatemetabolically produced heat by evapo- rate of metabolismwas not significantlydiffer- ration. It is expressedas a percentage,calculated as ent for uninsulatedand insulated turkeys at 0ø the heat lost by evaporationdivided by the heat pro- and 22øC.However, at 35 øC,insulated turkeys ducedby metabolism(100 H,/Hm; Calderand King exhibited a significantly higher average meta- 1974).Total thermal conductance(mW.[cm2]-'.[øC]-') bolic rate than uninsulated turkeys (Table 2, measuresall the heat lost by the animal, including Fig. 1). A significant, three-way interaction of evaporativeheat loss,and is the inverseof insulation. head insulation, temperature, and time period It is estimatedusing the values for heat production, suggeststhat the effectsof head insulation be- and ambient and body temperatures.Dry thermal conductanceis a measure of all nonevaporative means camemore pronouncedthe longer the subject of heat loss: radiation, convection and conduction. If was exposedto the chamber conditionsat hot totalconductance is exceededby heatproduction, heat temperatures. is storedin the tissuesof the animal and body tem- Uninsulated turkeys demonstratea signifi- peraturerises. Each of thesecomponents of thermal cantly greater ability than insulated turkeys to balancemay be varied by the animal to copewith dissipateexcess metabolic heat by evaporation increased head and neck insulation. at 35 øC,but not at lower temperatures(Fig. 2). Thermoregulatorytrials were conductedtwice for Total thermal conductance increased with tem- eachturkey at eachof threeambient temperatures (0 ø, perature. It alsowas greater for insulatedbirds 22ø, 35øC),one time as a control, and the other with overall (Table 1, Fig. 3), but this differencewas the turkey'shead insulated. These temperatures were chosento be below, within, and above,respectively, significantonly at the highesttemperature (Ta- the zoneof thermalneutrality (Gray and Prince1988). ble 2). Dry thermal conductancedecreased with The temperaturesalso are within the range that tur- increasingtemperature in the uninsulatedbirds keys experiencein the wild. I conducted48 trials. (Fig. 4). Insulated birds showed a similar pat- This matcheddesign compares the metabolicvalues tern of conductances at 0 ø and 22øC, but had April 1996] BareHeads and Thermoregulation 313

TABLE1. Partial results of repeated-measuresAN- OVAs showingstatistically significant sources of variation in the dependent measures(centered in bold) of thermal balance listed.

+1 +1 +1 +[ +1 +1 +1 +1 +1 +1 +1 Source of variation df F-value Oxygen consumption Size 1 5.96* Insul x Temp 2 4.16' Insul x Temp x Time 4 4.15' Cooling capacity Temp 2 135.00'** Insul x Temp 2 8.87* Thermal conductance (total) Insul 1 7.19 * Temp 2 367.00'** Insul x Temp 2 8.03* Thermal conductance(dry) Temp 2 9.45* Insul x Temp 2 7.96* Core body temperature Insul 1 6.62* Temp 2 22.10'* Insul x Temp 2 10.39' Leg temperature Temp 2 150.00'** Feathertemperature Temp 2 42.01'** Head-skin temperature Insul 1 12.09' Temp 2 16.28'** Frontal-caruncletemperature Temp 2 6.39* Dewlap temperature Temp 2 6.83* Insul x Temp 2 7.79**

*, P < 0.05; **, P < 0.01; ***, P < 0.001. significantly higher values than uninsulated birds at 35øC. Head insulation resulted in significantly greatercore-body temperature changes of birds at 35 øC,but not at the lower temperaturestested (Fig. 5). Head insulationserved to keep head skin warmer at 0øC(uninsulated, 31.2 + 1.0øC; insulated, 36.7 + 0.3øC), but did not result in significantlyhigher skin temperaturesat 22ø and 35øC.Insulated turkeys at 22øCwere significantly more likely to have increaseddewlap temperatures than uninsulated birds (uninsulated, 32.1 + 0.7øC; insulated, 35.2 + 0.3øC insulated),but thiswas not trueat 0ø or 35øC.Body skin,feather, frontal caruncle, and leg temperatures all increasedwith increasingambient 314 RICHARDBUCHHOLZ [Auk, Vol. 113

OoC 22ø(2 35• .4 OoC 22oc 35oc ß ß

ß ß 0.6

[3[3[3 [] [] ß []

0.5[] [] [] ß

[] []

[]

12345678 12345 678 2345 678 2345 678 Individual Turkeys Individual Turkeys Fig. 1. Oxygen consumptionof eight Wild Tur- Fig. 3. Total thermal conductancesof eight Wild keyswith headsuninsulated (empty squares) and then Turkeyswith headsuninsulated (empty squares) and with headsinsulated (filled squares)at three ambient then with heads insulated (filled squares)at three temperatures. Turkeys presented in order of in- ambient temperatures.Turkeys presented in order of creasedbody mass. increasedbody mass.

temperatures,but were not affectedby the insulation treatment (Table 1). OoC 2 2oc 35oc Across and within each temperature treat-

[] 150 [] ment,head insuiation had no effect on the proportion of instantaneousobservations during which the subjectswere standing, panting, or had their head tucked in back feathers or under [] [] the wing (Mann-Whitney U-tests,n = 16, all P > 0.05). Panting only occurredat 35øC.The lOO ß [] ß frequencyof panting was difficult to observe because the birds often held their necks for- ß ß ward and down so that the view from the small window was blockedby the bird's body. There- fore, it was not possibleto evaluate associations 50 ß [] between panting frequency and thermal balance. Nevertheless, upon opening the chamber at the end of the 35øCtrials, I observedpanting and an elongatedsnood (only visible in uninsulated trials) in all individuals. Snood elon- gation did not occur at other temperatures. 1234 678 12345 678 2345678 Although the proportion of observationsin Individual Turkeys which the subjectswere standing was not in- Fig. 2. Cooling capacitiesof eight Wild Turkeys fluenced by the insulation treatment, the fre- with headsuninsulated (empty squares) and then with quency of this behavior did have some effect headsinsulated (filled squares)at three ambient tem- on the dependent thermal variables. The fre- peratures.Turkeys presentedin order of increased quency of standing had a significant positive body mass. correlation with head skin temperatures at 0øC April 1996] BareHeads and Thermoregulation 315

0.4 OoC 22oc 35oc 0oc 22oc ß 35o(

[]

ß ß ß m•m[][] ß

ß [] [] o -0.2 m•um [] [] [][] [] ß

[] I- mm [] []•m [] m[] ß ß

[]

-0.6 2345 678 234 678 12345678 12345678 12345678 12345678

Individual Turkeys IndividualTurkeys Fig. 4. Dry thermal conductancesof eight Wild Fig. 5. Changein corebody temperatureof eight Turkeyswith headsuninsulated (empty squares) and Wild Turkeyswith headsuninsulated (empty squares) then with heads insulated (filled squares)at three and then with headsinsulated (filled squares)at three ambienttemperatures. Turkeys presented in orderof ambient temperatures.Individuals whose body tem- increasedbody mass. perature changed indentically when insulated and uninsulatedare representedby a squarecontaining a vertical line. Turkeys presentedin order of in- (both insulation treatments combined, rs= 0.54, creasedbody mass. P=0.05) and may be associatedwith lower rates of metabolicheat production,although the latter correlationonly approachesstatistical sig- temperature. Although the insulated birds nificance (r, = -0.50, P=0.07). At 22ø and 35øC, managedto increasedry and evaporativether- feather temperature inversely correlated with mal conductancesdespite the head insulation, standingfrequency (rs=-0.66, P=0.01 and rs= the lower cooling capacitiesof insulated birds -0.52, P= 0.05, respectively). isevidence of their inabilityto dissipateenough Head tucking occurredin 8 of 16 trials by six heat to offsetmetabolic heat production.These of the eight individuals at 0øC. Higher fre- results demonstrate that the unfeathered heads quenciesof head tucking were positively cor- and necksof male Wild Turkeys, and possibly relatedwith changesin dewlaptemperature (rs = the fleshy structureson the head, contribute to 0.50, P=0.05), and negatively correlated with heat dissipationat high ambienttemperatures. changesin skin temperature (r, = -0.58, P = Contrary to expectations,under cold con- 0.04), changesin core body temperature(rs = ditions head and neck insulationdid not sig- -0.54, P = 0.05),metabolic heat production(r, = nificantly reduce thermal conductance or in- -0.56, P = 0.05), and both total and dry con- creasemetabolic heat production. Under cold ductance (rs = -0.56, P= 0.05 and r, = -0.66, conditions,free-living Wild Turkeysoften con- P = 0.02, respectively). tract the skin at the back of their necks, effectively drawingthe featheredskin at the baseof DISCUSSION the neck up and over much of the usually bare areasof the back of the neck (pers. obs.). The A dramatic cost of insulated heads and necks captive Wild Turkeys in my study exhibited occursin male Wild Turkeysat high tempera- similar behavior, possiblyexplaining the ab- tures.Insulated birds had higher metabolicrates sence of a difference in thermal conductance and markedly increasedcore body tempera- between uninsulated and insulated birds at 0øC. tures relative to uninsulated males at the same Becausewinter starvation can be an important 316 RICHARDBUCm•OLZ [Auk, Vol. 113 source of mortality for turkey populations in ent temperature(ca. 30øC)and exposureto di- the northern part of their distribution (Healy rect sunlight may reduce male fertility by as 1992a),reducing heat lossfrom the head may much as 10%in broad-breastedbronze turkeys, enhanceturkey survivorship. At night, thermal the domesticbreed mostsimilar in appearance conductancemay be further decreasedby tuck- to Wild Turkeys(Kosin and Mitchell 1955).Wil- ing the head under the wing or back feathers. son and Woodard (1955) found that all domestic In my study, four of the eight uninsulated in- turkeys were subjectto hyperthermia at ambi- dividuals at 0øC were seen with their heads ent temperaturesabove 32øC; this was particu- tucked during at least one of the observation larly true of large males.In addition, body tem- periods,and three of thesehad lower metabolic perature and water consumptionby domestic rates than the remaining individuals. La- turkeyswere inverselycorrelated with the per- Rochelie et al. (1982) found a similar effect of cent of shade cover provided at ambient tem- head tucking in BlackVultures (Coragypsatra- peraturesabove 35øC (Wilson et al. 1955,Wilson tus), which also have unfeathered heads. Ad- and Woodard1955). Wild Turkeysexperienced ditional studies of the effects of cold environ- heat stressat 35øCin the laboratoryduring my mental conditions(e.g. low temperature,high study. All malesresponded to hot chambertem- wind speed,low insolation)and artificialhead peraturesby panting,dropping their wings,and insulationon headtucking and thermalbalance extending their necksand snoods.The smallest in Wild Turkeysare neededto confirm the ef- male even became frantic at the very end of ficacyof this tacticfor reducingheat loss. both high-temperaturetrials and was removed Anatomicaladaptations used to modify heat immediately. Behavioralchanges that occur in losshave been describedin other avian species. free-living wild malesunder hot conditionsalso Ptarmigan (Lagopusspp.), which live at high suggestthat activity is limited by high ambient latitudes and altitudes where the difference be- temperatures. tween body temperatureand ambient temper- Males call ("gobble")to attractfemales most ature can be large (e.g. > 60øC), often have often before dawn and during early morning featheredlegs and feet (Johnsgard1983). Gulls (Hoffman 1990). This is especiallytrue in the (Laridae) have counter-current heat-exchange presence of clear skies (Healy 1992b), when mechanisms that reduce heat loss from the feet males would risk greater heat and insolation under cold conditions (Baudinette et al. 1976). later in the day. Mature male Wild Turkeys in The Wood Stork (Mycteria americana)and Tur- northern Florida seem to avoid bright sun at key Vulture (Cathartesaura) use their unfeath- any time of the year and, in summer,often are ered legs to dissipateheat at hot temperatures found standingin heavy shadewith their dew- and are able to enhance this mechanism of heat laps and necksbright red and extended,while loss by defecating on their legs to promote panting heavily (pers.obs.). Also, the malesare evaporativeheat loss(Kahl 1963, Hatch 1970). more reluctant to flee under these conditions Ducks may use the large surfacearea of their and can be approachedmore closelythan when bills to dissipateheat (Hagan and Heath 1980). it is cooler.Females engaged in gular fluttering The Wild Turkey is the only speciesin which under hot conditions but did not seem dis- the value of unfeathered heads and necks for tressedby the heat. From these cursory obser- heat dissipationhas been demonstratedexper- vationsit appearsthat the sexesdiffer in their imentally. susceptibilitiesto heat stress,possibly resulting Previous studiesof Wild Turkey metabolism in fitnessconflicts between the sexesduring the have ignored the metabolismof Wild Turkeys breeding seasonwhen malesand femalesmust at temperaturesabove 25øC(Gray and Prince be together. 1988,Oberlag et al. 1990).The adaptivebenefit Males gobble on warm days in winter, sug- of unfeathered heads demonstratedhere suggestingthat they are more responsiveto warm geststhat peak effective temperaturesduring temperatures and less dependent on photope- the reproductiveseason, especially in habitats riod to stimulate breeding than are females without shade, may limit Wild Turkey distri- (Healy 1992b).Also, in most areasof their dis- bution or population density. These resultsare tribution, male turkeys exhibit a peak in gob- reinforced by early studieson the temperature bling behaviorseveral weeks before females nest requirementsof domesticturkeys. High ambi- (reviewed by Hoffman 1990).The degreeof sex April1996] BareHeads and Therrnoregulation 317 differences in reproductive timing may vary Threadgill DissertationFellowship from the College with latitude. Davis (1994: 117) found that Wild of Liberal Arts and Sciencesat the University of Flor- Turkeysin centralTexas gobble synchronously ida. with female receptivity, while southern Texas malespeak in gobbling well before femalesare LITERATURE CITED usuallywilling to mate.The remarkableability ABACUSCONCEPTS, Inc. 1989. SuperANOVA. Berke- of hen turkeysto storesperm for severalmonths ley, California. (Verma and Cherms 1964) may allow females BAUDINETTE,R. V., J.P. LOVERIDGE,K. J. WILSON, C. to mate well before they would normally ovu- D. MILLS, AND K. SCHMIDT-NEILSEN.1976. Heat late and thereby circumvent the reduction in loss from the feet of Herring Gulls at rest and fertility experiencedby malesat warmer tem- during flight. Am. J. Physiol. 230:920-924. peratures(Kosin and Mitchell 1955). Of course BOYCE,M.S. 1990. The red queen visits SageGrouse sexualdifferences in reproductivetiming might leks. Am. Zool. 30:263-270. be explainedby other ecologicaldifferences be- BRODSKY,L.M. 1988. Ornament size influences mat- tween the sexes. Nevertheless there are several ing successin male Rock Ptarmigan. Anim. Be- hav. 36:662-667. lines of evidence to suggestthat the daily and BUCHHOLZ,R. 1994. Adaptive functions of fleshy seasonaltemporal patterning of male display ornamentationin Wild Turkeysand relatedbirds. behavior may be shapedto avoid overheating. Ph.D. dissertation, Univ. Florida, Gainesville. Males are faced with a thermoregulatory BUCHHOLz,R. 1995. Female choice, parasiteloadand quandaryunder hot, sunny conditions.Resting male ornamentation in the Wild Turkey. Anim. quietly in the shade maintains sublethal body Behav. 50:929-943. temperatures,but doesnot allow feeding, fight- CALI•ER,W. A., ANIVJ. R. KING. 1974. Thermal and ing for accessto mates,or displayingto females. caloricrelations of birds. Pages259-413 in Avian These latter activitiesare also functionally and biology,vol. 4 (D. S. Farner,J. R. King, and K. C. adaptivelynecessary, but resultin metabolicheat Parkes, Eds.) Academic Press, New York. productionand exposureto solarradiation. Field CROWE,T.M. 1979. Adaptive morphologicalvaria- tion in Helmeted Guineafowl Nurnida rneleagris studiesof the behavior of Wild Turkeys relative and Crested Guineafowl Guttera pucherani.Ibis to environmental conditions, including radia- 121:313-320. tive heat load and wind speed, are needed to CROWE,T. M., AND A. A. CROWE. 1979. Anatomy of understand how males trade-off thermal needs the vascularsystem of the head and neck of the with feeding and mating success.The resultsof Helmeted Guineafowl Nurnidarneleagris. J. Zool. my study suggestthat the bare headsand necks (Lond.) 188:221-223. of male Wild Turkeys enable them to manage CROWE,T. M., ANt) P. C. WITHERS. 1979. Brain tem- these conflictinggoals more successfully. peratureregulation in Helmeted Guineafowl.S. Afr. J. Sci. 75:362-365. DAVIS, W. D. 1994. Breeding chronology in Rio ACKNOWLEDGMENTS Grande turkey hens. Final report. Job No. 7.07, FederalAid ProjectNo. W-126-R-2.Texas Parks I gratefully acknowledge the guidance of my doc- and Wildlife Department, Austin. toral committee: H. J. Brockmann, E. Greiner, R. Kil- DICKSON,J.A. 1992. The Wild Turkey: Biology and tie, D. Levey and B. McNab. J. Anderson,F. Bonna- management. Stackpole Books, Harrisburg, curso,P. Fuller, B. McNab, and R. Samudio provided Pennsylvania. invaluable advice,assistance, and equipment.C. Res- GRAY, B. T., AND H. H. PRINCE. 1988. Basal metab- trepo provided statisticaladvice and moral support. olism and energeticcost of metabolismin Wild Thanks go to F. Nordlie, F. Tomson, the Department Turkeys. J. Wildl. Manage. 52:133-137. of Zoology, the Florida Museum of Natural History, HAGAN,A. A., ANDJ. E. HEATH. 1980. Regulation of and C. Wilcox and her crew for providing housing heat lossin the duck by vasomotionin the bill. facilities or animal care. The National Wild Turkey J. Therm. Biol. 5:95-101. Federationprovided Wild Turkey transportboxes. My HATCH,D.E. 1970. Energyconserving and heat dis- researchwas funded by the Department of Zoology sipationmechanisms of the Turkey Vulture. Auk at the University of Florida, a Frank M. Chapman 87:111-124. Memorial Fund Grant in Ornithology from the Amer- HEALY,W.M. 1992a. Population influences: envi- ican Museum of Natural History, a Grant-in-Aid of ronment.Pages 129-143 in The Wild Turkey: Bi- Researchfrom Sigma Xi, the Animal BehaviorFund ology and management (J. G. Dickson, Ed.). of the University of Florida Foundation, and a StackpoleBooks, Harrisburg, Pennsylvania. 318 RICHARDBUCHHOLZ [Auk, Vol. 113

HEALY,W. M. 1992b. Behavior.Pages 46-65 in The Influence of seasonaltemperatures on Wild Tur- Wild Turkey: Biology and management (J. G. key metabolism.J. Wildl. Manage. 54:663-667. Dickson, Ed.). Stackpole Books, Harrisburg, SOK•L,R. R., aND F. J. ROHLF. 1981. Biometry, 2nd Pennsylvania. ed. W. H. Freeman and Co., New York. HILLGARTH,N. 1990. Parasites and female choice in SPURIOER,M. F., M. BOYCE,AND B. F. I¾IANLY. 1991. the Ring-neckedPheasant. Am. Zool. 30:227-233. Effectsof parasiteson matechoice by captiveSage HOFFMAN,R.W. 1990. Chronology of gobbling and Grouse(Centrocercus urophasianus). Pages 389-398 nestingactivities of Merriam'sWild Turkeys.Proc. in Ecology,behaviour and evolution of bird-par- National Wild Turkey Symp.6: 25-31. asite interactions(J. Loye and M. Zuk, Eds.)Ox- JOHNSGARD,P. 1983. The grouseof the world. Univ. ford Univ. Press, Oxford. Nebraska Press, Lincoln. gERM.A, O.P., AND F. C. CHERMS.1964. The appear- KAHL,M.P., JR. 1963. Thermoregulationin the Wood ance of sperm and their persistencyin storage Stork with specialreference to the role of the tubules of turkey hens after a single insemina- legs.Physiol. Zool. 36:141-151. tion. Poult. Sci. 44:609-613. KOSIN, I. L., aND MITCHELL, M. S. 1955. Ambient WELTY,J. C. 1975. The life of birds. W. B. Saunders temperatureas a factor in turkey reproduction. Co., Philadelphia. Poult. Sci. 34:499-505. WILSON,W. O., W. H. EDWARDS,T. L. PLAISTER,J. HIL- LAROCHELLE,J., J. DELSON,AND K. SCHMIDT-NIELSEN. LERMAN,aND A. WOODARD. 1955. The shade re- 1982. Temperatureregulation in the BlackVul- quirementof growing turkeys.Poult. Sci.34:505- ture. Can. J. Zool. 60:491-494. 508. LIGON, J., R. THORNHILL, M. Zone, aND K. JOHNSON. WILSON, W. O., AND A. WOODARD. 1955. Some factors 1990. Male-male competition, ornamentation, affectingbody temperaturesof turkeys.Poult. Sci. and the role of testosterone in sexual selection 34:369-371. in Red Jungle Fowl. Anim. Behav.40:367-373. ZuK, M. 1991. Sexualornaments as animal signals. MARTIN,P., AND BATESON,P. 1986. Measuring be- Trends Ecol. & Evol. 6:228-231. haviour. CambridgeUniv. Press,Cambridge. ZOK, M., K. JOHNSON,R. THORNHILL, AND J. LIGON. McNAB,B.K. 1988a. Complicationsinherent in scal- 1990a. Mechanisms of female choice in Red Jun- ing the basalrate of metabolismin mammals.Q. gle Fowl. Evolution44:477-485. Rev. Biol. 63:25-54. ZUK, M., J. D. LIGON,AND R. THORNHiLL. 1992. Effects McNAB, B. K. 1988b. Energy conservationin a tree- of experimentalmanipulation of male secondary kangaroo(Dendrolagus matschiei) and the red pan- sex characterson female mate preference in Red da (Ailurusfulgens). Physiol. Zool. 61:280-292. Jungle Fowl. Anim. Behav.44:999-1006. MORRISON,P. R., aND W. J. TIETZ. 1957. Cooling and ZUK, M., R. THORNHILL,R., J. LIGON, AND K. JOHNSON. thermal conductivity in three small Alaskan 1990b. Parasitesand mate choicein Red Jungle mammals. J. Mammal. 38:78-86. Fowl Am. Zool. 30:235-244. OBERLAG,D. F., P. J. PEKINS,AND W. W. MAUTZ. 1990.