Temporal Patterns of Territorial Behavior and

Temporal Patterns of Territorial Behavior and Circulating Testosterone in the Lapland Longspur and Other Arctic Passerines Author(s): Kathleen Hunt, John C. Wingfield, Lee B. Astheimer, William A. Buttemer and Thomas P. Hahn Source: American Zoologist, Vol. 35, No. 3 (Jun., 1995), pp. 274-284 Published by: Oxford University Press Stable URL: http://www.jstor.org/stable/3884063 Accessed: 24-02-2016 19:33 UTC REFERENCES Linked references are available on JSTOR for this article: http://www.jstor.org/stable/3884063?seq=1&cid=pdf-reference#references_tab_contents You may need to log in to JSTOR to access the linked references.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to American Zoologist.

http://www.jstor.org

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

Amer. Zool., 35:274-284 (1995) Patterns of Territorial Temporal in the Lapland Testosterone Longspur

Behavior and Other

and Circulating Arctic Passerines1

Kathleen

Lee B. Astheimer2, John C. Wingfield, Hunt, William and Thomas P. Hahn3 A. Buttemer,2

Department of Zoology, NJ-15, University of Washington, Seattle, Washington 98195 in relation to circulating Synopsis. We studied territorial aggression tes? tosterone levels in free-living birds of four species in northern Alaska. The Lapland Calcarius is an abundant lapponicus, breeding longspur, at lower latitudes, passerine on the arctic tundra. Unlike many passerines male Lapland longspurs do not defend a "multiple-purpose territory" that serves to provide nest sites, food and shelter. Rather, after arrival on the breeding grounds, they perform aerial display flights over a loosely defined "nest area" for a very brief period of two days or so, showing tolerance of other males. This song display may be involved in courtship. During this phase, male longspurs show a brief and pronounced peak in circulating testosterone toward simulated territorial levels, and are not aggressive intrusions Males then their mates for (STIs). "guard" sexually receptive about ten days, during which they are highly aggressive toward STIs, but do not sing as much. During the next phase, incubation, the males become males. Their circulating testosterone levels very tolerant of conspecific decline to baseline levels, and they generally do not sing or display aggres? sion in response to STIs. Three other passerines, the white-crowned spar? American tree sparrow, Spizella row, Zonotrichia leucophrys gambelii, show patterns arborea, and savannah sparrow, Passerculus sandwichensis, of territorial aggression typical of species studied at lower latitudes. Welldefined territories are defended for several weeks, during which there is a prolonged of testosterone. These three peak in plasma concentrations to sing and display aggression even late in the season, species continue unlike the longspurs. The peak of testosterone in the longspurs occurs with the peak in song display, while in mid-latitude simultaneously species it occurs with the peak in reproductive These data suggest that aggression. the interrelationship of testosterone in Lapland longspurs and aggression with multiple-purpose may be different from that of passerines territories, and may be related to the constraints of breeding in the open arctic tundra. Introduction Male certain

passerine breeding

show birds commonly behaviors that vary in

1From the Symposium Endocrinology of Arctic Birds and Mammals presented at the Annual Meeting ofthe American Society of Zoologists, 27-30 December 1993, at Los Angeles, California. 2 Present address of Lee B. Astheimer and William A. Buttemer is: Department of Biomedical Research and Department of Biological Sciences, University of Wollongong, Wollongong, New South Wales 2500, Australia. 3 Present address of Thomas P. Hahn is: Department of Psychology, Johns Hopkins University, Baltimore, Maryland 21218.

and duration from species to spe? intensity cies. For instance, early in the breeding sea? son, males typically spend much time sing? and defending their ing, courting females, territory from rival males, and will reduce or cease these activities later in the season. Such behaviors are important for synchronizing the male's and female's physiology and behavior, from final maturation of the gonads to the shift from sexual to parental behavior. These complex behavioral inter? actions have been shown to have an intimate relationship with steroid hormones, such as testosterone (T), estradiol, and their metabolites (?.#., Balthazart, 1983; Harding

274 This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

Territoriality

in Arctic

et al, and Ramenofsky, 1988; Wingfield and Moore, 1985; Wingfield 1987). Fur? the temporal of circu? thermore, patterns hormone levels in lating (e.g., testosterone) different to be linked to species appear breeding strategies and mating systems, sug? of behav? gesting a broad interrelationship ioral ecology and endocrine mechanisms et al, most However, (Wingfield 1990). on the interrelationships of investigations and aggression have focused on testosterone zone species in which the breed? temperate season (i.e., 2-5 ing may be protracted the have addressed None tempo? months). of repro? ral progression and integration ductive events in the Arctic, where breeding seasons are usually only 4-5 weeks. such as the Species from mid-latitudes, Sound white-crowned Puget sparrow (Zonotrichia leucophrys pugetensis) and song have multisparrow (Melospiza melodia), territories that males defend ple-purpose the breeding season. Territorial throughout defense includes and singing, patrolling, with challenging males. agonistic encounters the territorial During aggressive encounters, male approaches the intruder and gives short flights (fluttersongs, threat postures, ing wings and short flights from perch to around the intruder), and direct perch within 5 m until the attacks, remaining intruder is expelled from the territory (see and Moore, and Wingfield 1987; Wingfield Hahn, 1994). Note that males also use song as an advertisement or courtship signal to attract mates. These behaviors have been shown to be activated and prolonged by tes? tosterone (e.g., Balthazart, 1983; Harding et and Moore, al, 1988; Wingfield 1987; Wingfield, 1994). Once plasma levels of tes? tosterone decline at the end ofthe breeding then high levels of spontaneous season, and of courtship-related activi? aggression, and Ball, ties, also decline (see Nowicki are typical of pas? 1989). These behaviors serines breeding at northern mid-latitudes. Most species that have been studied breed in wooded or semi-open habitat (shrubs and thickets), and many maintain multiple-purfor several months. pose territories Arctic birds, in contrast, have a very short of hormonebreeding season. Comparison behavior in this temporally adaptations

275

Passerines

restricted nesting cycle with the more proin lowertracted breeding cycle common a number of ques? latitude raises species tions. Do breeding behaviors (such as ter? ritorial and mate-guarding song, aggression, or and courtship) have different intensities breed? duration in birds that must complete Do cir? ing in the fleeting arctic summer? levels show a pattern culating testosterone in any way (e.g., in the that is different correlates amplitude, timing, or behavioral T peak) from that seen in of the seasonal lower-latitude species? If so, what aspects in of the Arctic might be most important that result pressures exerting the selective in different Here we breeding strategies? the testosterone-behavior intercompare of a the arctic relationships passerine, truly Lapland longspur, with other passerines that breed both in the Arctic and at lower lati? tudes. As we are comparing several different we have arranged this aspects of behavior, paper by topics. We first discuss the natural history of our four species, and then describe methods and results for temporal patterns of testosterone, and then for aggressive behavior. Finally, we compare the four spe? cies and discuss the ecological of context these hormone-behavior relationships. Natural General

History

methods

We studied a population of Lapland long? spurs (Calcarius lapponicus, "longspurs"), Gambel's white-crowned sparrows (Zono? trichia leucophrys gambelii), American tree and savannah sparrows (Spizella arborea) at sparrows (Passerculus sandwichensis) Toolik Lake (68?38'N, 149?38"W), Alaska, of 1989-1994. Birds during the summers in Potter traps baited with were captured seeds or in Japanese mist nets, and each individual (except for savannah sparrows) was banded with a unique combination of color bands for subsequent observation and in the field. In each year, we identification identified mated pairs and observed behav? ioral patterns and general reproductive biol? ogy. Natural

history ofthe Lapland Longspur The Lapland longspur is a small passerine that breeds throughout much of the Arctic

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

276

K. HuNT?:r^L.

tundra region. It has a circumpolar distri? bution and is thought to be the most abun? dant arctic breeding bird (Bent, 1968). Males arrive on the breeding grounds in mid-May 1960) and within a few days (e.g., Irving, they show display flights accompanied by song over an apparent "territory" (or at least they show these aerial displays over the same m in small patch of ground, about 50-100 L. B. Astheimer and K. Hunt, diameter; A displaying unpublished observations). male typically flies up to a height of about 50 m, holds his wings out, and glides downward in a large spiral while singing several times. Before starting the glide, he often will fly in a large circle that roughly outlines the perimeter of his "territory", uttering a series of short call notes. During this phase they do sometimes (but not always) chase other males out of their territories. They also actively court females, and thus these aerial function. displays may serve a courtship occurs Pairing usually rapidly, following which display flights are greatly reduced. As the females begin nest-building and become males follow their mates sexually receptive, to prevent extraclosely as they mate-guard often pair copulations, chasing other males phase lasts 6-8 away. This mate-guarding days. Females then incubate the eggs for 1213 days, during which the male sings only or follows rarely, and no longer mate-guards the female. Until mid-incubation (typically the female is capable of renestmid-June), and ing if the clutch is lost (L. B. Astheimer K. Hunt, unpublished observations). Our field observations indicate that after the eggs hatch, both parents feed the young, which fledge at about 10 days of age. Adult males and females continue to feed fledgat about 30 days. lings until independence, During this parental phase, song frequency and territory defense by males declines to near zero, though occasional song bouts are still heard. Males begin to forage in small and will tolerate other groups ("cruising"), males close to the nest (sometimes even tolerating another male singing a few meters from the nest!). In this late stage, both sexes forage over a wide area, and are frequently found up to 1 km away from their own nests, often in other longspurs' territories (and near their nests) with little aggression from the

will also congregate at They rich sites for food, such as at our unusually seed-baited trap sites. male do not Lapland longspurs Clearly, defend a classical terrimultiple-purpose appear to show a brief tory, and instead series of territorial behaviors that begin with the aerial displays and songs, and finish with behavior. Our general obser? mate-guarding at Toolik vations ofthe Lapland longspurs for this Lake concur with data published species breeding at Barrow, Alaska (Custer and Pitelka, 1977); Bylot Island, Northwest Canada (Drury, 1961) and EqaTerritories, West Greenland Nunaat, (Fox et lungmiut al, 1987). It should be pointed out that the aerial displays and lack of a clearly defined multiple-purpose territory may be typical of on grassland/ other passerines breeding habitats at lower latitudes? shrub-steppe the bobolink, for example, oryDolichonyx zivorus and the dickcissel, (Bent, 1965), Spiza americana (Zimmerman, 1966). residents.

Natural history of White-crowned, Tree, and Savannah Sparrows Gambel's white-crowned sparrow, Amer? ican tree sparrow and savannah sparrow are abundant the northern species throughout North American continent. Gambel's whitehave crowned and savannah sparrows of the far south breeding ranges extending distri? Arctic circle (Bent, 1968). Breeding tree sparrow is conbution ofthe American fined to arctic and subarctic regions, but it does not breed on open fell-field tundra or the high Arctic (Bent, 1968) as does the Lap? the land longspur. One of these species, white-crowned sparrow, has been studied in control detail in relation to the endocrine of the activation mechanisms underlying territorial testosterone by (e.g., aggression Wingfield and Moore, 1987). White-crowned and American tree sparrows sparrows territories at Too? defend multiple-purpose lik Lake, usually around patches of willow (Salix sp.) that may reach a height of up to 2 m in sheltered river beds and gullies. also defend multipleSavannah sparrows on tussock tundra in territories purpose lower flat areas often associated with patches of willow.

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

Territoriality

in Arctic

These three species of passerines are sim? in the broad ilar to the Lapland longspurs outline of their breeding season, with a sim? in a single brood, ilar clutch size (4-6) similar renesting until mid-June, possible and similar of parental duties. division Males sing at high frequency and also mateguard early in the season. These behaviors then decline the parental phase. during However, they differ from the Lapland longin the following spur primarily ways: First, the three sparrow species have lower breed? than the longspurs, due to the ing densities sparrows' preference for willow habitat. Our study site at Toolik Lake usually has 40-50 to an pairs of Lapland longspurs compared average of 10-15 pairs for each ofthe other three species. Second, all three sparrow spe? cies tend not to be as tolerant of conspecifics as the Lapland longspurs; the sparrows have more precise territory boundaries, defend their territories longer, are not as apt to form flocks during the breeding season, and will not tolerate close proximity of a conspecific even when foraging. Third, the sparrows do not have aerial song displays, singing instead from slightly elevated perches within their territories.

Temporal Profile of Testosterone in Relation to Breeding and Phenology Territorial Aggression Changes

in circulating

testosterone

levels

We took blood samples from trapped birds in 1989-1994 by puncture ofthe wing vein, microcollecting the blood into heparinized hematocrit tubes and keeping them on ice until within 4 hr). centrifuged (usually Plasma was stored at - 20?C and later transof Wash? ported frozen to the University ington in Seattle. We then measured plasma levels of testosterone by radioimmunoassay after partial purification of steroid extracts on diatomaceous micro colearth/glycol umns. This assay procedure, along with inter- and intra-assay has been variations, in detail by Ball and Wingfield described (1986). There

was considerable in dimorphism between male plasma levels of testosterone and female Lapland longspurs (Fig. 1). This

Passerines

5/7

5/175/27

277

6/6

6/166/26

7/6

Fig. 1. Plasma levels of testosterone in male and female Lapland longspurs (Calcarius lapponicus) in relation to calendar date at Toolik Lake, 1990. is typical of most passerine species studied to date (Wingfield and Farner, 1993). Males showed highest levels for a few days (or even less) early in each year (e.g., around 27 May in 1990, Fig. 1). Females never attained such high levels and showed very little temporal of circulating levels of testos? organization terone with calendar date (Fig. 1). Owing to the rapid temporal progression ofthe breed? ing cycle, Lapland longspurs appear to be in laying dates. We highly synchronous organized the hormone data from 1992 both by calendar date (Fig. 2, upper panel) and by stage in the breeding cycle (display, mateand feeding of young) guarding, incubation for individuals of known status breeding (Fig. 2, lower panel). The temporal patterns of testosterone in male Lapland longspurs identical when ana? appeared essentially lyzed both ways. Plasma levels of testoster-

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

K. Hunt

278 Male

Lapland

longspurs,

et al.

1992

25 20 C o 15 c o 5 co 10 o (0 0) 5

Late

Early H *T 1 5/26 5/31 6/5 6/10 6/15 6/20 6/25 6/30 Calendar date

0

Fig. 3. Plasma levels of testosterone in male Lapland longspurs (Calcarius lapponicus), Gambel's whitecrowned sparrows (Zonotrichia leucophrys gambelii) and American tree sparrows, (Spizella arborea), early (pre-breeding) and late (parental phase) in the breeding season. Bars are means and vertical lines the standard errors.

25 ??? 20 c 15 105_J_^J w?. FEMALE: BuikjLay IncubateFeed/brood MALE:DisplayGuard Sentry/ Feed/cruise Reproductive

stage

Fig. 2. Comparison of plasma levels of testosterone in relation to calendar date (upper panel) and stage in the reproductive cycle (lower panel) of male Lapland longspurs, (Calcarius lapponicus) at Toolik Lake, 1992. Sample size = 44. one were highest during the three days when and were showing aerial displays males (the "display singing at greatest frequency phase"; mean of 6.95 ng/ml ? 1.5 SE), with several individuals showing very high T lev? els of over 10 ng/ml. These levels were higher than in both the pre-settlement phase before the display phase (mean immediately of 1.90 ng/ml ?0.43), and the mate-guardafter (mean of 1.64 ing phase immediately was ng/ml ? 0.66). By the time incubation were underway, plasma levels of testosterone basal (mean of 1.09 ng/ml ? 0.33). Analysis with single-factor ANOVA revealed signif? icant differences among the four phases (P = Further analysis with Fisher's Pro0.0004). Difference test tected Least Significant

the display phase to have signifi? showed cantly higher T levels than the pre-settleor incubation phases, ment, mate-guarding, while the latter three phases did not differ from each other. Male white-crowned sparrows and Amer? ican tree sparrows also showed declines in levels of testosterone as the season pro= 2.392, P = 0.0404, gressed (1989, Fig. 3, t df = 9; and t = 3.212, P = 0.0075, df = 9 Student's For Mest, 2-tailed). respectively, ease of comparison, we include the 1992 = longspur data on the same figure ("early" = The data for display, "late" incubation). the sparrow species are essentially identical to those presented and Farner by Wingfield for a population of Z.l gambelii (1978) in the of Fairbanks, Alaska breeding vicinity levels as (64?N). The decline of testosterone the parental phase of the breeding cycle ensues (late season) is typical of avian spe? cies in which males provide significant et al, 1990). parental care (e.g., Wingfield testosterone levels in the spar? However, rows were elevated for 2-3 weeks early in the breeding season males were (when territories and establishing courting females, and Farner, 1978; Wingfield and Wingfield Moore, 1987)?much longer than the 2-3 levels of testos? days of elevated circulating terone in male longspurs. Quantification

of aggressive

behavior

Males of all species were tested with sim? ulated territorial intrusions (STI) to assess

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

Territoriality

in Arctic

This procedure detail elsewhere adult a free-living (Wingfield, 1985). Briefly, male is challenged by playing back taperecorded conspecific songs, and the aggres? ofthe male is recorded as he sive behavior tries to locate and drive off the apparent intruder. In many studies, a caged, live con? specific decoy is placed alongside the speaker to provide visual as well as auditory stimuli. males fly up to a perch (e.g., in a Typically, willow), sing in reply, and then approach the source with threats and wing quivers all these behaviors to (flights), showing the breeding varying degrees throughout season (e.g., Wingfield, 1985; Wingfield and and Hahn, 1994). Moore, 1987; Wingfield this pattern of response may not However, the Lap? hold for all species. For example, land longspur prefers open tundra where may only be 10 cm high. These vegetation birds thus rarely show flights but rather fly down to the decoy and walk around in threat postures 1961; K. Hunt, unpub? (Drury, lished observations). Male longspurs often do not sing during STI, and almost always come very close to the decoy (< 1 m), even when not aggressive. of these Because in behavioral we will describe differences, some detail the STI procedure used with the

their aggressive behavior. in has been described

longspurs. Adult male Lapland longspurs were tested with a STI as follows: A caged adult male was placed 20 m from the focal longspur male's nest (or in the center of an apparent as the area over which territory, designated he performed most of his song flights). Con? at Toolik Lake in specific song, recorded over a speaker 1991, was played placed the decoy's alongside cage. The playback the natural of tape mimicked frequency the Lapland longspur STI, singing. During an observer stood 20 m from the decoy (40 m from the nest) and recorded the focal male's for a ten-minute response period. Blinds were not used, as Lapland longspurs at Toolik Field Station are accustomed to humans. Four different behaviors were recorded: 1. Number of songs throughout the ten-minute period. Because of the dual role of song as a courtship signal as well as a territorial signal, song was not automatito be indicative of aggrescally considered

279

Passerines

sion. 2. Time spent within 5 m ofthe decoy (as in Wingfield, 1985). We regarded time within 5 m of the decoy as an indicator of general interest in the decoy, but not nec? essarily of aggression. 3. Time spent in threat posture (defined as body held horizontally, and bill pointed drooped wings slightly ahead or tilted up?see straight Drury, defined as 1961); and 4. Number ofattacks, physical striking of the decoy's cage with the bill or feet. We considered these last two variables to be indicators of aggression. STIs were performed on Lapland long? spurs at three major phases in the breeding cycle: during the song display phase immediately after arrival, during the mate-guardphase. ing phase, and during the incubation STIs were performed on white-crowned to sparrows early in the season (comparable display and mate-guarding phases in Lap? land longspur) and late in the season (during incubation and feeding of young). STIs were on savannah and tree sparrows performed only late in the season, during incubation and feeding of young. All Lapland longspur individuals were of known status. For the precisely breeding other three species, individual breeding sta? tus was not always known, and was inferred from the general status. We population regard this as quite reliable, since breeding is tightly synchronized in all these species at Toolik. Results

of simulated

territorial

intrusions

Male Lapland longspurs gave the greatest number of songs during the display phase test, H = 10.045, P (Fig. 4, Kruskal-Wallis = = df 0.0066, 2), but were virtually silent and incubation. Male during mate-guarding also less time within 5 m longspurs spent ofthe decoy during the display phase com? and incubation pared with mate-guarding = 6.894, P = (Fig. 4, H 0.0318). Our impression was that during the display phase, males continued their usual aerial song displays with little concern for the STI. In contrast, during the mate-guarding phase, males spent much time by the decoy, and time in threat and the number of attacks were posture highest (Fig. 4, H = 19.004, P = 0.0001 and H = 11.645, P = 0.003 respectively). During the incubation phase, posture, aggressive

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

K. Hunt

280

Male

Lapland

etal.

Longspur

STI

10

201 O O 01 13 |15 o tn

8i

E to 10

o XI E 3

S m 01 3 C

X

Display

Mateguard

4H

2i

Incubate

10n 01 k. 3 *?* CO o a ??# m 01 k. ?C

Display

Mateguard

Incubate

Display

Mateguard

Incubate

251

201

15H

c 10H m a> ?>?* 3 C ????

51

Display

Mateguard

Incubata

Fig. 4. Behavioral responses of male Lapland longspurs (Calcarius lapponicus) to simulated territorial intrusions. Bars are means and vertical lines the standard errors. attacks, and song were all low, but males to the decoy, close remained generally "interested" but not aggressive. appearing Males often foraged near the decoy, or sat still by the decoy's cage, sometimes apparanalysis ently going to sleep. Subsequent U tests (Table 1) gen? with Mann-Whitney test erally concurred with the Kruskal-Wallis did not results, although some comparisons

quite reach significance (e.g, songs and attacks in display vs. mate-guarding). Males of species defending multiple-purand also territories pose gave songs the simulated intruder (Figs. 5 approached and 6). White-crowned sparrows showed no in number of songs in changes significant STIs performed early in the season vs. late in the season (Fig. 5, upper panel). How-

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

Territoriality

in Arctic

281

Passerines

Table 1. Mann-Whitney U-test results for the responses ofmale Lapland longspurs to STI during the display phase (D), mate-guarding (M), and incubation (I)* D vs.I

M vs.I

D vs.M

2.024 (0.0430) # Songs 1.676(0.0937) Time in 5 m 0.688(0.4913) 2.572(0.0101) 4.204 (0.0001) 2.194 (0.0282) Threat posture # Attacks_1.766(0.0775)_3.220(0.0013)_0.153(0.8785) * The table shows Z-values, corrected for ties, with P-values in parentheses. Sample display, n = 22 for mate-guarding, and n = 25 for incubation. males the simulated ever, approached intruder more closely early in the season than late (P < 0.02, Mann-Whitney U test). For STIs performed during the parental and feeding of young), phase (incubation male white-crowned tree spar? sparrows, rows, and savannah sparrows all gave more songs than male Lapland longspurs (Figs. 4, 5 and 6). Male savannah also sparrows around intruder showed the many flights

Male White-CrownedSparrow STI

3.136(0.0017) 2.200(0.0278) 0.372 (0.7099) sizes were n = 6 for

(Fig. 6) and spent almost 50 percent of the STI period within 5 m ofthe tape recorder, even though all birds tested were feeding young. 16 i4i 12 10 8 6 4 2 0

Z.I.gambelii

P. sandwichensis

Z.l.gambelii

P. sandwichensis

40 30 i 20 10 H Early

Late 0

16i 14 12 10 8 CD (0 O o

6" 42" Early

Late

Fig. 5. Behavioral responses of male Gambel's whitecrowned sparrows (Zonotrichia leucophrys gambelii) to simulated territorial intrusions, both early (prebreeding) and late (parental phase) in the breeding sea? son. Bars are means and vertical lines the standard errors. E = early and L = late in the breeding season.

sandwichensis Fig. 6. Behavioral responses of male Gambel's whitecrowned sparrows (Zonotrichia leucophrys gambelii), American tree sparrows (Spizella arborea) and savan? nah sparrows (Passerculus sandwichensis) to simulated territorial intrusions late in the breeding season. Bars are means and vertical lines the standard errors.

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

K.HUUTETAL.

282

Discussion and Comparison of testosterone in Lapland longspurs aggression other species

vs.

The Lapland longspurs do show some dif? to the whiteferences when compared savannah and American tree, crowned, in differed particularly sparrows. Longspurs 1) their type of territory defense, 2) the pheof their testosterone peak, 3) the nology and duration of their aggressive nature of to STI, and 4) the correlation responses the testosterone profile with behavior. behavioral 1. Territory defense. ?Our show that the longspur males observations defend loose "nest areas" with indistinct of the close and are tolerant boundaries, males at all times of conspecific proximity After midother than during mate-guarding. fluctuate to incubation, territory boundaries the point of vanishing, and males frequently of with small flocks off-territory forage In contrast, males "cruising." neighboring white-crowned and song sparrows breeding at mid-latitudes show a more gradual shift from high song rate to low song rate upon entering the parental phase, greater aggres? sion toward conspecifics at all times in the to period, and less of a tendency breeding flock or to "cruise" (Wingfield and Hahn, Amer? 1994). Our data for white-crowned, ican tree, and savannah sparrows at Toolik Lake clearly show a pattern similar to those tested at mid-latitudes, and dif? passerines ferent from the Lapland longspur. 2. Testosterone. ?The males longspur showed an unusually to five-fold high (up higher than in white-crowned sparrows; Figs. 1 and 2, and Wingfield and Farner, 1978; and Moore, 1987) and brief (1-3 Wingfield versus 2-3 weeks) peak in testosterone. days In fact, Lapland longspurs have the highest levels of circulating testosterone of any wild and Far? species studied to date (Wingfield ner, 1993), and the peak is so brief in dura? tion that we may have missed it in many individuals sampled during the first few days after arrival at Toolik Lake. 3. Response to STL? Lapland longspurs showed high song rates during STI only at the very first stage ofthe season, the display stage, when their spontaneous song rate is

STI highest. Longspur song rate during after this. In contrast, declined abruptly white-crowned sparrows sang in response to the season, and whiteSTI throughout all tree, and savannah crowned, sparrows sang more than longspurs during late-season STI (Figs. showed 4-6). Longspurs the toward as greatest decoy, aggression measured threat and postures attacks, by Note that this peak during mate-guarding. in STI-measured aggression lagged about a week behind the peak in spontaneous song rate. Longspurs do not tend to respond to STI with flights around the decoy. In con? trast, the white-crowned, song, and savan? nah sparrows with numerous responded variation among flights. There is noticeable the three sparrow species in all the behav? iors measured, with the white-crowns being least like the longspurs and the savannah This may sparrows most like the longspurs. relate to the savannah possibly sparrows' similar tussock-nesting habitat. 4. Testosterone and behavior. ?The peak of circulating testosterone levels in long? spurs (10-20 ng/ml) occurs when the males are singing the most (display not phase), when they are most aggressive toward STI Testosterone (mate-guarding phase). declines to an intermediate level by the time the males are most aggressive. In contrast, in the other three species peaks testosterone at about 3-5 ng/ml, and stays at that level for 2-3 weeks while the male is most aggres? sive and is also mate-guarding (e.g., Wing? field and Moore, 1987; Wingfield and Hahn, 1994). Our impression is that in Lapland long? division spurs there is a slight temporal between maximum song and maximum aggression, with song peaking a week earlier. Maximum with the song rate correlates brief testosterone longspurs' unusually high, correlates with peak; maximum aggression the more usual 2-4 ng/ml level of testos? terone seen a week or two later. Testoster? one is known to affect both singing and aggression in many species, so much so that a type of aggression, song is often considered and testosterone is sometimes said to pro? mote both behaviors How? simultaneously. is not only an ever, song has a dual role?it aggressive behavior toward rival males, but

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

Territoriality

in Arctic

also a courtship behavior that serves to attract potential mates. Male Lapland long? spurs, in fact, may be using song more for than for aggression, as indicated courtship to sing in aerial display, by their tendency and not to sing during aggressive responses to STIs. It is tempting to see a possible duallevel effect of testosterone in Lapland long? spurs, with very high levels of testosterone promoting song and aerial displays, and thus perhaps courtship of females, and lower lev? els promoting inter-male At aggression. these are present however, just correlations, and controlled are needed to experiments examine these questions more closely. The Arctic

as a "unique"

ecosystem

Overall, it is clear that Lapland longspurs are different in several of their aspects behavior and endocrinology when com? American pared with white-crowned, tree, and savannah sparrows. Why might this be different? are arctic species Longspurs limited to breeding on the tun? specialists the other three species have dra, while extensive in non-arctic breeding ranges as well. For these, the low-arctic regions tundra is the extreme northern extent of their the differences discussed ranges. Might above be due to special selective pressures seen in the Arctic? The Arctic has several features that might affect breeding biology, including: 1. Very short summer season (breeding limited to mid-May to mid-July). 2. Low vegetation; the tallest trees are willow m in 0.5-2 riparian thickets, height. 3. Essentially continuous daylight in June. 4. Large insect bloom, also in June. 5. Erratic snowstorms probable in the early breeding season, and possible anytime. Most of these factors, individually, are not unique to the Arctic. For instance, birds also experience freliving at high elevations and birds of tem? quent spring snowstorms, perate grasslands also live in low vegetation. If the Arctic is indeed unique, it is in the combination of these factors and continu? ous daylight throughout the breeding period. In particular, the low, open tundra habitat combined with a very short breeding season

283

Passerines

may lead to certain ological adaptations,

behavioral as follows:

and physi?

1. There can be one clutch only per season, if with a limited window for renesting nest is lost. Therefore, mate-guarding may be intense but brief. Individuals may switch rapidly from courtship to mateat the beginning of guarding behaviors the season. 2. Abundant treeless tundra habitat means that nest sites may not be limited and territory quality will possibly not be very variable. Thus, strong defense of a ter? ritory may not be necessary. 3. The very short time for arrival, setting up territories, and finding a mate, in conwith the abundant tundra hab? junction itat, may lead to priority being given to and mate-guarding instead of courtship defense. Territories territory may have loose boundaries and may be defended only briefly. 4. The treeless "grassland-like" habitat may or other promote flight song displays aerial displays by males. 5. Ground-nesting may result in high nest with accompanying selection predation, for cryptic pressure nests, "wariness" around the nest, and sentry behavior. tundra nesting habitat Thus, the abundant combined with the very short window for mates have led to finding may longspurs aerial song dis? develop their grassland-like with their loose "nest area" play combined territories. Their brief, high testosterone peak may serve as a trigger to switch behav? iors rapidly from migration to song and then to mate-guarding, thus initiating breeding very rapidly. The Lapland longspur is just one arctic On the basis of this one example specialist. we cannot answer the broader question addressed of papers; by this collection whether the mechanisms and/or namely, and endocri? temporal patterns of behavior nology are different in arctic animals from those in temperate-zone animals. We sug? between Lap? gest that specific comparisons land longspurs, other arctic-specialized breeders Cal(such as Smith's longspurs, carius pictus, redpolls, Acanthis flammea, and snow buntings, Plectrophonax nivalis)

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions

K. HUNT?TAL.

284

and lower-latitude grassland breeders (such C. mccownii, as McCown's longspurs, C. ornatus, chestnut-collared longspurs, and bobolinks, oryzivorus, Dolychonix the hormoneothers) will reveal whether of this arctic specialist behavior interactions to the Arctic, unique adaptations represent or are more general features of birds inhab? iting grassland and open tundra biomes. Acknowledgments reported herein were investigations supported by grant numbers DPP-8901228, to DPP-9300771 and DPP-9023834, J.C.W., and by an NSF Graduate Research Two to K.H. Fellowship anonymous comments on valuable reviewers provided We are also grateful to Lynn the manuscript. with the Erckmann for her expert assistance hormone assays. The

References Ball, G. F. and Wingfield, J. C. 1986. Changes in plasma levels of sex steroids in relation to multiple broodedness and nest site density in male starlings. Physiological Zoology 60:191-199. Balthazart, J. 1983. Hormonal correlates of behavior. In D. S. Farner, J. R. King and K. C. Parkes (eds.), Avian biology, Vol. 7, pp. 221-365. Academic Press, New York. Bent, A. C. 1965. Life histories of North American Blackbirds, Orioles, Tanagers, and Allies. Dover Publications, New York. Bent, A. C. 1968. Life histories of North American Cardinals, Grosbeaks, Buntings, Towhees, Finches, Sparrows and Allies, Vol. 1-3. Dover Publications, New York. Custer, T. W. and F. Pitelka. 1977. Demographic features of a Lapland Longspur population near Barrow Alaska. The Auk 94:505-525. Drury, W. H. Jr. 1961. Studies ofthe breeding biology of horned lark, water pipit, Lapland longspur, and snow bunting on Bylot Island, Northwest Terri? tories, Canada. Bird-Banding 32:1-46. Fox, A. D., I. S. Francis, J. Madsen, and J. M. Stroud. 1987. The breeding biology ofthe Lapland bun-

ting Calcarius lapponicus in West Greenland dur? ing two contrasting years. Ibis 129:541-552. Harding, C. F., M. J. Walters, D. Collado, and K. Sheridan. 1988. Hormone specificity and activation of social behavior in male red-winged blackbirds. Hormones and Behavior 22:402-418. Irving, L. 1960. Birds of Anaktuvuk Pass, Kobuk, and Old Crow. A study in Arctic adaptation. United States National Museum Bulletin 217, Smithson? ian Institution, Washington D.C. Nowicki, S. and G. F. Ball. 1989. Testosterone induc? tion of song in photosensitive and photorefractory male sparrows. Hormones and Behavior 23:514525. Wingfield, J. C. 1985. Short-term changes in plasma levels of hormones during establishment and defense of a breeding territory in male song spar? rows, Melospiza melodia. Hormones and Behav? ior 19:174-187. Wingfield, J. C. 1994. Regulation of territorial behav? ior in the sedentary song sparrow, Melospiza melo? dia morphna. Hormones and Behavior 28:1-15. Wingfield, J. C. and D. S. Farner. 1978. The annual cycle in plasma irLH and steroid hormones in feral populations ofthe white-crowned sparrow. Zono? trichia leucophrys gambelii. Biology of Reproduc? tion 19:1046-1056. Wingfield, J. C. and D. S. Farner. 1993. Endocri? nology of reproduction in wild species. In D. S. Farner, J. R. King, and K. C. Parkes (eds.), Avian biology, Vol. 9, pp. 163-327. Academic Press, New York. Wingfield, J. C. and T. P. Hahn. 1994. Testosterone and territorial behaviour in sedentary and migra? tory sparrows. Animal Behaviour 47:77-89. Wingfield, J. C, R. E. Hegner, A. M. Dufty, Jr., and G. F. Ball. 1990. The "challenge hypothesis": Theoretical implications for patterns of testoster? one secretion, mating systems, and breeding strat? egies. American Naturalist 136:829-846. Wingfield, J. C. and M. C. Moore. 1987. Hormonal, social, and environmental factors in the repro? ductive biology of free-living male birds. In D. Crews (ed.), Psychobiology of reproductive behav? ior: An evolutionary perspective, pp. 149-175. Prentice Hall, New Jersey. Wingfield, J. C. and M. Ramenofsky. 1985. Hor? monal and environmental control of aggression in birds. In R. Gilles and J. Balthazart (eds.), Neu? robiology, pp. 92-104. Springer-Verlag, Berlin. Zimmerman, J. L. 1966. Polygyny in the dickcissel. The Auk 83:534-546.

This content downloaded from 158.121.247.60 on Wed, 24 Feb 2016 19:33:05 UTC All use subject to JSTOR Terms and Conditions