| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavioral Ecology Vol. 15 No. 2: 219-222
Behavioral Ecology vol. 15 no. 2 © International Society for Behavioral Ecology 2004; all rights reserved
Replacement female house sparrows regularly commit infanticide: gaining time or signaling status?
Departamento de Ecología Evolutiva, Museo Nacional de Ciencias Naturales, C.S.I.C., José Gutiérrez Abascal 2, 28006 Madrid, Spain
Address correspondence to J. P. Veiga. E-mail: jpveiga{at}mncn.csic.es.
Received 4 March 2002; accepted 12 March 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although the killing of unrelated young (usually designed as infanticide) has been typically considered a male behavior, recent research has shown that females may commit infanticide even more frequently than do males. In rodents and primates, female infanticide represents a strategy associated to competition for resources or infant exploitation, but little is known about the causes and reproductive consequences of the killing of conspecifics by females in other vertebrates. In the present article, I focus on infanticide committed by females that replace mates of territorial males in a population of the house sparrow. I show that (1) replacement females regularly committed infanticide, (2) experienced females committed infanticide more frequently than did novel females and tended to select polygynous males to take over their nests, and (3) laying date and reproductive success after a territory takeover did not differ between infanticidal and noninfanticidal females. These results seem to indicate that infanticide has not evolved in females because of the short-term reproductive benefits it accrues to the perpetrator. I suggest that the killing of unrelated young by females relates to dominance status among potential female breeders and that this behavior benefits the perpetrators in terms of mate selection.
Key words: female-female competition, female infanticide, female replacement, house sparrows, sexual selection.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The killing of unrelated young (usually designed as infanticide) has been typically considered a male behavior. Recent research has shown, however, that females may commit infanticide even more frequently than do males. However, although male infanticide represents a sexual strategy (Hausfater and Hrdy, 1984
; Hrdy, 1979; Packer and Pusey, 1983; Sherman, 1981; van Schaik, 2000), the most common proximate causes of infanticide in females are the competition for breeding resources and the infant exploitation (Blumstein, 2000; Digby, 2000). In the first instance, by killing the dependent young of other females living either in the same group or in adjacent territories, infanticidal females gain direct access to the limited resources (Digby, 2000). Infant exploitation refers to cannibalism, in which the infanticidal females benefit from using the victim as a food item (Hrdy, 1979; Sherman, 1981). Only for some birds, it has been suggested that the infanticidal behavior in females may have been sexually selected: in great reed warblers (Hansson et al., 1997; Hasselquist, 1998) and house sparrows (Veiga, 1990b), females mated with polygynous males kill the dependent young of other females in the mating unit to reallocate male parental care toward the infanticidal female's nestlings. Infanticidal females would also benefit if they could start a breeding attempt without having to wait until nestlings in the replacement nest have fledged. Although this time advantage hypothesis has been put forward for males in both mammals and birds (van Schaik, 2000; Veiga, 2000), it may also apply to females, especially in multibrooded species such as the house sparrow and when the fledgling production decreases abruptly with season. In all these cases, infanticide provides immediate benefits to killers, so one would expect that infanticidal replacement females should improve their breeding success with respect to noninfanticidal females.
In mammals with multifemale groups and cooperative breeding groups, female infanticide, although ultimately owing to resource competition, seems to be primary related to the increase or maintenance of social status (Digby, 2000). In the present article, I suggest that the infanticidal behavior may function as an indicator trait to other females or to males chosen as mates. Intruding avian females that commit infanticide may show her dominant status with respect to victimized resident females and so reduce the risk of direct fighting for nest possession. Also, female killers may be evaluated as good-quality partners by victimized males so that they would be more easily accepted into a male's territory than noninfanticidal lower-quality females. Thus, the infanticidal behavior would have evolved in females by sexual selection mechanisms similar to that proposed for honest signaling traits of males (for review, see Andersson, 1992).
In the present article, I investigate female infanticide during a 10-year study of a house sparrow population breeding in a nesting colony in which infanticide by both males and females has been previously reported. Although male infanticide is usually committed by intruding males at nests of widowed females (Veiga, 1990a), female infanticide has been recorded mostly in a context of female-female competition by the parental investment of polygynous males (Veiga, 1990b). Intrasexual competition for limited nesting resources is strong in both males and females in this population, and agonistic interactions are commonly observed throughout the entire breeding season in both sexes. Here, I focus on infanticide committed by females that replace mates of territorial males. I address the following questions: (1) whether replacement females regularly commit infanticide, (2) whether experienced and, presumably, more competent females commit infanticide more frequently than do novel females, (3) whether infanticidal females select specific males, and (4) whether infanticidal females gained a time advantage to start a subsequent breeding attempt and/or they had greater annual reproductive success than did replacement noninfanticidal females.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
During 10 consecutive years (1986–1995), I studied a house sparrow colony made up of 60–80 pairs nesting in nest-boxes at Collado Villalba, Spain. Males are predominantly monogamous, and most individuals undertake two or three reproductive attempts per season. For this study, I selected males that carried out two or more breeding attempts per season. Cases of infanticide attributable to interference between polygynously mated females have not been considered because this type of infanticide is not associated to mate replacement (Check and Robertson 1991
; Hansson et al., 1997; Veiga, 1990b).
I routinely inspected nests during laying, incubation, and nestling periods. During the nestling period, I checked nests between three and five times. Each time I observed a nestling with any suspicious sign of having suffered an infanticide attempt, I checked the nest daily during the following 1 or 2 days to obtain additional evidence of infanticide. In some cases, I made direct observations from a blind to identify the eventual infanticide. I considered that infanticide occurred when I detected nestlings with pecking marks, hematomas, or featherless backs and crowns, wounds typically produced by conspecific adults (Veiga, 1990a). In three cases, infanticide was directly witnessed by me, and in one additional case it was suspected because nestlings were missing after the continuous presence of an intruder at the nest was recorded. For this study, I considered only nests in which the identity of both initial partners could be unambiguously established because they were color-ringed (see below). Cases of simultaneous replacement of both pair members have been excluded from the present analysis. The occurrence or absence of female replacement and infanticide could be determined in 250 nests. In most instances the sex that committed the infanticide could not be directly ascertained. A logistic regression of the presence or absence of infanticide in relation to the occurrence of male replacement or female replacement indicated that either male or female replacement closely associated with infanticide (female replacement, ß = 1.79, p =.002; male replacement, ß = 1.18, p =.0003). Then, I assume that when infanticide associates to replacement of one of the partners, the killing was committed by the replacing sex. Noninfanticidal females usually replaced female owners after a breeding failure of the previous owner or after her brood had fledged. In one case the replacing females laid eggs when nestlings were near to fledge.
From 70–90%, depending on the year, of all females that started a breeding attempt had been captured with mist-nets and color-ringed before the start of the breeding season (from February–March) or in previous breeding seasons. As laying season starts in mid April–May, most replacements occurred during May–June. In consequence, females that were already ringed when replacement occurred had some previous mating and breeding experience; I assumed unringed females were novel ones. However, not all replacing females that had been ringed in February–March had been recorded as breeders before they took over a nest. These females were not included in the sample of females of known breeding experience. A similar age-assignment procedure had been previously validated comparing morphological traits of individuals of estimated age with other of known age (Veiga, 1993). In any case, uncertainties about female reproductive experience will increase Type II error, making the results more conservative.
Breeders were identified by direct observation of nests in each reproductive attempt. Observations were mainly conducted shortly before the start of laying and during the nestling period. When nestlings were 12 days old, I measured and ringed them. As chicks usually left the nest at 13–15 days, I considered that the number of nestlings that were alive in this last visit fledged successfully. Relative laying date is defined as the number of days elapsed from the start of the previous clutch in the replacement nest until the initiation of the clutch produced by the replacement female. The absolute laying date is the number of days elapsed after the first clutch recorded in the current breeding season.
I used logistic stepwise regression based on the likelihood ratio test (Sokal and Rohlf, 1987) to analyze the effect of independent variables on a dependent variable with two possible values. The variation of relative laying date and fledgling production in relation to replacement or infanticide was analyzed with ANOVA models. Absolute laying date is included as a covariate to control for the variability of fledgling production with time of season (see Figure 2). Although actual values are shown in figures, the number of fledglings was standardized each year to have zero mean and 1 SD in statistical analyses.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Female replacement occurred in 106 of 250 attempts in which the presence or absence of female replacement could be unambiguously established. Female infanticide occurred in 22 of these replacements, whereas passive replacement (i.e., without infanticide) occurred in 84 of the attempts. The killing of all brood members occurred in 16 out of these 22 cases. Thus, infanticide by females occurred in 20.75% of the nests in which female replacement occurred and in 8.8% of all attempts. However, because of the method used to ascertain the presence of infanticide (see Methods), its frequency may have been underestimated. For example, some replacement females could commit infanticide at the incubation stage, when it is not possible to discriminate between infanticide and other sources of egg losses (nest desertion, predation, etc.). Also it possible that a few replacement females committing infanticide did not bred during the rest of the breeding season, so they could not be detected.
Replacements by females of known breeding experience occurred in 79 instances, so that the analysis of the frequency of infanticide in relation to male polygynous status and female experience was restricted to this partial sample. A logistic stepwise regression model for the occurrence of female infanticide versus passive replacement in relation to female experience and male mating status indicated that experienced females committed infanticide more frequently than did novel females (-2Log LR = 7.1, p =.007) (Figure 1). Second, polygynous males tended to suffer infanticide more frequently than did monogamous males, although the tendency was only marginally significant (-2Log LR = 3.05, p =.08). However, a simple comparison between both types of males showed a significant effect (
, p =.015). Nonetheless, the interaction between female experience and male status was significant (female–experience x male–status: -2Log LR = 6.2, p =.013), indicating that the experienced females were the ones that committed infanticide preferentially at the nests of polygynous males (Figure 1). The occurrence of passive female replacement did not depend on male status: replacement took place in 34.8% (n = 198) of the nests of monogamous males and in 28.8% (n = 52) in the case of polygynous males (
, p = 0.41).
|
The relative laying date after replacement did not differ between infanticidal and noninfanticidal females after controlling for absolute laying date in the nest where they replaced (infanticidals: 33.47 days ± 15.12 SD; noninfanticidals: 37.33 days ± 11.56 SD; ANOVA: F1,67 = 0.17, p =.68). Thus, female killers gained a similar small time advantage with respect to noninfanticidal replacement females. The fledgling production after replacement did not differ between infanticidal and noninfanticidal females (ANCOVA: infanticide effect, F1,64 = 0,17, p =.68; date effect, F1,64 = 0.76, p =.39; interaction, F1,64 = 0.46, p =.5) (Figure 2).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Female house sparrows regularly committed infanticide when taking over a nest and replacing the previous female owner. Experienced females committed infanticide, as an alternative strategy to passive replacement, more frequently than do novel females. If previous breeding experience is a correlate of intrasexual dominance in the competition for breeding resources, as it has frequently recorded in male birds (Beletsky et al., 1995
; Hegner and Wingfield, 1987; Rohwer and Rohwer, 1978; Searcy and Yasukawa, 1995; Veiga et al., 2001), the infanticidal behavior would be underlying the presence of a hierarchy among females that live in the same nesting colony. A similar scenario has been depicted for social rodents, primates, and carnivores, in which dominant females kill the young of other females more frequently than subordinates do (Andrews, 1998; Clutton-Brock et al., 1998; Digby, 2000; Muroyama and Thierry, 1996; Wasser and Starling, 1986). However, although in mammals the food resources are the target of infanticidal females, in the house sparrow, females apparently might be searching specific males or some male characteristics. Increased paternal care is not an obvious advantage of this strategy because even becoming primary female, the infanticidal would receive similar aid than if she would have mated with a monogamous male, and she would assume the risk of suffering infanticide by the secondary female (Veiga, 1990b). A greater protection of the offspring is probably neither an advantage of choosing a polygynous male because, as shown in this study, such males are more frequently targets of infanticide than monogamous ones. Polygynous males could be selected, as it has been demonstrated in other bird species, because their genetic superiority or because they defended better breeding resources (Andersson, 1992; Kempenaers et al., 1992; Pleszczynska, 1978).
Whatever benefit an infanticidal female is securing by taking over an active nest, the pertinent question is why they kill young instead of waiting until they have left the nest. The most obvious response would be that by killing young, females immediately can start a breeding attempt, thus gaining time that may be crucial for breeding success. However, my results indicate that infanticidal females gained only 4 days, in average, with respect to noninfanticidal females, a time period that did not translate into differential reproductive success. This small time difference resulted from the fact that infanticide usually took place with mid-grown or older nestling. Why females did not kill younger chicks more frequently is unclear, although it could be related to the occurrence of displaying maneuvers of males in the last part of the nestling period (Veiga, 1992).
Another possibility to explain why females commit infanticide is that killers were subordinate females that tried to provoke a perception of high risk in the victim that would induce her to search for safer sites (Sherman, 1981). This possibility involves a surreptitious behavior that I have not observed in the infanticidal female house sparrows. On the contrary, in several infanticidal acts I witnessed, the female attacked the nestlings in presence of both parents, showing brave and dominant behavior. Thus, it seems more plausible that an infanticidal act may serve as a status signal directed to competitors, which may preclude more direct and risky agonistic interactions to establish dominance relationships. Supporting this, it has been shown in some primates that in the rare occasions in which a subordinate female commits infanticide or after the death of the dominant female, the perpetrator rose to a top-ranking position (Andrews, 1998; Roda and Mendes Pontes, 1998).
The infanticidal act may also function as an indicator to males of the experience of a potential mate. A male could prefer an infanticidal high-quality female even after she killed his young because mating with her has a superior reproductive reward than maintaining a lower-quality female. Interestingly, I witnessed two occasions in which a male was reluctant to defend his brood from the infanticidal female. On the other hand, as infanticidal females did not raise more young than did other females, the argued male preference for the infanticidal female implies that either her offspring survived better than those of the noninfanticidal females, or that the newly formed pair should start additional reproductive attempts that compensate the male for the previous losses caused by the killer. This aspect, however, could no be addressed in the current study and deserves a more specific research to be adequately tested.
Even though infanticide does not render short-term reproductive benefits to killers, it cannot be discarded that this behavior had more subtle effects on fitness. For example, if infanticidal females acquire polygynous more attractive mates, then the primary consequences of being infanticidal would be the production of sons that inherit the attractiveness of their father (see Weatherhead and Robertson 1979). This aspect, however, needs further research before a robust conclusion can be drawn. On the other hand, it is possible that although infanticidal females had to commit infanticide in order to get a breeding opportunity, noninfanticidal females managed to replace without displaying aggressiveness toward nestlings. The proper test of this possibility should be a comparison between the fitness of a female committing infanticide and the same female passively replacing. Such a comparison is extremely difficult to conduct in a free living population. However, the fact that infanticidal females are more experienced than are noninfanticidal ones does not lend much support to this possibility.
In conclusion, female house sparrows regularly killed the young of the female owners they replaced. Infanticide was committed more frequently by experienced females than by novel females, the former preferentially taking over the nests of polygynous males. This seems to indicate that there is a female preference for specific males, and that the killing of unrelated young by females may represent a behavior that relates to the status of the potential female breeders.
|
ACKNOWLEDGEMENTS |
|---|
I thank three anonymous referees for their constructive comments. During data processing and manuscript writing, I was supported by projects PB97-1249 (DGICYT) and BOS2001-0703 (MCYT).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Andersson M, 1992. Sexual selection. Princeton, New Jersey: Princeton University Press.
Andrews J, 1998. Infanticide by a female black lemur, Eulemur macaco, in disturbed habitat on Nosy Be, North-Western Madagascar. Folia Primatol 69:14-17.
Beletsky LD, Gori DF, Freeman S, Wingfield JC, 1995. Testosterone and polygyny in birds. In: Current ornithology, vol. 12 (Power DM, ed). New York: Plenum Press; 1–4.
Blumstein DT, 2000. The evolution of infanticide in rodents: a comparative analysis. In: Infanticide by males and its implications (van Schaik CP, Janson CH, eds). Cambridge: Cambridge University Press; 178–197.
Clutton-Brock TH, Brotherton NM, Smith R, McIlrath GM, Kansky R, Gaynor D, O'Riain MJ, Skinner JD, 1998. Infanticide and expulsion of females in a cooperative mammal. Proc R Soc Lond B 265:2291-2295.[Medline]
Check AA, Robertson RJ, 1991. Infanticide in female tree swallows: a role for sexual selection. Condor 93:454-457.
Digby L, 2000. Infanticide by female mammals: implications for the evolution of social systems. In: Infanticide by males and its implications (van Schaik CP, Janson CH, eds). Cambridge: Cambridge University Press; 178–197.
Hansson B, Bensch S, Hasselquist D, 1997. Infanticide in great red warblers: secondary females destroy eggs of primary females. Anim Behav 54:297-304.[CrossRef][ISI][Medline]
Hasselquist D, 1998. Polygyny in great red warblers: a long-term study of factors contributing to male fitness. Ecology 79:2376-2390.[CrossRef][ISI]
Hausfater G, Hrdy SB, 1984. Infanticide: comparative and evolutionary perspectives. New York: Aldine.
Hegner RE, Wingfield JC, 1987. Social status and circulating levels of hormones in flocks of house sparrows. Ethology 76:1-14.
Hrdy SB, 1979. Infanticide among animals: a review, classification and examination of the implications for the reproductive strategies of females. Ethol Sociobiol 1:13-40.[CrossRef][ISI]
Kempenaers B, Verheyen M, Van den Broek M, Burke T, 1992. Extra-pair paternity results from female preference of high quality males in the blue tit. Nature 362:537-539.[CrossRef]
Muroyama Y, Thierry B, 1996. Fatal attack on an infant by an adult female Tonkean macaque. Int J Primatol 17:219-227.
Packer C, Pusey AE, 1983. Adaptations of female lions to infanticide by incoming males. Am Nat 121:716-728.[CrossRef][ISI]
Pleszczynska WK, 1978. Microgeographic prediction of polygyny in the lark bunting. Science 201:935-937.
Roda SA, Mendes Pontes AR, 1998. Polygyny and infanticide in common marmosets in a fragment of the Atlantic forest of Brazil. Folia Primatol 69:372-376.
Rohwer S, Rohwer FC, 1978. Status signalling in Harris' sparrows: experimental deceptions achieved. Anim Behav 26:1012-1022.[CrossRef]
Searcy WA, Yasukawa K, 1995. Polygyny and sexual selection in red-winged blackbirds. Princeton, New Jersey: Princeton University Press.
Sherman PW, 1981. Reproductive competition and infanticide in Belding's ground squirrels and other animals. In: Natural selection and social behavior (Alexander RD, Tinkle DW, eds). New York: Chiron Press; 311–331.
Sokal Rr, Rohlf FJ, 1987. Introduction to biostatistics. New York: W.H. Freeman.
van Schaik CP, 2000. Infanticide by male primates: the sexual selection hypothesis revisited. In: Infanticide by males and its implications (van Schaik CP, Janson CH, eds). Cambridge: Cambridge University Press; 27–60.
Veiga JP, 1990a. Infanticide by male and female house sparrows. Anim Behav 39:496-502.[CrossRef]
Veiga JP, 1990b. Sexual conflict in the house sparrow: interference between polygynously mated females versus asymmetric male investment. Behav Ecol Sociobiol 27:345-350.[CrossRef]
Veiga JP, 1992. Why are house sparrows predominantly monogamous: a test of hypotheses. Anim Behav 43:361-370.
Veiga JP, 1993. Badge size, phenotypic quality, and reproductive success in the house sparrow: a study on honest advertisement. Evolution 47:1161-1170.[CrossRef][ISI]
Veiga JP, 2000. Infanticide by male birds. In: Infanticide by males and its implications (van Schaik CP, Janson CH, eds). Cambridge: Cambridge University Press.
Veiga JP, Moreno J, Cordero PJ, Mínguez E, 2001. Territory size and polygyny in the spotless starling: resource holding potential or social inertia? Can J Zool 79:1951-1956.[CrossRef]
Wasser SK, Starling AK, 1986. Reproductive competition among female yellow baboons. In: Primate ontogeny, cognition and social behavior (Else JG, Lee PC, eds). Cambridge: Cambridge University Press; 343–354.
Weatherhead PJ, Robertson RJ, 1979. Offspring quality and the polygyny threshold:. "the sexy son hypothesis." Am Nat 113:201-208.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||