- Email: [email protected]
Female song sparrow, Melospiza melodia, response to simulated conspecific and heterospecific intrusion across three seasons
ANIMAL BEHAVIOUR, 2000, 59, 551–557 doi:10.1006/anbe.1999.1369, available online at http://www.idealibrary.com on
Female song sparrow, Melospiza melodia, response to simulated conspecific and heterospecific intrusion across three seasons MICHELLE M. ELEKONICH
Department of Psychology, University of Washington (Received 13 September 1999; initial acceptance 23 September 1999; final acceptance 29 November 1999; MS. number: A8319R)
To investigate female responses to territorial intrusion I presented female song sparrows with either a simulated female song sparrow intrusion or a simulated spotted towhee, Pipilo maculatus, intrusion as a control during either the prebreeding, breeding or postmoult seasons. Aggressive and nonaggressive behaviours and vocalizations were compared between intrusion types and across seasons. Principle components analysis suggested that female responses fell into three categories: (1) responses directed towards the intruder, mostly aggressive; (2) responses directed towards the mate; and (3) lack of response to the intruder. In every season, females responded more aggressively to simulated female song sparrow intrusion than simulated towhee intrusion. Responses directed towards female song sparrow intruders dropped across the three seasons and were significantly higher in the prebreeding season than in the breeding or postmoult seasons.
Males are viewed as the more aggressive sex in many animal species due to male specializations for intrasexual competition (reviewed in Archer 1988) and most studies of aggression in birds have focused on males of north temperate species. Most studies of female aggression focused on cases where females were expected to be under male-like selection pressures and evolve male-like specializations for intrasexual competition (e.g. sex-role reversed species: Jenni & Collier 1972; polygynous species: Yasukawa & Searcy 1982; Wittenberger & Tilson 1980; species where females regularly sing: Richison 1983, 1986; Beletsky 1982). Although various investigators have compared male intrasexual territorial aggression to female nest defence, comparisons like these are inappropriate because aggression serves different functions in each context and has been shaped by different selection pressures. Only in a few avian species has aggression in males and females been tested in comparable ways to allow direct comparison (Holmes et al. 1989; Wicklund & Village 1992; Slagsvold 1993; Levin 1996a, b). Due to differences in parental investment, natural selection often produces distinct reproductive and life history strategies in the sexes (Trivers 1972). Hrdy (1981; Hrdy & Williams 1983) argues that sex differences when they occur are most likely to be present in traits used to compete with conspecifics. Sex
differences should occur not only with respect to the development of particular behaviours, but also with respect to the function, hormonal control and social context of those behaviours. In song sparrows, the function, hormonal control, seasonality and context of male–male aggression have been particularly well studied (e.g. Wingfield et al. 1987; Wingfield & Hahn 1994), and females are known to engage in intraspecific aggression (Nice 1943; Arcese 1989b). In the present study I examined female response to territorial intrusion by female conspecifics to allow a direct comparison between male and female responses to simulated intrusions throughout the year. Although female defences against intraspecific parasitism have received considerable attention (e.g. Gowaty & Wagner 1988; Sandell & Smith 1997), there is no observational or genetic evidence for egg dumping in song sparrows (Nice 1943; Keller 1996). Thus, it is unlikely that female–female aggression in song sparrows functions to protect against intraspecific brood parasitism. Reported functions of female aggression in birds include: defence of a nest or young (reviewed in Montgomerie & Weatherhead 1988), reduction of intraspecific or interspecific brood parasitism (e.g. Gowaty & Wagner 1988), maintenance of pair or family group bonds (e.g. Ritchison 1983) and limiting female competitors’ access to mated territorial males (e.g. Hurley & Robertson 1984), parental care or defence provided by males (e.g. Yasukawa & Searcy 1982), or access to feeding and/or breeding territories (e.g. Hurley & Robertson 1984; Levin 1996a, b).
Correspondence and present address: M. M. Elekonich, Department of Entomology, MC-118, University of Illinois, 505 South Goodwin, Urbana, IL 61801, U.S.A. (email: [email protected]). 0003–3472/00/030551+07 $35.00/0
2000 The Association for the Study of Animal Behaviour
551
2000 The Association for the Study of Animal Behaviour
552
ANIMAL BEHAVIOUR, 59, 3
If female song sparrow aggression functions primarily for defence of the territory and its resources, as it does for males (Wingfield 1985; Stoddard et al. 1991; Wingfield & Hahn 1994), the seasonal characteristics of female territorial aggression should parallel those of male territorial aggression. In Washington State where these experiments occurred song sparrows are a resident species and many pairs remain on their territories throughout the year. Thus, if this hypothesis is true, female aggression should occur throughout the year and peak when territories are being settled (i.e. prior to nesting in the spring and again in the autumn as is the case for males; Beecher et al. 1994). If the primary function of female territoriality is to defend the food and nest sites available on the territory then female aggression should be directed equally towards conspecifics and heterospecifics that use the same resources.
METHODS
Sites and Subjects The experiments were conducted at the Skagit Valley Wildlife Recreation Area, near Conway, Washington, about 80.5 km north of Seattle in 1992, 1993 and 1994, and at Discovery Park, a 3.2-km2 undeveloped park along Puget Sound in Seattle, Washington, in 1990, 1994 and 1995. Song sparrows are year-round residents at both sites with most pairs remaining on their territories throughout the year. Song sparrows are small emberizine finches with a body mass of 18–25 g. The sexes are monomorphic in plumage. While males are generally larger, considerable overlap in size exists between the sexes. Males sing well-developed songs (Nice 1937, 1943). In contrast, song sparrow females sing only rarely (Nice 1943; Arcese et al. 1988). Song sparrows have a variety of calls, some of which are female specific and some of which are shared by males and females (Nice 1943; Elekonich 1997, 1998). Primary territory settlement occurs in the early spring prior to breeding. Juveniles of both sexes also attempt to settle into already established groups of territories in the autumn (Arcese 1987, 1989a, c; Beecher et al. 1994). I caught female song sparrows in the spring, summer and autumn (prebreeding, breeding and postmoult seasons) using food-baited, walk-in traps or mist nets. They were banded with a unique combination of colour bands and a numbered U.S. Fish and Wildlife Service band. I sexed females by laparotomy if their sex could not be determined by their use of female-specific vocalizations (Nice 1943; Elekonich 1997, 1998) and by observations of copulations between females and banded males. During the breeding season, I located song sparrow nests and checked each nest every 2 days to ascertain the stage in the breeding cycle of each female. I assumed an unbanded female on a nest was the same bird between successive nest checks. All unbanded subjects were subsequently banded during the simulated intrusion trials (see below).
Simulated Intrusions Female song sparrows were presented with either a simulated female song sparrow intrusion (N=49), or a simulated spotted towhee, Pipilo maculatus, intrusion (N=30) as a control for general arousal to playback, the presence of observers, and species specificity of the aggressive response. All presentations occurred between 0700 and 1200 hours. Prebreeding tests generally occurred between late February and early April. Breeding season tests occurred between April and August, when female subjects were on nests incubating eggs. Most birds moulted in August to early September and postmoult tests began immediately following that year’s moult and concluded in mid-October. To facilitate a direct comparison between the sexes, the experimental protocol was designed to match closely previous experiments using male song sparrows (see Wingfield 1985; Stoddard et al. 1991), but to allow for differences in female responses, I used a full behavioural ethogram including nonaggressive behaviours. Male song sparrows display stronger behavioural and hormonal reactions to a live song sparrow presentation than to either a mount alone, song alone, or a mount presented with song (Wingfield & Wada 1989). Thus, I presented resident females with female song sparrow vocalizations played on a Marantz PMD-221 tape recorder through a Sony speaker placed next to a cage containing a live female song sparrow. To capture females to obtain blood samples for later hormone analysis (see Elekonich 1997), both the cage and speaker were placed at the base of a mist net which was open when the trial began. Male song sparrows react more intensely to stimuli presented at the territory centre (Stoddard et al. 1991). Therefore, I presented simulated female song sparrow intrusions in the centre of the territory during the spring (prebreeding) and autumn (postmoult) seasons. During the breeding season, however, the nest is the centre of female activity on the territory and a female’s territory boundaries may not always match those of her mate, so presentations occurred 10–15 m from the nest towards the centre of the territory. This distance ensured that a female was within hearing distance of the playback, but was far enough to ensure that female responses to territorial intrusion were measured rather than nest defence. The playback tape for simulated female song sparrow intrusions consisted of groups of female vocalizations selected from field recordings to mimic the length and mixture of call types of naturally observed female vocalizations. Including vocalizations used at the nest, Nice (1943) suggested that females have 15 calls, 12 of which are shared with males. Sound spectrographic analysis of calls recorded during observations of unmanipulated birds indicates that females away from the nest actually use only six calls, four of which are shared with males (Elekonich 1997, 1998). Two of these shared calls, the ‘chitter’ and ‘growl’, are used during aggressive interactions. Chitters are multipurpose calls used by females in aggressive and nonaggressive contexts. Females use chitters for inciting male territorial aggression, during early stages of female–female
ELEKONICH: FEMALE RESPONSE TO INTRUSION
Table 1. Behaviours scored during simulated female song sparrow and towhee intrusions Behaviour
Definition
Foraging
Feeding: bird moves about pecking at ground and consuming seeds or gleaning insects from plants and trees. Female walks or flies closer to caged decoy or mount and net. Female flies from one position to another. Low, single-note calls used to maintain contact between the pair or family group. High-pitched, single-note call similar to ‘tseets’, which is used to signal alarm such as when a predator is present. Chets are single notes of the same type as those given in the more complex chitter vocalization; chitters and chets are often given together; multipurpose calls given prior to female copulation solicitation, when female gets off the nest during incubation, and during aggressive encounters; and to incite males. Buzzing sound, ‘zhee, zhee, zhee’ of Nice (1943), used primarily during copulation solicitation. Low-pitched, broadband call used by both sexes during aggressive encounters; precedes actual physical aggression. Female crouches with wing tips pointing down, head forward, the crest raised or sleeked, tail down, waving or flipping wings while facing intruder; often accompanied by growl vocalizations. Female perches on branch or stands on ground, gives chitters and buzzes, crouches with tail up and ‘vibrates’ or ‘shivers’ wings; mate is often nearby or arrives soon after the female begins this display. Female performs behaviour associated with nesting or caring for young; including incubating eggs, feeding or brooding nestlings or feeding fledglings. The closest distance the focal female came to the caged bird/mount and speaker during the trial.
Approaches Flight Low chip (contact call) High chip (alarm call) Chitter/chet
Buzz Growl Wing-wave threat posture Copulation solicitation Parental care Closest approach
aggression, during copulation solicitations and to alert the male when a female leaves the nest (Nice 1943; Arcese 1989b, c; Wingfield & Hahn 1994; Elekonich 1997, 1998). Chitters are formed from a number of single notes, which I term ‘chets’ (Elekonich 1997, 1998), and consist of a rapid downsweep followed by an upsweep. Females use these notes alone or grouped with other calls and their function is similar to that of the full chitter call (Elekonich 1997, 1998). Males do not use the chitter call during the prebreeding or breeding seasons but have been observed to use it during autumn territorial interactions (Elekonich 1998). The growl is the primary aggressive signal. Both males and females use the growl during fully escalated aggressive interactions just prior to actual physical contact between the combatants (Nice 1943; Arcese 1989b; Elekonich 1997, 1998). Females use the ‘sweep buzz’ call primarily during copulation solicitations, but occasionally also use it during early stages of territorial aggression, perhaps to alert their mates (Elekonich 1997). Both males and females share low-chip (contact) and high-chip (alarm) calls (Nice 1943; Elekonich 1997, 1998). The high-chip call is a classic ‘tseet’ type alarm call (Marler 1955). I prepared stimulus tapes using vocalizations from six different female song sparrows to reduce pseudoreplication (Kroodsma 1989a, b, 1990). These vocalizations were recorded in 1991 and are representative of the frequency, structure, distribution by call type and length observed in response to female playback (for reference: mean call lengthSE including pauses=0.700.51 s). Groups of vocalizations were separated by 2 s of silence. Within a group, calls were presented in naturally occurring combinations. Within a call type, pauses were as they naturally occurred. The stimulus tape included 56% chitters and chets, 22% low chips (contact calls), 13%
growls, 9% buzzes and 0% high chips (alarm calls; see Elekonich 1998 for sonagrams of each call type). I recorded the following behavioural responses during the experiment: closest approach, foraging, flights, threat displays, parental care and vocalizations (see Table 1). All female subjects were observed on the territory within 24 h prior to the experiment, but due to their cryptic behaviour it was often difficult to locate the focal female at the very beginning of the trial. Because long latencies to respond could reflect variance in female location relative to the speaker/caged bird at the beginning of the trial, latency to respond was not measured. Trials continued for 30 min or until the subject was netted as described. I chose simulated spotted towhee intrusions as a behavioural control because spotted towhees are sympatric with song sparrows and use similar feeding and nesting resources. Spotted towhees are also a resident species in western Washington. Thus, unlike many of the other species of sparrows, towhees were found at both sites year-round. However, due to the difficulty of humanely keeping towhees in captivity, I presented a towhee mount, rather than a caged towhee, and taped vocalizations of a spotted towhee at the centre of the territory or 10–15 m from the nest as described above. I made the tape of towhee vocalizations from songs and calls from two individuals recorded in the field. Due to the effectiveness of the tape in attracting towhees, towhee presentations lasted 20 min (versus 30 min for song sparrow presentations) or until the level of towhee response necessitated removal of the towhee mount to keep it intact for future presentations. I sampled each subject only once during the experiment to preclude the possibility that previous exposure to the net/speaker/decoy changed a female’s response. To
553
554
ANIMAL BEHAVIOUR, 59, 3
Table 2. Component loadings for principle components 1–3 following square-root transformation of the behaviours measured and equamax rotation of the components Behaviour
Factor 1
Factor 2
Factor 3
Approach Flights Closest approach Wing-wave threat display Buzzes Chitters and chets Growls Low chip (contact) calls High chip (alarm) calls Parental care Foraging Copulation solicitation
0.866 0.817 −0.670 0.756 0.663 0.129 0.482 0.031 0.630 −0.070 0.300 −0.090
0.166 0.295 −0.212 −0.205 −0.011 −0.183 −0.374 0.629 −0.017 0.777 0.653 0.078
0.025 0.252 0.116 0.287 0.439 0.758 0.452 −0.089 0.041 −0.069 0.027 0.836
control for day effects and distribute treatments equally across the study site, I matched female song sparrows for age and stage in the breeding cycle on noncontiguous territories and presented them with either a song sparrow or towhee simulated intrusion.
Analysis After dropping incomplete trials, 55 trials were analysed (32 female song sparrow presentations and 25 towhee presentations). Although song sparrow presentations lasted 30 min (in an effort to get hormone samples for another study: Elekonich 1997), I used only the first 20 min of the song sparrow trials. I counted each occurrence of each behaviour for all measures except closest approach, alarm calls, contact calls and parental care. I scored alarm calls, contact calls and parental care as 1 (occurred) or 0 (did not occur) for each minute of the trial (see Table 1). I analysed these data using SYSTAT for Windows (version 8.0). Data were square-root transformed to meet the assumptions for parametric tests, then subjected to principle components analysis with an equamax rotation. I further analysed the three factors (of 12) which had eigen values greater than 1 and at least three variables with component loadings greater than 3 (Wilkinson et al. 1996). Factor 1 accounted for 30.2% of the variance. Factors 2 and 3 accounted for 15.1 and 15.4% of the variance, respectively. Approaches, flights, growls, wing-wave threats, closest approach, buzzes and high chips (alarm calls) loaded most heavily onto factor 1 (Table 2). Although factor 1 could simply represent arousal, it may represent aggression. Wing-wave threat displays and growls are well known as the best indicators of escalated aggression for males and females (Nice 1937, 1943; Arcese 1989b; Elekonich 1997) and both loaded heavily onto this factor. Approaches, flights and closest approach are all indicators of increased movement, mostly towards the simulated intruder. The combination of approaches, flights, growls and wing-wave threat displays is observed during natural territorial encounters between females (Nice 1943; Arcese
1989b; Elekonich 1997) and is also common to male aggression (Nice 1937, 1943; Stoddard et al. 1991; Wingfield 1985). Because these presentations continued longer than territorial interactions normally do, the presence of alarm calls is not entirely surprising. Parental behaviours, foraging and low chips (contact calls) loaded heavily onto factor 2. Growls loaded negatively onto factor 2, suggesting that factor 2 is representative of a nonaggressive state where the subject’s focus is on something other than the simulated intruder (Table 2). Buzzes, copulation solicitations, chitters and growls all loaded heavily onto factor 3. Except growls, all of these vocalizations are involved in either inciting male territorial aggression or courtship (Elekonich 1997, 1998) suggesting that factor 3 represents behaviour directed towards the male mate during the context of a simulated female intrusion. I performed t tests using the factor scores to compare female responses to song sparrow versus towhee simulated intrusions within each season. Because the groups typically had unequal variances, I used the separate variance t value. I used a Dunn–Sidak correction to control for the effects of multiple comparisons. I evaluated changes in song sparrow females’ responses to simulated female song sparrow intrusion across the seasons with analysis of variance (ANOVA) followed by a post-hoc Tukey’s HSD test for each factor.
RESULTS
Song Sparrow versus Spotted Towhee Simulated Intrusions within Each Season The females’ responses represented by factor 1 were significantly higher to simulated intrusions by female song sparrows than to those of towhees in each of the three seasons, prebreeding, breeding and postmoult (Table 3). Females responses to simulated intrusion by female song sparrows and towhees did not differ significantly during any season for factor 2. Females responses represented by factor 3 were significantly higher to female song sparrow intrusion than to towhee intrusion in the postmoult season, but did not differ significantly in either the prebreeding or the breeding seasons (Table 3).
Seasonality of Female Song Sparrow Response to Simulated Female Intrusion Factor 1 varied significantly across the three seasons (F2,34 =3.56, P=0.039). Female aggressive responses to simulated female song sparrow intrusion were significantly higher in the prebreeding (Tukey’s HSD: df=34, P=0.040) season. Factor 2 scores were higher during the breeding season but did not differ significantly across seasons (F2,34 =2.78, P=0.076). Factor 3 scores were highest in the prebreeding season and dropped across the three seasons but these seasonal changes were not significant (F2,34 =1.57, P=0.224).
ELEKONICH: FEMALE RESPONSE TO INTRUSION
Table 3. Female responses to simulated female song sparrow (S) and spotted towhee (T) intrusions within each season Principle component Season
Factor 1
Factor 2
Factor 3
Prebreeding t N df P S>T?
4.12 13.1 10,6 0.004 >
1.07 13.5 10,6 0.66 =
0.40 12 10,6 0.97 =
Breeding t N df P S>T?
3.82 24.8 19,10 0.002 >
1.26 26.9 19,10 0.52 =
0.118 19.5 19,10 0.99 =
Postmoult t N df P S>T?
3.86 9.8 8,6 0.01 >
2.52 12 8,6 0.08 =
2.94 11.8 8,6 0.04 =
DISCUSSION Female song sparrows presented with a simulated intrusion by a female conspecific either responded to the intruder, signalled their mates or did not respond to the intruder, choosing instead to forage or care for young. Principle component 1 indicated that female song sparrows overall directed more behaviours (aggressive growls and wing-wave threat displays) towards an intruder than towards their mates in the prebreeding season than in the breeding or postbreeding seasons. Male song sparrow aggression is similarly highest during the prebreeding season, then declines during the breeding season, and increases again in early autumn (postmoult) when firstyear birds attempt to gain territories (Arcese 1987, 1989a, c; Beecher et al. 1994; Wingfield & Hahn 1994). Within each season females responded more strongly and aggressively to a simulated intrusion by a conspecific than a heterospecific as indicated by factor 1. These data suggest that female aggression, like male aggression, functions to defend access to resources on the territory and is directed specifically to conspecifics and not generalized to all birds using similar resources. Alternatively, females might react to heterospecifics that compete for resources, but females in these populations did not consider towhees resource competitors, or towhees are resource competitors but the towhee model and vocalizations were not adequate representations of a towhee. The former is more likely, as the tape/mount were so effective at attracting towhees that on a number of occasions I had to remove the mount to keep it intact for future trials. Arcese (1989a) suggested that male song sparrows defend access to females rather than resources on the territory. Similarly females may respond less to towhee intrusion than to song sparrow intrusion because they are
defending something other than the territory’s resources. Females could defend access to the male and his contributions of defence of the nest and territory and feeding of the nestlings and fledglings. Alternatively, female aggression could defend against intraspecific brood parasitism (e.g. Gowaty & Wagner 1988). However, intraspecific brood parasitism has not been observed in song sparrows (Keller 1996). Arcese (1989b) also observed a decrease in female aggression at the start of the breeding season, which he suggested was due to time constraints arising from the greater reproductive effort of females. Female song sparrow parental investment consists of building the nest, laying and incubating the eggs, and defending, brooding and feeding the young. In contrast, male song sparrows’ contribution includes only defending the nest and feeding the young (Nice 1943). Once a female begins to build a nest, any time she spends defending the territory comes at the cost of reproduction or self-maintenance. When Arcese (1989b) artificially increased the amount of food on territories, females spent more time guarding the territory and chasing floaters (birds without territories) than females whose territories did not have supplemental food. These data suggest that females are indeed defending resources on a territory and that a trade-off between territorial and reproductive behaviours exists for females. The lack of increase in female aggression following the moult (in autumn) when juveniles are attempting to settle onto territories (e.g. Beecher et al. 1994) suggests that females may not benefit from autumn territoriality. However, the within-season analysis of factor 3 suggests that in autumn females may be signalling their mates to defend the territory for them. Furthermore, female floaters are rare in these populations (Elekonich 1997). Thus, there may be less selective pressure on females to set up territories in their first year, because any female that survives the winter will breed. Alternatively, the resource females defend may be more seasonal, or vary more within a season. Dickinson & Lein (1987) found that nest site selection by female red-winged blackbirds, Agelaius phoeniceus, determines territory boundaries. Similarly, in several cases the site of a female song sparrow’s nest caused her mate to fight the neighbouring male in order to increase his territory to include the nest (Elekonich 1997). However, if the structural features of territories that determine the location of good nest sites often change between autumn and spring, females may not benefit by defending a territory boundary in autumn. Nonaggressive responses to simulated intrusions were most likely directed to mates or young as suggested by principle components 2 and 3. The responses represented by factor 2 consist of both self-maintenance and care of nestlings. Thus, it is not surprising that factor 2 scores peaked in the breeding season (although not significantly) when nutritional needs for producing eggs and/or for feeding young were highest. Buzzes, copulation solicitations, chitters and growls all loaded onto factor 3, which appears to represent signals directed towards the female’s mate. Buzzes are most often
555
556
ANIMAL BEHAVIOUR, 59, 3
used during copulation solicitations (Elekonich 1997, 1998). Although sometimes used during territorial interactions, females most often use chitters to incite male–male aggression, during copulation solicitation and when leaving the nest (Nice 1943; Wingfield & Hahn 1994; Elekonich 1997, 1998). Given the role of inciting in courtship and pair bond maintenance (Lorenz 1958), it is not surprising that chitters loaded heavily onto principle component 3, rather than principle component 1, which appears to represent signals directed towards female intruders. Especially in the early spring, natural male– male and female–female territorial interactions often occur (Nice 1937, 1943; Arcese 1989b; Elekonich 1997). During these interactions, females may use the vocalizations represented by factor 3 to call their mates to help defend the territory. In the present study, when males did respond to their mate’s chitters during simulated female song sparrow intrusions, they often perched nearby and watched rather than performing any aggressive displays. Although females’ responses reflected by factor 3 were higher to song sparrows than towhees in every season, the difference was only significant in the autumn, and growls loaded less heavily onto factor 3 than the other vocalizations. Furthermore, females gave more growls per minute of playback in the spring. Thus, the loading of growls onto factor 3 may reflect the natural situation of pair defence of boundaries, with increased female territoriality in spring as reflected by factor 1 and increased defence by males only in autumn. In summary, female song sparrows appear to defend territorial resources. But investigation of nonaggressive behaviours suggests that there are trade-offs for females between territorial behaviour and reproduction. These trade-offs may select for females that require fewer interactions to establish territorial boundaries and conclude aggressive interactions more quickly than males. The results of this study suggest that sex differences may exist for each parameter (e.g. seasonality and context) of behaviour. Furthermore, the inclusion of nonaggressive behaviours and the presentation of different types of stimuli provide a more complete picture of the natural contexts of aggressive behaviour in song sparrows. Acknowledgments I thank John Garrett, manager of the Skagit Wildlife Area, the Washington Department of Wildlife, the Seattle Parks Department and staff of Discovery Park for hosting the field work; Liz Campbell, Cully Nordby, Chris Hill, Sally Boeve, Jennie Maughn, Tami Nelson, Chris Robertson and Sean Schmidt for assistance in the field; Jim Ha and Sam Beshers for statistical advice, and Mike Beecher, Melissa Fleming, Sean O’Donnell, Philip Stoddard, Judy Stamps and two anonymous referees for comments. This research was supported in part by grants from Sigma Xi and the American Psychological Association. The research presented here was described in Animal Research Protocol No. 2207-02, approved on 22 June 1993 by the Institutional Animal Care and Use Committee of the University of Washington.
References Arcese, P. 1987. Age, intrusion pressure and defence against floaters by territorial male song sparrows. Animal Behaviour, 35, 773–784. Arcese, P. 1989a. Territory acquisition and loss in male song sparrows. Animal Behaviour, 37, 45–55. Arcese, P. 1989b. Intrasexual competition and the mating system in primarily monogamous birds: the case of the song sparrow. Animal Behaviour, 38, 96–111. Arcese, P. 1989c. Intrasexual competition, mating system and natal dispersal in song sparrows. Animal Behaviour, 38, 958– 979. Arcese, P., Stoddard, P. K. & Hiebert, S. M. 1988. The form and function of song in female song sparrows. Condor, 90, 44–50. Archer, J. 1988. The Behavioural Biology of Aggression. Cambridge: Cambridge University Press. Beecher, M. D., Campbell, S. E. & Stoddard, P. K. 1994. Correlation of song learning and territory establishment strategies in the song sparrow. Proceedings of the National Academy of Sciences U.S.A., 91, 1450–1454. Beletsky, L. D. 1982. Vocalizations of female northern orioles. Condor, 84, 445–447. Dickinson, T. E. & Lein, M. R. 1987. Territory dynamics and patterns of female recruitment in red-winged blackbirds (Agelaius phoeniceus). Canadian Journal of Zoology, 65, 465–471. Elekonich, M. M. 1997. Female–female territorial aggression and its hormonal control in the song sparrow. Ph.D. thesis, University of Washington. Elekonich, M. M. 1998. Song sparrow males use female-typical vocalizations in fall. Condor, 100, 145–148. Gowaty, P. & Wagner, S. 1988. Breeding season aggression of male and female eastern bluebirds to models of potential conspecific and interspecific egg dumpers. Ethology, 78, 238–250. Holmes, R. T., Sherry, T. W. & Reitsma, L. 1989. Population structure, territoriality and overwinter survival of two migrant warbler species in Jamaica. Condor, 91, 545–561. Hrdy, S. B. 1981. The Woman That Never Evolved. Cambridge, Massachusetts: Harvard University Press. Hrdy, S. B. & Williams, G. C. 1983. Behavioral biology and the double standard. In: Social Behavior of Female Vertebrates (Ed. by S. K. Wasser), pp. 1–18. New York: Academic Press. Hurly, T. A. & Robertson, R. J. 1984. Aggressive and territorial behaviour in female red-winged blackbirds. Canadian Journal of Zoology, 62, 148–153. Jenni, D. A. & Collier, G. 1972. Polyandry in the American jacana (Jacana spinosa). Auk, 89, 743–765. Keller, L. 1996. Inbreeding in an insular population of song sparrows. Ph.D. thesis, University of Wisconsin, Madison. Kroodsma, D. E. 1989a. Suggested experimental designs for song playbacks. Animal Behaviour, 37, 600–609. Kroodsma, D. E. 1989b. Inappropriate experimental designs impede progress in bioacoustic research: a reply. Animal Behaviour, 38, 717–719. Kroodsma, D. E. 1990. Using appropriate experimental designs for intended hypotheses in ‘song’ playbacks, with expamples for testing effects of song repertoire sizes. Animal Behaviour, 40, 1138–1150. Levin, R. 1996a. Song behaviour and reproductive strategies in a duetting wren, Thryothorus nigricapillus: 1. Removal experiments. Animal Behaviour, 52, 1093–1106. Levin, R. 1996b. Song behaviour and reproductive strategies in a duetting wren, Thryothorus nigricapillus: II. Playback experiments. Animal Behaviour, 52, 1107–1117. Lorenz, K. Z. 1958. The evolution of behavior. Scientific American, 199, 67–83. Marler, P. 1955. Characteristics of some animal calls. Nature, 176, 6–8.
ELEKONICH: FEMALE RESPONSE TO INTRUSION
Montgomerie, R. D. & Weatherhead, P. J. 1988. Risks and rewards of nest defence in birds. Quarterly Review of Biololgy, 63, 167–187. Nice, M. M. 1937. Studies in the Life History of the Song Sparrow. I. A Population Study of the Song Sparrow and Other Passerines. New York: Dover. Nice, M. M. 1943. Studies in the Life History of the Song Sparrow. II. The Behaviour of the Song Sparrow and Other Passerines. New York: Dover. Richison, G. 1983. The function of singing in female black-headed grosbeaks (Pheucticus melancephalus). Auk, 100, 105–116. Richison, G. 1986. The singing behavior of female northern cardinals. Condor, 88, 156–159. Sandell, M. I. & Smith, H. G. 1997. Female aggression in the European starling during the breeding season. Animal Behaviour, 53, 13–23. Slagsvold, T. 1993. Female–female aggression and monogamy in great tits Parus major. Ornis Scandinavica, 22, 319–326. Stoddard, P. K., Beecher, M. D., Horning, C. L. & Campbell, S. E. 1991. Recognition of individual neighbors by song in the song sparrow, a species with song repertoires. Behavioral Ecology and Sociobiology, 29, 211–215. Trivers, R. L. 1972. Parental investment and sexual selection. In: Sexual Selection and the Descent of Man (Ed. by B. Campbell), pp. 139–179. New York: Aldine.
Wiklund, C. G. & Village, A. 1992. Sexual and seasonal variation in territorial behaviour of kestrels, Falco tinnunculus. Animal Behaviour, 43, 823–830. Wilkinson, L., Blank, G. & Gruber, C. 1996. Desktop Data Analysis with SYSTAT. Upper Saddle River, New Jersey: Prentice Hall. Wingfield, J. C. 1985. Short-term changes in plasma levels of hormones during establishment and defence of a breeding territory in male song sparrows, Melospiza melodia. Hormones and Behavior, 19, 174–187. Wingfield, J. C. & Hahn, T. P. 1994. Testosterone and territorial behaviour in sedentary and migratory sparrows. Animal Behaviour, 47, 77–89. Wingfield, J. C. & Wada, M. 1989. Male–male interactions increase both luteinizing hormone and testosterone in the song sparrow, Melospiza melodia: specificity, time course and possible neural pathways. Journal of Comparative Physiology A, 166, 189–194. Wingfield, J. C., Ball, G. F., Dufty, A. M., Jr, Hegner, R. E. & Ramenofsky, M. 1987. Testosterone and aggression in birds: tests of the ‘challenge hypothesis’. American Scientist, 75, 602–608. Wittenberger, J. F. & Tilson, R. L. 1980. The evolution of monogamy: hypotheses and evidence. Annual Review of Ecology and Systematics, 11, 197–232. Yasukawa, K. & Searcy, W. A. 1982. Aggression in female redwinged blackbirds: a strategy to ensure male parental investment. Behavioral Ecology and Sociobiology, 11, 13–17.
557
Female song sparrow, Melospiza melodia, response to simulated conspecific and heterospecific intrusion across three seasons MICHELLE M. ELEKONICH
Department of Psychology, University of Washington (Received 13 September 1999; initial acceptance 23 September 1999; final acceptance 29 November 1999; MS. number: A8319R)
To investigate female responses to territorial intrusion I presented female song sparrows with either a simulated female song sparrow intrusion or a simulated spotted towhee, Pipilo maculatus, intrusion as a control during either the prebreeding, breeding or postmoult seasons. Aggressive and nonaggressive behaviours and vocalizations were compared between intrusion types and across seasons. Principle components analysis suggested that female responses fell into three categories: (1) responses directed towards the intruder, mostly aggressive; (2) responses directed towards the mate; and (3) lack of response to the intruder. In every season, females responded more aggressively to simulated female song sparrow intrusion than simulated towhee intrusion. Responses directed towards female song sparrow intruders dropped across the three seasons and were significantly higher in the prebreeding season than in the breeding or postmoult seasons.
Males are viewed as the more aggressive sex in many animal species due to male specializations for intrasexual competition (reviewed in Archer 1988) and most studies of aggression in birds have focused on males of north temperate species. Most studies of female aggression focused on cases where females were expected to be under male-like selection pressures and evolve male-like specializations for intrasexual competition (e.g. sex-role reversed species: Jenni & Collier 1972; polygynous species: Yasukawa & Searcy 1982; Wittenberger & Tilson 1980; species where females regularly sing: Richison 1983, 1986; Beletsky 1982). Although various investigators have compared male intrasexual territorial aggression to female nest defence, comparisons like these are inappropriate because aggression serves different functions in each context and has been shaped by different selection pressures. Only in a few avian species has aggression in males and females been tested in comparable ways to allow direct comparison (Holmes et al. 1989; Wicklund & Village 1992; Slagsvold 1993; Levin 1996a, b). Due to differences in parental investment, natural selection often produces distinct reproductive and life history strategies in the sexes (Trivers 1972). Hrdy (1981; Hrdy & Williams 1983) argues that sex differences when they occur are most likely to be present in traits used to compete with conspecifics. Sex
differences should occur not only with respect to the development of particular behaviours, but also with respect to the function, hormonal control and social context of those behaviours. In song sparrows, the function, hormonal control, seasonality and context of male–male aggression have been particularly well studied (e.g. Wingfield et al. 1987; Wingfield & Hahn 1994), and females are known to engage in intraspecific aggression (Nice 1943; Arcese 1989b). In the present study I examined female response to territorial intrusion by female conspecifics to allow a direct comparison between male and female responses to simulated intrusions throughout the year. Although female defences against intraspecific parasitism have received considerable attention (e.g. Gowaty & Wagner 1988; Sandell & Smith 1997), there is no observational or genetic evidence for egg dumping in song sparrows (Nice 1943; Keller 1996). Thus, it is unlikely that female–female aggression in song sparrows functions to protect against intraspecific brood parasitism. Reported functions of female aggression in birds include: defence of a nest or young (reviewed in Montgomerie & Weatherhead 1988), reduction of intraspecific or interspecific brood parasitism (e.g. Gowaty & Wagner 1988), maintenance of pair or family group bonds (e.g. Ritchison 1983) and limiting female competitors’ access to mated territorial males (e.g. Hurley & Robertson 1984), parental care or defence provided by males (e.g. Yasukawa & Searcy 1982), or access to feeding and/or breeding territories (e.g. Hurley & Robertson 1984; Levin 1996a, b).
Correspondence and present address: M. M. Elekonich, Department of Entomology, MC-118, University of Illinois, 505 South Goodwin, Urbana, IL 61801, U.S.A. (email: [email protected]). 0003–3472/00/030551+07 $35.00/0
2000 The Association for the Study of Animal Behaviour
551
2000 The Association for the Study of Animal Behaviour
552
ANIMAL BEHAVIOUR, 59, 3
If female song sparrow aggression functions primarily for defence of the territory and its resources, as it does for males (Wingfield 1985; Stoddard et al. 1991; Wingfield & Hahn 1994), the seasonal characteristics of female territorial aggression should parallel those of male territorial aggression. In Washington State where these experiments occurred song sparrows are a resident species and many pairs remain on their territories throughout the year. Thus, if this hypothesis is true, female aggression should occur throughout the year and peak when territories are being settled (i.e. prior to nesting in the spring and again in the autumn as is the case for males; Beecher et al. 1994). If the primary function of female territoriality is to defend the food and nest sites available on the territory then female aggression should be directed equally towards conspecifics and heterospecifics that use the same resources.
METHODS
Sites and Subjects The experiments were conducted at the Skagit Valley Wildlife Recreation Area, near Conway, Washington, about 80.5 km north of Seattle in 1992, 1993 and 1994, and at Discovery Park, a 3.2-km2 undeveloped park along Puget Sound in Seattle, Washington, in 1990, 1994 and 1995. Song sparrows are year-round residents at both sites with most pairs remaining on their territories throughout the year. Song sparrows are small emberizine finches with a body mass of 18–25 g. The sexes are monomorphic in plumage. While males are generally larger, considerable overlap in size exists between the sexes. Males sing well-developed songs (Nice 1937, 1943). In contrast, song sparrow females sing only rarely (Nice 1943; Arcese et al. 1988). Song sparrows have a variety of calls, some of which are female specific and some of which are shared by males and females (Nice 1943; Elekonich 1997, 1998). Primary territory settlement occurs in the early spring prior to breeding. Juveniles of both sexes also attempt to settle into already established groups of territories in the autumn (Arcese 1987, 1989a, c; Beecher et al. 1994). I caught female song sparrows in the spring, summer and autumn (prebreeding, breeding and postmoult seasons) using food-baited, walk-in traps or mist nets. They were banded with a unique combination of colour bands and a numbered U.S. Fish and Wildlife Service band. I sexed females by laparotomy if their sex could not be determined by their use of female-specific vocalizations (Nice 1943; Elekonich 1997, 1998) and by observations of copulations between females and banded males. During the breeding season, I located song sparrow nests and checked each nest every 2 days to ascertain the stage in the breeding cycle of each female. I assumed an unbanded female on a nest was the same bird between successive nest checks. All unbanded subjects were subsequently banded during the simulated intrusion trials (see below).
Simulated Intrusions Female song sparrows were presented with either a simulated female song sparrow intrusion (N=49), or a simulated spotted towhee, Pipilo maculatus, intrusion (N=30) as a control for general arousal to playback, the presence of observers, and species specificity of the aggressive response. All presentations occurred between 0700 and 1200 hours. Prebreeding tests generally occurred between late February and early April. Breeding season tests occurred between April and August, when female subjects were on nests incubating eggs. Most birds moulted in August to early September and postmoult tests began immediately following that year’s moult and concluded in mid-October. To facilitate a direct comparison between the sexes, the experimental protocol was designed to match closely previous experiments using male song sparrows (see Wingfield 1985; Stoddard et al. 1991), but to allow for differences in female responses, I used a full behavioural ethogram including nonaggressive behaviours. Male song sparrows display stronger behavioural and hormonal reactions to a live song sparrow presentation than to either a mount alone, song alone, or a mount presented with song (Wingfield & Wada 1989). Thus, I presented resident females with female song sparrow vocalizations played on a Marantz PMD-221 tape recorder through a Sony speaker placed next to a cage containing a live female song sparrow. To capture females to obtain blood samples for later hormone analysis (see Elekonich 1997), both the cage and speaker were placed at the base of a mist net which was open when the trial began. Male song sparrows react more intensely to stimuli presented at the territory centre (Stoddard et al. 1991). Therefore, I presented simulated female song sparrow intrusions in the centre of the territory during the spring (prebreeding) and autumn (postmoult) seasons. During the breeding season, however, the nest is the centre of female activity on the territory and a female’s territory boundaries may not always match those of her mate, so presentations occurred 10–15 m from the nest towards the centre of the territory. This distance ensured that a female was within hearing distance of the playback, but was far enough to ensure that female responses to territorial intrusion were measured rather than nest defence. The playback tape for simulated female song sparrow intrusions consisted of groups of female vocalizations selected from field recordings to mimic the length and mixture of call types of naturally observed female vocalizations. Including vocalizations used at the nest, Nice (1943) suggested that females have 15 calls, 12 of which are shared with males. Sound spectrographic analysis of calls recorded during observations of unmanipulated birds indicates that females away from the nest actually use only six calls, four of which are shared with males (Elekonich 1997, 1998). Two of these shared calls, the ‘chitter’ and ‘growl’, are used during aggressive interactions. Chitters are multipurpose calls used by females in aggressive and nonaggressive contexts. Females use chitters for inciting male territorial aggression, during early stages of female–female
ELEKONICH: FEMALE RESPONSE TO INTRUSION
Table 1. Behaviours scored during simulated female song sparrow and towhee intrusions Behaviour
Definition
Foraging
Feeding: bird moves about pecking at ground and consuming seeds or gleaning insects from plants and trees. Female walks or flies closer to caged decoy or mount and net. Female flies from one position to another. Low, single-note calls used to maintain contact between the pair or family group. High-pitched, single-note call similar to ‘tseets’, which is used to signal alarm such as when a predator is present. Chets are single notes of the same type as those given in the more complex chitter vocalization; chitters and chets are often given together; multipurpose calls given prior to female copulation solicitation, when female gets off the nest during incubation, and during aggressive encounters; and to incite males. Buzzing sound, ‘zhee, zhee, zhee’ of Nice (1943), used primarily during copulation solicitation. Low-pitched, broadband call used by both sexes during aggressive encounters; precedes actual physical aggression. Female crouches with wing tips pointing down, head forward, the crest raised or sleeked, tail down, waving or flipping wings while facing intruder; often accompanied by growl vocalizations. Female perches on branch or stands on ground, gives chitters and buzzes, crouches with tail up and ‘vibrates’ or ‘shivers’ wings; mate is often nearby or arrives soon after the female begins this display. Female performs behaviour associated with nesting or caring for young; including incubating eggs, feeding or brooding nestlings or feeding fledglings. The closest distance the focal female came to the caged bird/mount and speaker during the trial.
Approaches Flight Low chip (contact call) High chip (alarm call) Chitter/chet
Buzz Growl Wing-wave threat posture Copulation solicitation Parental care Closest approach
aggression, during copulation solicitations and to alert the male when a female leaves the nest (Nice 1943; Arcese 1989b, c; Wingfield & Hahn 1994; Elekonich 1997, 1998). Chitters are formed from a number of single notes, which I term ‘chets’ (Elekonich 1997, 1998), and consist of a rapid downsweep followed by an upsweep. Females use these notes alone or grouped with other calls and their function is similar to that of the full chitter call (Elekonich 1997, 1998). Males do not use the chitter call during the prebreeding or breeding seasons but have been observed to use it during autumn territorial interactions (Elekonich 1998). The growl is the primary aggressive signal. Both males and females use the growl during fully escalated aggressive interactions just prior to actual physical contact between the combatants (Nice 1943; Arcese 1989b; Elekonich 1997, 1998). Females use the ‘sweep buzz’ call primarily during copulation solicitations, but occasionally also use it during early stages of territorial aggression, perhaps to alert their mates (Elekonich 1997). Both males and females share low-chip (contact) and high-chip (alarm) calls (Nice 1943; Elekonich 1997, 1998). The high-chip call is a classic ‘tseet’ type alarm call (Marler 1955). I prepared stimulus tapes using vocalizations from six different female song sparrows to reduce pseudoreplication (Kroodsma 1989a, b, 1990). These vocalizations were recorded in 1991 and are representative of the frequency, structure, distribution by call type and length observed in response to female playback (for reference: mean call lengthSE including pauses=0.700.51 s). Groups of vocalizations were separated by 2 s of silence. Within a group, calls were presented in naturally occurring combinations. Within a call type, pauses were as they naturally occurred. The stimulus tape included 56% chitters and chets, 22% low chips (contact calls), 13%
growls, 9% buzzes and 0% high chips (alarm calls; see Elekonich 1998 for sonagrams of each call type). I recorded the following behavioural responses during the experiment: closest approach, foraging, flights, threat displays, parental care and vocalizations (see Table 1). All female subjects were observed on the territory within 24 h prior to the experiment, but due to their cryptic behaviour it was often difficult to locate the focal female at the very beginning of the trial. Because long latencies to respond could reflect variance in female location relative to the speaker/caged bird at the beginning of the trial, latency to respond was not measured. Trials continued for 30 min or until the subject was netted as described. I chose simulated spotted towhee intrusions as a behavioural control because spotted towhees are sympatric with song sparrows and use similar feeding and nesting resources. Spotted towhees are also a resident species in western Washington. Thus, unlike many of the other species of sparrows, towhees were found at both sites year-round. However, due to the difficulty of humanely keeping towhees in captivity, I presented a towhee mount, rather than a caged towhee, and taped vocalizations of a spotted towhee at the centre of the territory or 10–15 m from the nest as described above. I made the tape of towhee vocalizations from songs and calls from two individuals recorded in the field. Due to the effectiveness of the tape in attracting towhees, towhee presentations lasted 20 min (versus 30 min for song sparrow presentations) or until the level of towhee response necessitated removal of the towhee mount to keep it intact for future presentations. I sampled each subject only once during the experiment to preclude the possibility that previous exposure to the net/speaker/decoy changed a female’s response. To
553
554
ANIMAL BEHAVIOUR, 59, 3
Table 2. Component loadings for principle components 1–3 following square-root transformation of the behaviours measured and equamax rotation of the components Behaviour
Factor 1
Factor 2
Factor 3
Approach Flights Closest approach Wing-wave threat display Buzzes Chitters and chets Growls Low chip (contact) calls High chip (alarm) calls Parental care Foraging Copulation solicitation
0.866 0.817 −0.670 0.756 0.663 0.129 0.482 0.031 0.630 −0.070 0.300 −0.090
0.166 0.295 −0.212 −0.205 −0.011 −0.183 −0.374 0.629 −0.017 0.777 0.653 0.078
0.025 0.252 0.116 0.287 0.439 0.758 0.452 −0.089 0.041 −0.069 0.027 0.836
control for day effects and distribute treatments equally across the study site, I matched female song sparrows for age and stage in the breeding cycle on noncontiguous territories and presented them with either a song sparrow or towhee simulated intrusion.
Analysis After dropping incomplete trials, 55 trials were analysed (32 female song sparrow presentations and 25 towhee presentations). Although song sparrow presentations lasted 30 min (in an effort to get hormone samples for another study: Elekonich 1997), I used only the first 20 min of the song sparrow trials. I counted each occurrence of each behaviour for all measures except closest approach, alarm calls, contact calls and parental care. I scored alarm calls, contact calls and parental care as 1 (occurred) or 0 (did not occur) for each minute of the trial (see Table 1). I analysed these data using SYSTAT for Windows (version 8.0). Data were square-root transformed to meet the assumptions for parametric tests, then subjected to principle components analysis with an equamax rotation. I further analysed the three factors (of 12) which had eigen values greater than 1 and at least three variables with component loadings greater than 3 (Wilkinson et al. 1996). Factor 1 accounted for 30.2% of the variance. Factors 2 and 3 accounted for 15.1 and 15.4% of the variance, respectively. Approaches, flights, growls, wing-wave threats, closest approach, buzzes and high chips (alarm calls) loaded most heavily onto factor 1 (Table 2). Although factor 1 could simply represent arousal, it may represent aggression. Wing-wave threat displays and growls are well known as the best indicators of escalated aggression for males and females (Nice 1937, 1943; Arcese 1989b; Elekonich 1997) and both loaded heavily onto this factor. Approaches, flights and closest approach are all indicators of increased movement, mostly towards the simulated intruder. The combination of approaches, flights, growls and wing-wave threat displays is observed during natural territorial encounters between females (Nice 1943; Arcese
1989b; Elekonich 1997) and is also common to male aggression (Nice 1937, 1943; Stoddard et al. 1991; Wingfield 1985). Because these presentations continued longer than territorial interactions normally do, the presence of alarm calls is not entirely surprising. Parental behaviours, foraging and low chips (contact calls) loaded heavily onto factor 2. Growls loaded negatively onto factor 2, suggesting that factor 2 is representative of a nonaggressive state where the subject’s focus is on something other than the simulated intruder (Table 2). Buzzes, copulation solicitations, chitters and growls all loaded heavily onto factor 3. Except growls, all of these vocalizations are involved in either inciting male territorial aggression or courtship (Elekonich 1997, 1998) suggesting that factor 3 represents behaviour directed towards the male mate during the context of a simulated female intrusion. I performed t tests using the factor scores to compare female responses to song sparrow versus towhee simulated intrusions within each season. Because the groups typically had unequal variances, I used the separate variance t value. I used a Dunn–Sidak correction to control for the effects of multiple comparisons. I evaluated changes in song sparrow females’ responses to simulated female song sparrow intrusion across the seasons with analysis of variance (ANOVA) followed by a post-hoc Tukey’s HSD test for each factor.
RESULTS
Song Sparrow versus Spotted Towhee Simulated Intrusions within Each Season The females’ responses represented by factor 1 were significantly higher to simulated intrusions by female song sparrows than to those of towhees in each of the three seasons, prebreeding, breeding and postmoult (Table 3). Females responses to simulated intrusion by female song sparrows and towhees did not differ significantly during any season for factor 2. Females responses represented by factor 3 were significantly higher to female song sparrow intrusion than to towhee intrusion in the postmoult season, but did not differ significantly in either the prebreeding or the breeding seasons (Table 3).
Seasonality of Female Song Sparrow Response to Simulated Female Intrusion Factor 1 varied significantly across the three seasons (F2,34 =3.56, P=0.039). Female aggressive responses to simulated female song sparrow intrusion were significantly higher in the prebreeding (Tukey’s HSD: df=34, P=0.040) season. Factor 2 scores were higher during the breeding season but did not differ significantly across seasons (F2,34 =2.78, P=0.076). Factor 3 scores were highest in the prebreeding season and dropped across the three seasons but these seasonal changes were not significant (F2,34 =1.57, P=0.224).
ELEKONICH: FEMALE RESPONSE TO INTRUSION
Table 3. Female responses to simulated female song sparrow (S) and spotted towhee (T) intrusions within each season Principle component Season
Factor 1
Factor 2
Factor 3
Prebreeding t N df P S>T?
4.12 13.1 10,6 0.004 >
1.07 13.5 10,6 0.66 =
0.40 12 10,6 0.97 =
Breeding t N df P S>T?
3.82 24.8 19,10 0.002 >
1.26 26.9 19,10 0.52 =
0.118 19.5 19,10 0.99 =
Postmoult t N df P S>T?
3.86 9.8 8,6 0.01 >
2.52 12 8,6 0.08 =
2.94 11.8 8,6 0.04 =
DISCUSSION Female song sparrows presented with a simulated intrusion by a female conspecific either responded to the intruder, signalled their mates or did not respond to the intruder, choosing instead to forage or care for young. Principle component 1 indicated that female song sparrows overall directed more behaviours (aggressive growls and wing-wave threat displays) towards an intruder than towards their mates in the prebreeding season than in the breeding or postbreeding seasons. Male song sparrow aggression is similarly highest during the prebreeding season, then declines during the breeding season, and increases again in early autumn (postmoult) when firstyear birds attempt to gain territories (Arcese 1987, 1989a, c; Beecher et al. 1994; Wingfield & Hahn 1994). Within each season females responded more strongly and aggressively to a simulated intrusion by a conspecific than a heterospecific as indicated by factor 1. These data suggest that female aggression, like male aggression, functions to defend access to resources on the territory and is directed specifically to conspecifics and not generalized to all birds using similar resources. Alternatively, females might react to heterospecifics that compete for resources, but females in these populations did not consider towhees resource competitors, or towhees are resource competitors but the towhee model and vocalizations were not adequate representations of a towhee. The former is more likely, as the tape/mount were so effective at attracting towhees that on a number of occasions I had to remove the mount to keep it intact for future trials. Arcese (1989a) suggested that male song sparrows defend access to females rather than resources on the territory. Similarly females may respond less to towhee intrusion than to song sparrow intrusion because they are
defending something other than the territory’s resources. Females could defend access to the male and his contributions of defence of the nest and territory and feeding of the nestlings and fledglings. Alternatively, female aggression could defend against intraspecific brood parasitism (e.g. Gowaty & Wagner 1988). However, intraspecific brood parasitism has not been observed in song sparrows (Keller 1996). Arcese (1989b) also observed a decrease in female aggression at the start of the breeding season, which he suggested was due to time constraints arising from the greater reproductive effort of females. Female song sparrow parental investment consists of building the nest, laying and incubating the eggs, and defending, brooding and feeding the young. In contrast, male song sparrows’ contribution includes only defending the nest and feeding the young (Nice 1943). Once a female begins to build a nest, any time she spends defending the territory comes at the cost of reproduction or self-maintenance. When Arcese (1989b) artificially increased the amount of food on territories, females spent more time guarding the territory and chasing floaters (birds without territories) than females whose territories did not have supplemental food. These data suggest that females are indeed defending resources on a territory and that a trade-off between territorial and reproductive behaviours exists for females. The lack of increase in female aggression following the moult (in autumn) when juveniles are attempting to settle onto territories (e.g. Beecher et al. 1994) suggests that females may not benefit from autumn territoriality. However, the within-season analysis of factor 3 suggests that in autumn females may be signalling their mates to defend the territory for them. Furthermore, female floaters are rare in these populations (Elekonich 1997). Thus, there may be less selective pressure on females to set up territories in their first year, because any female that survives the winter will breed. Alternatively, the resource females defend may be more seasonal, or vary more within a season. Dickinson & Lein (1987) found that nest site selection by female red-winged blackbirds, Agelaius phoeniceus, determines territory boundaries. Similarly, in several cases the site of a female song sparrow’s nest caused her mate to fight the neighbouring male in order to increase his territory to include the nest (Elekonich 1997). However, if the structural features of territories that determine the location of good nest sites often change between autumn and spring, females may not benefit by defending a territory boundary in autumn. Nonaggressive responses to simulated intrusions were most likely directed to mates or young as suggested by principle components 2 and 3. The responses represented by factor 2 consist of both self-maintenance and care of nestlings. Thus, it is not surprising that factor 2 scores peaked in the breeding season (although not significantly) when nutritional needs for producing eggs and/or for feeding young were highest. Buzzes, copulation solicitations, chitters and growls all loaded onto factor 3, which appears to represent signals directed towards the female’s mate. Buzzes are most often
555
556
ANIMAL BEHAVIOUR, 59, 3
used during copulation solicitations (Elekonich 1997, 1998). Although sometimes used during territorial interactions, females most often use chitters to incite male–male aggression, during copulation solicitation and when leaving the nest (Nice 1943; Wingfield & Hahn 1994; Elekonich 1997, 1998). Given the role of inciting in courtship and pair bond maintenance (Lorenz 1958), it is not surprising that chitters loaded heavily onto principle component 3, rather than principle component 1, which appears to represent signals directed towards female intruders. Especially in the early spring, natural male– male and female–female territorial interactions often occur (Nice 1937, 1943; Arcese 1989b; Elekonich 1997). During these interactions, females may use the vocalizations represented by factor 3 to call their mates to help defend the territory. In the present study, when males did respond to their mate’s chitters during simulated female song sparrow intrusions, they often perched nearby and watched rather than performing any aggressive displays. Although females’ responses reflected by factor 3 were higher to song sparrows than towhees in every season, the difference was only significant in the autumn, and growls loaded less heavily onto factor 3 than the other vocalizations. Furthermore, females gave more growls per minute of playback in the spring. Thus, the loading of growls onto factor 3 may reflect the natural situation of pair defence of boundaries, with increased female territoriality in spring as reflected by factor 1 and increased defence by males only in autumn. In summary, female song sparrows appear to defend territorial resources. But investigation of nonaggressive behaviours suggests that there are trade-offs for females between territorial behaviour and reproduction. These trade-offs may select for females that require fewer interactions to establish territorial boundaries and conclude aggressive interactions more quickly than males. The results of this study suggest that sex differences may exist for each parameter (e.g. seasonality and context) of behaviour. Furthermore, the inclusion of nonaggressive behaviours and the presentation of different types of stimuli provide a more complete picture of the natural contexts of aggressive behaviour in song sparrows. Acknowledgments I thank John Garrett, manager of the Skagit Wildlife Area, the Washington Department of Wildlife, the Seattle Parks Department and staff of Discovery Park for hosting the field work; Liz Campbell, Cully Nordby, Chris Hill, Sally Boeve, Jennie Maughn, Tami Nelson, Chris Robertson and Sean Schmidt for assistance in the field; Jim Ha and Sam Beshers for statistical advice, and Mike Beecher, Melissa Fleming, Sean O’Donnell, Philip Stoddard, Judy Stamps and two anonymous referees for comments. This research was supported in part by grants from Sigma Xi and the American Psychological Association. The research presented here was described in Animal Research Protocol No. 2207-02, approved on 22 June 1993 by the Institutional Animal Care and Use Committee of the University of Washington.
References Arcese, P. 1987. Age, intrusion pressure and defence against floaters by territorial male song sparrows. Animal Behaviour, 35, 773–784. Arcese, P. 1989a. Territory acquisition and loss in male song sparrows. Animal Behaviour, 37, 45–55. Arcese, P. 1989b. Intrasexual competition and the mating system in primarily monogamous birds: the case of the song sparrow. Animal Behaviour, 38, 96–111. Arcese, P. 1989c. Intrasexual competition, mating system and natal dispersal in song sparrows. Animal Behaviour, 38, 958– 979. Arcese, P., Stoddard, P. K. & Hiebert, S. M. 1988. The form and function of song in female song sparrows. Condor, 90, 44–50. Archer, J. 1988. The Behavioural Biology of Aggression. Cambridge: Cambridge University Press. Beecher, M. D., Campbell, S. E. & Stoddard, P. K. 1994. Correlation of song learning and territory establishment strategies in the song sparrow. Proceedings of the National Academy of Sciences U.S.A., 91, 1450–1454. Beletsky, L. D. 1982. Vocalizations of female northern orioles. Condor, 84, 445–447. Dickinson, T. E. & Lein, M. R. 1987. Territory dynamics and patterns of female recruitment in red-winged blackbirds (Agelaius phoeniceus). Canadian Journal of Zoology, 65, 465–471. Elekonich, M. M. 1997. Female–female territorial aggression and its hormonal control in the song sparrow. Ph.D. thesis, University of Washington. Elekonich, M. M. 1998. Song sparrow males use female-typical vocalizations in fall. Condor, 100, 145–148. Gowaty, P. & Wagner, S. 1988. Breeding season aggression of male and female eastern bluebirds to models of potential conspecific and interspecific egg dumpers. Ethology, 78, 238–250. Holmes, R. T., Sherry, T. W. & Reitsma, L. 1989. Population structure, territoriality and overwinter survival of two migrant warbler species in Jamaica. Condor, 91, 545–561. Hrdy, S. B. 1981. The Woman That Never Evolved. Cambridge, Massachusetts: Harvard University Press. Hrdy, S. B. & Williams, G. C. 1983. Behavioral biology and the double standard. In: Social Behavior of Female Vertebrates (Ed. by S. K. Wasser), pp. 1–18. New York: Academic Press. Hurly, T. A. & Robertson, R. J. 1984. Aggressive and territorial behaviour in female red-winged blackbirds. Canadian Journal of Zoology, 62, 148–153. Jenni, D. A. & Collier, G. 1972. Polyandry in the American jacana (Jacana spinosa). Auk, 89, 743–765. Keller, L. 1996. Inbreeding in an insular population of song sparrows. Ph.D. thesis, University of Wisconsin, Madison. Kroodsma, D. E. 1989a. Suggested experimental designs for song playbacks. Animal Behaviour, 37, 600–609. Kroodsma, D. E. 1989b. Inappropriate experimental designs impede progress in bioacoustic research: a reply. Animal Behaviour, 38, 717–719. Kroodsma, D. E. 1990. Using appropriate experimental designs for intended hypotheses in ‘song’ playbacks, with expamples for testing effects of song repertoire sizes. Animal Behaviour, 40, 1138–1150. Levin, R. 1996a. Song behaviour and reproductive strategies in a duetting wren, Thryothorus nigricapillus: 1. Removal experiments. Animal Behaviour, 52, 1093–1106. Levin, R. 1996b. Song behaviour and reproductive strategies in a duetting wren, Thryothorus nigricapillus: II. Playback experiments. Animal Behaviour, 52, 1107–1117. Lorenz, K. Z. 1958. The evolution of behavior. Scientific American, 199, 67–83. Marler, P. 1955. Characteristics of some animal calls. Nature, 176, 6–8.
ELEKONICH: FEMALE RESPONSE TO INTRUSION
Montgomerie, R. D. & Weatherhead, P. J. 1988. Risks and rewards of nest defence in birds. Quarterly Review of Biololgy, 63, 167–187. Nice, M. M. 1937. Studies in the Life History of the Song Sparrow. I. A Population Study of the Song Sparrow and Other Passerines. New York: Dover. Nice, M. M. 1943. Studies in the Life History of the Song Sparrow. II. The Behaviour of the Song Sparrow and Other Passerines. New York: Dover. Richison, G. 1983. The function of singing in female black-headed grosbeaks (Pheucticus melancephalus). Auk, 100, 105–116. Richison, G. 1986. The singing behavior of female northern cardinals. Condor, 88, 156–159. Sandell, M. I. & Smith, H. G. 1997. Female aggression in the European starling during the breeding season. Animal Behaviour, 53, 13–23. Slagsvold, T. 1993. Female–female aggression and monogamy in great tits Parus major. Ornis Scandinavica, 22, 319–326. Stoddard, P. K., Beecher, M. D., Horning, C. L. & Campbell, S. E. 1991. Recognition of individual neighbors by song in the song sparrow, a species with song repertoires. Behavioral Ecology and Sociobiology, 29, 211–215. Trivers, R. L. 1972. Parental investment and sexual selection. In: Sexual Selection and the Descent of Man (Ed. by B. Campbell), pp. 139–179. New York: Aldine.
Wiklund, C. G. & Village, A. 1992. Sexual and seasonal variation in territorial behaviour of kestrels, Falco tinnunculus. Animal Behaviour, 43, 823–830. Wilkinson, L., Blank, G. & Gruber, C. 1996. Desktop Data Analysis with SYSTAT. Upper Saddle River, New Jersey: Prentice Hall. Wingfield, J. C. 1985. Short-term changes in plasma levels of hormones during establishment and defence of a breeding territory in male song sparrows, Melospiza melodia. Hormones and Behavior, 19, 174–187. Wingfield, J. C. & Hahn, T. P. 1994. Testosterone and territorial behaviour in sedentary and migratory sparrows. Animal Behaviour, 47, 77–89. Wingfield, J. C. & Wada, M. 1989. Male–male interactions increase both luteinizing hormone and testosterone in the song sparrow, Melospiza melodia: specificity, time course and possible neural pathways. Journal of Comparative Physiology A, 166, 189–194. Wingfield, J. C., Ball, G. F., Dufty, A. M., Jr, Hegner, R. E. & Ramenofsky, M. 1987. Testosterone and aggression in birds: tests of the ‘challenge hypothesis’. American Scientist, 75, 602–608. Wittenberger, J. F. & Tilson, R. L. 1980. The evolution of monogamy: hypotheses and evidence. Annual Review of Ecology and Systematics, 11, 197–232. Yasukawa, K. & Searcy, W. A. 1982. Aggression in female redwinged blackbirds: a strategy to ensure male parental investment. Behavioral Ecology and Sociobiology, 11, 13–17.
557