- Email: [email protected]
Offspring age and nest defence: test of the feedback hypothesis in the meadow pipit
ANIMAL BEHAVIOUR, 2001, 61, 297–303 doi:10.1006/anbe.2000.1574, available online at http://www.idealibrary.com on
Offspring age and nest defence: test of the feedback hypothesis in the meadow pipit VA u CLAV PAVEL & STANISLAV BURES {
Laboratory of Ornithology, Palacky´ University Olomouc (Received 23 September 1999; initial acceptance 3 November 1999; final acceptance 9 August 2000; MS. number: 6347R)
The feedback hypothesis has been proposed to explain variation in nest defence intensity in birds. In species in which the female builds the nest and incubates the eggs, this hypothesis predicts a higher level of nest defence initially for females, whereas males’ responses should increase when they start feeding nestlings. We studied changes in nest defence by both sexes during the nestling period in meadow pipits, Anthus pratensis. We placed a stuffed stoat, Mustela erminea, 5 m from a meadow pipit nest with nestlings aged either 2–4 or 7–12 days and recorded the nest defence behaviour of both parents for 10 min. Males came closer to the predator and mobbed more intensely for older nestlings whereas females defended the nest at a high intensity from the beginning of the nestling period. This finding agrees with the predictions of the feedback hypothesis. We also discuss possible functions of alarm calls and number of mobbing birds during nest defence.
will increase with nestling age as the nest becomes more conspicuous (Redondo & Carranza 1989; Onnebrink & Curio 1991). One might also hypothesize that the proximate stimulus that influences nest defence intensity in parent birds is feedback from the current contents of the nest. Based on that prediction McLean & Rhodes (1992) suggested the feedback hypothesis. They proposed that the variability in parental responses to predators is determined primarily by the interaction between parents and their nest and young. A complete nest without a clutch provides a static visual signal, which maintains parental activities at the nest, while a nest containing eggs provides stronger signals affecting incubation. As the breeding attempt progresses, the growing, moving and vocalizing chicks provide stronger signals, which rapidly reinforce a still higher level of parental responses (including nest defence). In species in which parents share parental activities unequally, the feedback hypothesis predicts that the sex that has more interactions with the nest should defend it more vigorously. This hypothesis also predicts that males and females will change the intensity of their nest defence differentially as their offspring grow. In some bird species (such as meadow pipits, Anthus pratensis) only the female builds the nest and incubates the eggs, while the male initially sporadically feeds her or the young nestlings (Halupka 1994). As the nestlings develop, their growing demands reinforce the male’s interactions with the nest (through increasing feeding activity). The
In many bird species parents increase the intensity of nest defence as their offspring get older (Anderson et al. 1980; Brunton 1990; Rytko ¨ nen et al. 1990). However, at the same time as the offspring grow, the breeding season progresses (Regelmann & Curio 1983), the opportunity to renest declines (Barash 1975), and in many studies (as a methodological artefact) the number of preceding visits by the experimenter to check the nest also increases (Knight & Temple 1986a). Only a few studies have separated these factors from one another, but the effect of offspring age was the most important in the majority of them (Redondo & Carranza 1989; Rytko ¨ nen et al. 1990). Furthermore male and female behaviour can change differentially as the offspring grow (Regelmann & Curio 1986; Martin & Horn 1993; Rytko ¨ nen et al. 1993). Several functional hypotheses have been proposed to account for the increase in nest defence intensity with offspring age. The reproductive value hypothesis predicts that parental investment in short-lived passerines increases during the breeding attempt as the current offspring’s chances of survival increase and the reproductive potential of the parent decreases (Maynard Smith 1977). This hypothesis explains the ultimate causation of parental behaviour, but what is the proximate factor that influences patterns of parental responses? The vulnerability hypothesis predicts that the risk of a predator detecting the nest (and hence also nest defence intensity) Correspondence: V. Pavel, Laboratory of Ornithology, Palacky´ University, trˇ. Svobody 26, 771 46 Olomouc, Czech Republic (email: [email protected]). 0003–3472/01/020297+07 $35.00/0
2001 The Association for the Study of Animal Behaviour
297
2001 The Association for the Study of Animal Behaviour
298
ANIMAL BEHAVIOUR, 61, 2
female’s feeding effort also increases but her interactions with the nest increase less than the male’s, because she has incubated the eggs and brooded the young nestlings (unpublished data; for the importance of males in feeding nestlings see, e.g. Hayes & Robertson 1989). In such species the hypothesis predicts a higher level of nest defence initially for the female than for the male, whereas the male should increase his defence when he starts feeding nestlings. The female’s responses should also increase as the nestlings grow, but less than the male’s, because she should vigorously defend a nest with young nestlings. We tested the prediction that the intensity of nest defence increases with the age of the offspring in the meadow pipit and that it increases more for males than for females. To control for the potentially confounding effect of renesting (when renesting is more costly for the female, she should defend the nest more intensively than the male at the young nestling stage if she still has an opportunity to renest) we did experiments in two areas: the Tydal area in Northern Europe (where there is time for only one breeding attempt) and the Jeseni´ky Mountains in Central Europe (where two breeding attempts are common). We evaluated the behaviour of parents defending their nest by exposing them to a predator. We assumed that the risk should increase rapidly as the parent gets closer or more accessible and vulnerable to the predator (Montgomerie & Weatherhead 1988).
METHODS The meadow pipit is a small monogamous passerine that breeds on the ground in open habitats. Only the female builds the nest and incubates the eggs (3–7 eggs, 13 days) while the male occasionally feeds her. Nestlings are cared for and fed initially mainly by the female; the male feeds them more intensively as they grow. Fledglings are fed by both parents for 1–2 weeks (Halupka 1994; unpublished data). We studied meadow pipit populations on alpine meadows 1400–1460 m above sea level in the Jeseni´ky Mountains (Czech Republic; 50N, 17E) in 1995–1998 and in an alpine ecosystem 730–950 m above sea level near Tydal, Sør Trondelag province (central Norway; 63N, 12E) in 1996 and 1998. Both study areas had similar vegetation cover and nest sites. The meadow pipit was the most abundant bird species with a density of ca. 1 breeding pair/ha, and the stoat, Mustela erminea, was one of the most important predators in both study areas (A. Moksnes, personal communication; unpublished data). We usually found the nests by flushing the incubating bird or observing a bird feeding nestlings. We determined nestling age from either the hatching day (day 0) or nestling weight (Coulson 1956). To determine the days of hatching and fledging, we checked the nests every 2–3 days. Because meadow pipits have no appreciable sex differences, we marked females with a white spot on both sides of the neck. We captured some females immediately after the nestlings hatched by covering both the nest and brooding female with a net. Others were caught while
Table 1. Ranked call and behavioural responses of the meadow pipit to the stuffed stoat Ranked response
Call responses 0 1
2 3 Behaviour 1 2 3 4 5
Description
No calling by parent Chirping ‘chutt’ or ‘chitt’ calls; used to maintain contact rather than to express anxiety Alarm calls near nest; a persistent ‘stitt-itt’ Distress and alarm calls near nest Parent watches the nest and stuffed stoat from a distance Parent walks or lands close to the nest and stuffed stoat and watches them Parent displays slight fluttering flight around the nest and stuffed stoat Parent hovers over the stuffed stoat Parent intensively dives or strikes at the stuffed stoat
feeding nestlings; we covered the nest with a cage, the trap door of which was released either by the approaching bird or by a remote control from a hide. We placed a stuffed stoat 5 m from the meadow pipit nest and recorded the defensive behaviour of the parents for 10 min from a hide 30–60 m away (we built the hide at least 1 h before the experiment to habituate the parents). To evaluate the influence of nestling age on nest defence intensity, we did two experiments at each nest: when the nestlings were young (2–4 days old) and old (7–12 days old). We did the experiments during periods of high feeding activity from 0600 to 1930 hours. A duration of 10 min corresponds with the feeding frequency of meadow pipits (5.8 times/h, Pedroli 1978; 10.0 times/h, unpublished data) and was sufficient to record parental nest defence activities. The procedure had no apparent adverse effects on the adults or nestlings. Parents started to feed the nestlings a few minutes after being caught and immediately after we removed the predator model. Nest defence varied from quietly watching from a distance to noisy contact dives or strikes at the stoat. Many studies have used the distance to the predator model, the latency to approach, whether the bird performs a risky display, calling rates or mobbing intensity to assess the risk to parents of defending their clutch or brood (e.g. Greig-Smith 1980; Regelmann & Curio 1983; Westneat 1989). After preliminary observations, we decided to use the following variables: (1) minimum distance, that is, the shortest distance between the bird and the stuffed stoat; (2) intensity of calling, that is, the calls of the parents ranked from 0 (no calling) to 3 (the most intensive alarm and distress calls; Table 1); (3) intensity of display, ranked from 1 (minimal activity) to 5 (maximal activity; cf. Hudson & Newborn 1990; Halupka & Halupka 1997; Table 1). We also recorded (4) the number of extrapair mobbers reflecting the number of mobbing or calling birds (except the parents) close to the stuffed predator (Halupka & Halupka 1997); and (5)
PAVEL & BURES { : NEST DEFENCE IN PIPITS
Table 2. Comparisons of nest defence activities of meadow pipit parents in the Jeseni´ky Mountains (J) and in the Tydal area (T) Nestlings 2–4 days old
Minimum distance Intensity of calling Intensity of display
Nestlings 7–12 days old
Both parents NJ =18 NT =23
Male NJ =11 NT =21
Female NJ =11 NT =21
Both parents NJ =18 NT =23
Male NJ =11 NT =21
Female NJ =11 NT =21
0.46 0.71 0.78
−1.11 0.50 1.19
1.07 0.08 0.04
0.07 0.80 0.95
−0.56 0.16 0.83
0.40 0.46 0.64
Z values from Mann–Whitney U tests are given. All P>0.2.
parent–nest distance (minimum distance between the defending parent and its nest) to control for their effects. During 1995–1998 we conducted 41 paired experiments (every nest was tested twice, when nestlings were young and old) on 18 nests in the Jeseni´ky Mountains and 23 nests in the Tydal area. Females were not marked until June 1996, therefore only 32 paired experiments (11 nests from the Jeseni´ky Mountains and 21 from the Tydal area) were available to compare the behaviour of males and females. The other nine paired experiments were included when we analysed the reactions of both parents. Because our results from previous studies of distraction displays (unpublished data) and from this study (Table 2) showed that nest defence intensity did not differ between the Jeseni´ky Mountains and the Tydal area (despite a marked difference in the opportunity to renest; see Introduction), we assumed that the opportunity to renest did not seriously influence sexual differences in defence behaviour, and we pooled data from the two study areas for the following analyses. We had two licences for this study: (1) from the Bird Ringing Centre Prague and the Ministry of Environment of the Czech Republic (for field research on birds in the National Park Pradeˇd in the Jeseniky Mountains). No special licence was necessary for study in the Tydal area. We used STATISTICA for Windows software (StatSoft 1998) for data analysis. For analyses of the riskdetermining variables we used nonparametric methods. To compare the behaviour of parents with young and old nestlings at the same nest, we computed Wilcoxon signed-ranks tests. To evaluate the relative relationships between single risk-determining variables and the number of extrapair mobbers we partialled out the effects of other factors. To do this we computed Pearson partial correlations between the compared variables, when data were rank ordered (Spearman partial rank order correlations). All tests were two tailed. RESULTS All risk-determining variables (minimum distance, intensity of calling and intensity of display) were significantly interdependent and related to the number of extrapair mobbers for both males and females and also for young and old nestlings, when single correlations were
computed. In all cases both intensity of calling and intensity of display decreased with increasing distance of the parent from the predator. Furthermore, minimum distance decreased and intensity of calling and intensity of display increased with increasing number of extrapair mobbers (Table 3). When the effects of all variables not in question were partialled out, only the interdependence of minimum distance and intensity of display was highly significant in all cases. Intensity of calling was significantly related only to intensity of display (in all cases except for males with old nestlings), but not to minimum distance (in all cases). Number of extrapair mobbers was marginally significantly correlated only with minimum distance of females with young nestlings and intensity of display of females with old nestlings. Only intensity of calling increased significantly with nestling age for both parents together. Although the directions of differences in minimum distance and intensity of display corresponded with increasing risk in old nestlings, the results were nonsignificant (Table 4). Furthermore, there were significantly more extrapair mobbers during experiments with old than with young nestlings (XSE=0.710.29 versus 2.000.49; paired t test: t40 = 2.45, P=0.019). The increase in risk taking with nestling age was more evident in males than in females (Fig. 1). Although intensity of calling was significantly stronger for old nestlings in both parents, minimum distance was significantly shorter for old nestlings only in males. An increase in the intensity of displays in males was nonsignificant (Fig. 1). Males decreased their minimum distance significantly more than females (Table 4). We found no differences in parent–nest distance between experiments with young and old nestlings for both parents together (XSE=7.411.58 versus 6.600.95; Wilcoxon signed-ranks test: Z=0.06, N=41, P=0.951) or for males (16.552.25 versus 14.361.84; Z=1.24, N=32, P=0.214) and females (7.011.96 versus 7.421.33; Z=1.04, N=32, P=0.300). Females were near the nest in all experiments, but we did not observe males in three experiments with young nestlings and two (the same nests as in the first case) with old nestlings (21 =0.22, P=0.641). As the males might have been present but hidden from our view, we recorded their responses as
299
300
ANIMAL BEHAVIOUR, 61, 2
Table 3. Spearman rank correlations between measures of meadow pipit nest defence behaviour (minimum distance to the predator, intensity of display and intensity of calling) and number of extrapair mobbers Male (N=32)
Nestlings 2–4 days old Distance–Calling Distance–Display Distance–Number Calling–Display Calling–Number Display–Number Nestlings 7–12 days old Distance–Calling Distance–Display Distance–Number Calling–Display Calling–Number Display–Number
Female (N=32)
rs
Partial r
rs
Partial r
−0.64*** −0.88*** −0.61*** 0.76*** 0.50** 0.62***
0.10 −0.73*** −0.18 0.52** 0.08 0.15
−0.58*** −0.80*** −0.58*** 0.64*** 0.46** 0.45**
−0.03 −0.68*** −0.40* 0.38* 0.22 −0.08
−0.66*** −0.90*** −0.73*** 0.67*** 0.56*** 0.77***
−0.16 −0.73*** −0.13 0.19 0.07 0.34
−0.49** −0.82*** −0.45** 0.73*** 0.65*** 0.67***
0.21 −0.77*** 0.17 0.48** 0.27 0.37*
Simple and partial correlation coefficients are shown. *P<0.05; **P<0.01; ***P<0.001. Table 4. Comparison of responses of both parents together and males and females separately to a stuffed stoat close to the nest when nestlings were young (2–4 days) and old (7–12 days) and of the change in response with nestling age for males and females
Both parents (N=41)
Male (N=32)
Female (N=32)
Change in response of males versus females (N=32)
1.93† 4.00** 1.52
2.32* 3.34** 1.86†
0.71 3.32** 0.58
2.03* 0.19 1.00
Young versus old nestlings
Minimum distance Intensity of calling Intensity of display
Z values from Wilcoxon signed-ranks tests are given. †P<0.1; *P<0.05; **P<0.001.
30 m for minimum distance and parent–nest distance, 0 for intensity of calling and 1 for intensity of display. These scores corresponded to the weakest recorded for observed males. DISCUSSION
Differences Between the Sexes Males and females differed in how the risk-determining variables minimum distance and intensity of display changed with nestling age (Table 4, Fig. 1). Males, but not females, changed their behaviour from watching the nest and predator from a distance when the nestlings were 2–4 days old to direct attacks on the predator when they were 7–12 days old. This finding is in accordance with predictions of the feedback hypothesis (McLean & Rhodes 1992). Because the female meadow pipit builds the nest and incubates the eggs her level of interaction with the nest is high before hatching, and she should defend it relatively intensively just after hatching. The male has few interactions with the nest before hatching, but feeds
the nestlings regularly; as his feeding effort increases, so should his nest defence activities. Similar results have been obtained with other species. Breitwisch (1988) observed that males, but not females, of the northern mockingbird, Mimus polyglottos, approached intruders more closely as the nestlings grew. Rytko ¨ nen et al. (1995) found that the nest defence intensity of female willow tits, Parus montanus, was related to their provisioning effort; although in males this relationship was nonsignificant, those males that visited the nest most frequently (but with the smallest load size and hence lower provisioning effort) were the most aggressive nest defenders. Such findings are consistent with our results and could be explained proximately by the feedback hypothesis as well as ultimately by the reproductive value hypothesis. Our results may also be affected by the occurrence of extrapair paternity in our populations. If males have any doubt about their paternity, their level of parental investment should be lower than that of females. If the probability of extrapair paternity changes during the breeding season, their parental investment should
Minimum distance (m)
PAVEL & BURES { : NEST DEFENCE IN PIPITS
30 25 20 15 10 5 0
Intensity of calling
3
2
1
0
Intensity of display
5 4 3 2 1
Young
Old Male
Young
Old Female
Figure 1. Minimum distance (m) from stuffed stoat, intensity of calling and intensity of display (ranked responses, see Table 1) of males and females with young and old nestlings. Bars indicate median values, error bars indicate interquartile range (all N=32).
change correspondingly. Although this prediction was not confirmed when confidence of paternity was experimentally decreased (e.g. in males of the eastern bluebird, Sialia sialis, MacDougall-Shackleton & Robertson 1998) it can add to the variation in male defence behaviour. However, as far as we know the level of extrapair paternity, if any, for meadow pipits has not been measured.
Do Alarm Calls Reflect Parental Risk? The strong increase in the intensity of calling with nestling age for both sexes corresponds with the results of other studies (Greig-Smith 1980; East 1981; Weatherhead 1989). However, the level of nest defence increased only for males. Furthermore, minimum distance was strongly negatively correlated with the intensity of display in both sexes at both nestling ages, while intensity of calling was related only to the intensity of display, not to the minimum distance, when the effect of other factors was partialled out. If the minimum distance to the predator is
an important factor that determines the risk taken by bird parents in the presence of a mammalian predator close to their nest, our findings cast serious doubt on the hypothesis that vocalizations are a relevant estimator of parental risk in birds (see also Welling et al. 1997). The risk to the parent will clearly increase as it displays more intensively, closer to the predator (Curio & Regelmann 1985; Redondo & Carranza 1989). But is this risk related to the intensity of calling? Although in some studies alarm calls have been associated with risk taken by parents, when they were distracting the predator from the nest (Anderson et al. 1980; Curio & Regelmann 1985; Westneat 1989), other studies have considered alarm calls to be a warning to the mate or offspring (Greig-Smith 1980; Knight & Temple 1986b). If the function of calling is to warn the nestlings, a parent may sometimes call more intensively, not because the predator is close by, but because the nestlings are further away. There was no such confounding effect of distance of parents from their nest on the intensity of calling in our study, as parent–nest distance did not differ between the two nestling ages. Furthermore, parents attended to the predator at both nestling ages from the far side of the nest (compare parent–nest distance in Results and minimum distance of parents from predator; Fig. 1). The most intensive calls, when a bird is confronted with a predator (distress calls), can also be explained by the calling for help hypothesis (distress calls are requests for aid from kin or reciprocal altruists, Rohwer et al. 1976), or by the predator attraction hypothesis (distress calls may attract a secondary predator that is capable of distracting the original predator, Ho ¨ gstedt 1983). These hypotheses, and our own results, suggest that calling rates are not necessarily directly associated with parental risk but are often related to the intensity of threat to the nest contents. This intensity can depend on the value of the nest contents, which increases during the breeding cycle.
Number of Mobbing Birds Some authors have suggested that the intensity of nest defence is influenced by the number of mobbers (e.g. Halupka & Halupka 1997). When the size of the mobbing group increases, the risk to individuals decreases, while overall nest defence is more effective. In many bird species the probability and intensity of mobbing and also the size of mobbing groups increase during the nesting period (e.g. Shields 1984) and thus simultaneously with offspring age. This effect can be explained by an increase in the number of males that respond at their nests (Gill & Sealy 1996), but also by a greater willingness of birds to mob the predator during the breeding than the nonbreeding period (Shedd 1983), or by an increasing number of birds that are available to mob (e.g. independent fledglings, finished or unsuccessful breeders; Francis et al. 1989). In our study the number of extrapair mobbers was also markedly higher with old than with young nestlings. However, stronger nest defence and calling intensity with old nestlings was not the result of an increasing number of mobbers. Although the number of mobbers increased with offspring age for both parents,
301
302
ANIMAL BEHAVIOUR, 61, 2
only males increased their responses. In contrast, nest defence was dependent on the size of the mobbing group only for females, not for males. These findings suggest that nest defence intensity is affected primarily by the age of the nestlings, although the number of extrapair mobbers had a marginal effect.
Conclusion McLean & Rhodes (1992) suggested that the proximate explanatory factor for the nest defence behaviour of parent birds is feedback from the current contents of the nest. Continual interaction between parents and their offspring may be, from an evolutionary point of view, the best proximate mechanism to ensure an optimal pattern of responses to predators. In accordance with the reproductive value hypothesis, the intensity of nest defence could increase in response to any developmental changes of offspring (perceived by parents as a proximate stimulus) that, together with passing time, indicate an increase in the nestlings’ probability of survival. Our results suggest that for the meadow pipit the level of interactions between parents and their offspring is probably the most important factor influencing parental responses. The male’s nest defence rapidly increased with increasing interactions with the nest when he started to feed the nestlings. In conclusion, interactions between parents and their nest could be the proximate mechanism that determines parental responses; therefore it should be considered in studies of nest defence in birds. Acknowledgments We are very grateful to Arne Moksnes and Eivin Røskaft for their help in arranging our stay in Norway. We also thank Karel Weidinger, Vladimi´r Remesˇ and anonymous referees for helpful comments on the manuscript. This work was supported by grants from the Grant Agency of the Czech Republic (206/97/1322) and the Czech Ministry of Education (VS 96019). References Anderson, M., Wiklund, C. G. & Rundgren, H. 1980. Parental defence of offspring: a model and an example. Animal Behaviour, 28, 536–542. Barash, D. P. 1975. Evolutionary aspects of parental behavior: distraction behavior of alpine accentor. Wilson Bulletin, 87, 367–373. Breitwisch, R. 1988. Sex differences in defence of eggs and nestlings by the northern mockingbird, Mimus polyglottos. Animal Behaviour, 36, 62–72. Brunton, D. H. 1990. The effects of nesting stage, sex, and type of predator on parental defense by killdeer (Charadrius vociferus): testing models of avian parental defense. Behavioral Ecology and Sociobiology, 26, 181–190. Coulson, J. C. 1956. Mortality and egg production of the meadow pipit with special reference to altitude. Bird Study, 3, 119–132. Curio, E. & Regelmann, K. 1985. The behaviour dynamics of great tits (Parus major) approaching a predator. Zeitschrift fu¨r Tierpsychologie, 69, 3–18.
East, M. 1981. Alarm calling and parental investment in the robin Erithacus rubecula. Ibis, 123, 223–230. Francis, A. M., Hailman, J. P. & Woolfenden, G. E. 1989. Mobbing by Florida scrub jays: behaviour, sexual asymmetry, role of helpers and ontogeny. Animal Behaviour, 38, 795–816. Gill, S. A. & Sealy, S. G. 1996. Nest defence by yellow warblers: recognition of a brood parasite and avian predator. Behaviour, 133, 263–282. Greig-Smith, P. W. 1980. Parental investment in nest defence by stonechat (Saxicola torquata). Animal Behaviour, 28, 604–619. Halupka, K. 1994. Incubation feeding in meadow pipit Anthus pratensis affects female time budget. Journal of Avian Biology, 25, 251–253. Halupka, K. & Halupka, L. 1997. The influence of reproductive season stage on nest defence by meadow pipits (Anthus pratensis). Ethology, Ecology and Evolution, 9, 89–98. Hayes, P. A. & Robertson, R. J. 1989. The impact of male parental care on female eastern kingbird reproductive success. Wilson Bulletin, 101, 462–467. Ho ¨ gstedt, G. 1983. Adaptation unto death: function of fear screams. American Naturalist, 121, 562–570. Hudson, P. J. & Newborn, D. 1990. Brood defence in precocial species: variations in the distraction displays of red grouse, Lagopus lagopus scoticus. Animal Behaviour, 40, 254–261. Knight, R. L. & Temple, S. A. 1986a. Methodological problems in studies of avian nest defence. Animal Behaviour, 34, 561–566. Knight, R. L. & Temple, S. A. 1986b. Nest defence in the American goldfinch. Animal Behaviour, 34, 887–897. MacDougall-Shackleton, E. A. & Robertson, R. J. 1998. Confidence of paternity and parental care by eastern bluebirds. Behavioral Ecology, 9, 201–205. McLean, I. G. & Rhodes, G. 1992. Enemy recognition and response in birds. In: Current Ornithology (Ed. by D. M. Power), pp. 173–211. New York: Plenum. Martin, K. & Horn, A. G. 1993. Clutch defense by male and female willow ptarmigan Lagopus lagopus. Ornis Scandinavica, 24, 261–266. Maynard Smith, J. 1977. Parental investment: a prospective analysis. Animal Behaviour, 25, 1–9. Montgomerie, R. D. & Weatherhead, P. J. 1988. Risk and rewards of nest defence by parent birds. Quarterly Review of Biology, 63, 167–187. Onnebrink, H. & Curio, E. 1991. Brood defense and age of young: a test of the Vulnerability Hypothesis. Behavioral Ecology and Sociobiology, 29, 61–68. Pedroli, J. C. 1978. Breeding success of the meadow pipit Anthus pratensis in the Swiss Jura. Ornis Scandinavica, 9, 168–171. Redondo, T. & Carranza, J. 1989. Offspring reproductive value and nest defense in the magpie (Pica pica). Behavioral Ecology and Sociobiology, 25, 369–378. Regelmann, K. & Curio, E. 1983. Determinants of brood defence in the great tit Parus major L. Behavioral Ecology and Sociobiology, 13, 131–145. Regelmann, K. & Curio, E. 1986. How do great tit (Parus major) pair mates cooperate in brood defence? Behaviour, 97, 10–36. Rohwer, S., Fretwell, S. D. & Tuckfield, R. C. 1976. Distress screams as a measure of kinship in birds. American Midland Naturalist, 96, 418–430. Rytko ¨ nen, S., Koivula, K. & Orell, M. 1990. Temporal increase in nest defence intensity of the willow tit (Parus montanus): parental investment or methodological artifact? Behavioral Ecology and Sociobiology, 27, 283–286. Rytko ¨ nen, S., Orell, M. & Koivula, K. 1993. Sex-role reversal in willow tit nest defence. Behavioral Ecology and Sociobiology, 33, 275–282. Rytko ¨ nen, S., Orell, M., Koivula, K. & Soppela, M. 1995. Correlation between two components of parental investment:
PAVEL & BURES { : NEST DEFENCE IN PIPITS
nest defence intensity and nestling provisioning effort of willow tits. Oecologia, 104, 386–393. Shedd, D. H. 1983. Seasonal variation in mobbing intensity in the black-capped chickadee. Wilson Bulletin, 95, 343–348. Shields, W. M. 1984. Barn swallow mobbing: self-defence, collateral kin defence, group defence or parental care? Animal Behaviour, 32, 132–148. StatSoft. 1998. STATISTICA for Windows (Computer program manual). Tulsa, Oklahoma: Statsoft.
Weatherhead, P. J. 1989. Nest defence by song sparrows: methodological and life history considerations. Behavioral Ecology and Sociobiology, 25, 129–136. Welling, P., Rytko ¨ nen, S., Koivula, K. & Orell, M. 1997. Song rate correlates with paternal care and survival in willow tits: advertisement of male quality? Behaviour, 134, 891–904. Westneat, D. F. 1989. Intensity of nest defence in indigo buntings increases with stage and not number of visits. Auk, 106, 747– 749.
303
Offspring age and nest defence: test of the feedback hypothesis in the meadow pipit VA u CLAV PAVEL & STANISLAV BURES {
Laboratory of Ornithology, Palacky´ University Olomouc (Received 23 September 1999; initial acceptance 3 November 1999; final acceptance 9 August 2000; MS. number: 6347R)
The feedback hypothesis has been proposed to explain variation in nest defence intensity in birds. In species in which the female builds the nest and incubates the eggs, this hypothesis predicts a higher level of nest defence initially for females, whereas males’ responses should increase when they start feeding nestlings. We studied changes in nest defence by both sexes during the nestling period in meadow pipits, Anthus pratensis. We placed a stuffed stoat, Mustela erminea, 5 m from a meadow pipit nest with nestlings aged either 2–4 or 7–12 days and recorded the nest defence behaviour of both parents for 10 min. Males came closer to the predator and mobbed more intensely for older nestlings whereas females defended the nest at a high intensity from the beginning of the nestling period. This finding agrees with the predictions of the feedback hypothesis. We also discuss possible functions of alarm calls and number of mobbing birds during nest defence.
will increase with nestling age as the nest becomes more conspicuous (Redondo & Carranza 1989; Onnebrink & Curio 1991). One might also hypothesize that the proximate stimulus that influences nest defence intensity in parent birds is feedback from the current contents of the nest. Based on that prediction McLean & Rhodes (1992) suggested the feedback hypothesis. They proposed that the variability in parental responses to predators is determined primarily by the interaction between parents and their nest and young. A complete nest without a clutch provides a static visual signal, which maintains parental activities at the nest, while a nest containing eggs provides stronger signals affecting incubation. As the breeding attempt progresses, the growing, moving and vocalizing chicks provide stronger signals, which rapidly reinforce a still higher level of parental responses (including nest defence). In species in which parents share parental activities unequally, the feedback hypothesis predicts that the sex that has more interactions with the nest should defend it more vigorously. This hypothesis also predicts that males and females will change the intensity of their nest defence differentially as their offspring grow. In some bird species (such as meadow pipits, Anthus pratensis) only the female builds the nest and incubates the eggs, while the male initially sporadically feeds her or the young nestlings (Halupka 1994). As the nestlings develop, their growing demands reinforce the male’s interactions with the nest (through increasing feeding activity). The
In many bird species parents increase the intensity of nest defence as their offspring get older (Anderson et al. 1980; Brunton 1990; Rytko ¨ nen et al. 1990). However, at the same time as the offspring grow, the breeding season progresses (Regelmann & Curio 1983), the opportunity to renest declines (Barash 1975), and in many studies (as a methodological artefact) the number of preceding visits by the experimenter to check the nest also increases (Knight & Temple 1986a). Only a few studies have separated these factors from one another, but the effect of offspring age was the most important in the majority of them (Redondo & Carranza 1989; Rytko ¨ nen et al. 1990). Furthermore male and female behaviour can change differentially as the offspring grow (Regelmann & Curio 1986; Martin & Horn 1993; Rytko ¨ nen et al. 1993). Several functional hypotheses have been proposed to account for the increase in nest defence intensity with offspring age. The reproductive value hypothesis predicts that parental investment in short-lived passerines increases during the breeding attempt as the current offspring’s chances of survival increase and the reproductive potential of the parent decreases (Maynard Smith 1977). This hypothesis explains the ultimate causation of parental behaviour, but what is the proximate factor that influences patterns of parental responses? The vulnerability hypothesis predicts that the risk of a predator detecting the nest (and hence also nest defence intensity) Correspondence: V. Pavel, Laboratory of Ornithology, Palacky´ University, trˇ. Svobody 26, 771 46 Olomouc, Czech Republic (email: [email protected]). 0003–3472/01/020297+07 $35.00/0
2001 The Association for the Study of Animal Behaviour
297
2001 The Association for the Study of Animal Behaviour
298
ANIMAL BEHAVIOUR, 61, 2
female’s feeding effort also increases but her interactions with the nest increase less than the male’s, because she has incubated the eggs and brooded the young nestlings (unpublished data; for the importance of males in feeding nestlings see, e.g. Hayes & Robertson 1989). In such species the hypothesis predicts a higher level of nest defence initially for the female than for the male, whereas the male should increase his defence when he starts feeding nestlings. The female’s responses should also increase as the nestlings grow, but less than the male’s, because she should vigorously defend a nest with young nestlings. We tested the prediction that the intensity of nest defence increases with the age of the offspring in the meadow pipit and that it increases more for males than for females. To control for the potentially confounding effect of renesting (when renesting is more costly for the female, she should defend the nest more intensively than the male at the young nestling stage if she still has an opportunity to renest) we did experiments in two areas: the Tydal area in Northern Europe (where there is time for only one breeding attempt) and the Jeseni´ky Mountains in Central Europe (where two breeding attempts are common). We evaluated the behaviour of parents defending their nest by exposing them to a predator. We assumed that the risk should increase rapidly as the parent gets closer or more accessible and vulnerable to the predator (Montgomerie & Weatherhead 1988).
METHODS The meadow pipit is a small monogamous passerine that breeds on the ground in open habitats. Only the female builds the nest and incubates the eggs (3–7 eggs, 13 days) while the male occasionally feeds her. Nestlings are cared for and fed initially mainly by the female; the male feeds them more intensively as they grow. Fledglings are fed by both parents for 1–2 weeks (Halupka 1994; unpublished data). We studied meadow pipit populations on alpine meadows 1400–1460 m above sea level in the Jeseni´ky Mountains (Czech Republic; 50N, 17E) in 1995–1998 and in an alpine ecosystem 730–950 m above sea level near Tydal, Sør Trondelag province (central Norway; 63N, 12E) in 1996 and 1998. Both study areas had similar vegetation cover and nest sites. The meadow pipit was the most abundant bird species with a density of ca. 1 breeding pair/ha, and the stoat, Mustela erminea, was one of the most important predators in both study areas (A. Moksnes, personal communication; unpublished data). We usually found the nests by flushing the incubating bird or observing a bird feeding nestlings. We determined nestling age from either the hatching day (day 0) or nestling weight (Coulson 1956). To determine the days of hatching and fledging, we checked the nests every 2–3 days. Because meadow pipits have no appreciable sex differences, we marked females with a white spot on both sides of the neck. We captured some females immediately after the nestlings hatched by covering both the nest and brooding female with a net. Others were caught while
Table 1. Ranked call and behavioural responses of the meadow pipit to the stuffed stoat Ranked response
Call responses 0 1
2 3 Behaviour 1 2 3 4 5
Description
No calling by parent Chirping ‘chutt’ or ‘chitt’ calls; used to maintain contact rather than to express anxiety Alarm calls near nest; a persistent ‘stitt-itt’ Distress and alarm calls near nest Parent watches the nest and stuffed stoat from a distance Parent walks or lands close to the nest and stuffed stoat and watches them Parent displays slight fluttering flight around the nest and stuffed stoat Parent hovers over the stuffed stoat Parent intensively dives or strikes at the stuffed stoat
feeding nestlings; we covered the nest with a cage, the trap door of which was released either by the approaching bird or by a remote control from a hide. We placed a stuffed stoat 5 m from the meadow pipit nest and recorded the defensive behaviour of the parents for 10 min from a hide 30–60 m away (we built the hide at least 1 h before the experiment to habituate the parents). To evaluate the influence of nestling age on nest defence intensity, we did two experiments at each nest: when the nestlings were young (2–4 days old) and old (7–12 days old). We did the experiments during periods of high feeding activity from 0600 to 1930 hours. A duration of 10 min corresponds with the feeding frequency of meadow pipits (5.8 times/h, Pedroli 1978; 10.0 times/h, unpublished data) and was sufficient to record parental nest defence activities. The procedure had no apparent adverse effects on the adults or nestlings. Parents started to feed the nestlings a few minutes after being caught and immediately after we removed the predator model. Nest defence varied from quietly watching from a distance to noisy contact dives or strikes at the stoat. Many studies have used the distance to the predator model, the latency to approach, whether the bird performs a risky display, calling rates or mobbing intensity to assess the risk to parents of defending their clutch or brood (e.g. Greig-Smith 1980; Regelmann & Curio 1983; Westneat 1989). After preliminary observations, we decided to use the following variables: (1) minimum distance, that is, the shortest distance between the bird and the stuffed stoat; (2) intensity of calling, that is, the calls of the parents ranked from 0 (no calling) to 3 (the most intensive alarm and distress calls; Table 1); (3) intensity of display, ranked from 1 (minimal activity) to 5 (maximal activity; cf. Hudson & Newborn 1990; Halupka & Halupka 1997; Table 1). We also recorded (4) the number of extrapair mobbers reflecting the number of mobbing or calling birds (except the parents) close to the stuffed predator (Halupka & Halupka 1997); and (5)
PAVEL & BURES { : NEST DEFENCE IN PIPITS
Table 2. Comparisons of nest defence activities of meadow pipit parents in the Jeseni´ky Mountains (J) and in the Tydal area (T) Nestlings 2–4 days old
Minimum distance Intensity of calling Intensity of display
Nestlings 7–12 days old
Both parents NJ =18 NT =23
Male NJ =11 NT =21
Female NJ =11 NT =21
Both parents NJ =18 NT =23
Male NJ =11 NT =21
Female NJ =11 NT =21
0.46 0.71 0.78
−1.11 0.50 1.19
1.07 0.08 0.04
0.07 0.80 0.95
−0.56 0.16 0.83
0.40 0.46 0.64
Z values from Mann–Whitney U tests are given. All P>0.2.
parent–nest distance (minimum distance between the defending parent and its nest) to control for their effects. During 1995–1998 we conducted 41 paired experiments (every nest was tested twice, when nestlings were young and old) on 18 nests in the Jeseni´ky Mountains and 23 nests in the Tydal area. Females were not marked until June 1996, therefore only 32 paired experiments (11 nests from the Jeseni´ky Mountains and 21 from the Tydal area) were available to compare the behaviour of males and females. The other nine paired experiments were included when we analysed the reactions of both parents. Because our results from previous studies of distraction displays (unpublished data) and from this study (Table 2) showed that nest defence intensity did not differ between the Jeseni´ky Mountains and the Tydal area (despite a marked difference in the opportunity to renest; see Introduction), we assumed that the opportunity to renest did not seriously influence sexual differences in defence behaviour, and we pooled data from the two study areas for the following analyses. We had two licences for this study: (1) from the Bird Ringing Centre Prague and the Ministry of Environment of the Czech Republic (for field research on birds in the National Park Pradeˇd in the Jeseniky Mountains). No special licence was necessary for study in the Tydal area. We used STATISTICA for Windows software (StatSoft 1998) for data analysis. For analyses of the riskdetermining variables we used nonparametric methods. To compare the behaviour of parents with young and old nestlings at the same nest, we computed Wilcoxon signed-ranks tests. To evaluate the relative relationships between single risk-determining variables and the number of extrapair mobbers we partialled out the effects of other factors. To do this we computed Pearson partial correlations between the compared variables, when data were rank ordered (Spearman partial rank order correlations). All tests were two tailed. RESULTS All risk-determining variables (minimum distance, intensity of calling and intensity of display) were significantly interdependent and related to the number of extrapair mobbers for both males and females and also for young and old nestlings, when single correlations were
computed. In all cases both intensity of calling and intensity of display decreased with increasing distance of the parent from the predator. Furthermore, minimum distance decreased and intensity of calling and intensity of display increased with increasing number of extrapair mobbers (Table 3). When the effects of all variables not in question were partialled out, only the interdependence of minimum distance and intensity of display was highly significant in all cases. Intensity of calling was significantly related only to intensity of display (in all cases except for males with old nestlings), but not to minimum distance (in all cases). Number of extrapair mobbers was marginally significantly correlated only with minimum distance of females with young nestlings and intensity of display of females with old nestlings. Only intensity of calling increased significantly with nestling age for both parents together. Although the directions of differences in minimum distance and intensity of display corresponded with increasing risk in old nestlings, the results were nonsignificant (Table 4). Furthermore, there were significantly more extrapair mobbers during experiments with old than with young nestlings (XSE=0.710.29 versus 2.000.49; paired t test: t40 = 2.45, P=0.019). The increase in risk taking with nestling age was more evident in males than in females (Fig. 1). Although intensity of calling was significantly stronger for old nestlings in both parents, minimum distance was significantly shorter for old nestlings only in males. An increase in the intensity of displays in males was nonsignificant (Fig. 1). Males decreased their minimum distance significantly more than females (Table 4). We found no differences in parent–nest distance between experiments with young and old nestlings for both parents together (XSE=7.411.58 versus 6.600.95; Wilcoxon signed-ranks test: Z=0.06, N=41, P=0.951) or for males (16.552.25 versus 14.361.84; Z=1.24, N=32, P=0.214) and females (7.011.96 versus 7.421.33; Z=1.04, N=32, P=0.300). Females were near the nest in all experiments, but we did not observe males in three experiments with young nestlings and two (the same nests as in the first case) with old nestlings (21 =0.22, P=0.641). As the males might have been present but hidden from our view, we recorded their responses as
299
300
ANIMAL BEHAVIOUR, 61, 2
Table 3. Spearman rank correlations between measures of meadow pipit nest defence behaviour (minimum distance to the predator, intensity of display and intensity of calling) and number of extrapair mobbers Male (N=32)
Nestlings 2–4 days old Distance–Calling Distance–Display Distance–Number Calling–Display Calling–Number Display–Number Nestlings 7–12 days old Distance–Calling Distance–Display Distance–Number Calling–Display Calling–Number Display–Number
Female (N=32)
rs
Partial r
rs
Partial r
−0.64*** −0.88*** −0.61*** 0.76*** 0.50** 0.62***
0.10 −0.73*** −0.18 0.52** 0.08 0.15
−0.58*** −0.80*** −0.58*** 0.64*** 0.46** 0.45**
−0.03 −0.68*** −0.40* 0.38* 0.22 −0.08
−0.66*** −0.90*** −0.73*** 0.67*** 0.56*** 0.77***
−0.16 −0.73*** −0.13 0.19 0.07 0.34
−0.49** −0.82*** −0.45** 0.73*** 0.65*** 0.67***
0.21 −0.77*** 0.17 0.48** 0.27 0.37*
Simple and partial correlation coefficients are shown. *P<0.05; **P<0.01; ***P<0.001. Table 4. Comparison of responses of both parents together and males and females separately to a stuffed stoat close to the nest when nestlings were young (2–4 days) and old (7–12 days) and of the change in response with nestling age for males and females
Both parents (N=41)
Male (N=32)
Female (N=32)
Change in response of males versus females (N=32)
1.93† 4.00** 1.52
2.32* 3.34** 1.86†
0.71 3.32** 0.58
2.03* 0.19 1.00
Young versus old nestlings
Minimum distance Intensity of calling Intensity of display
Z values from Wilcoxon signed-ranks tests are given. †P<0.1; *P<0.05; **P<0.001.
30 m for minimum distance and parent–nest distance, 0 for intensity of calling and 1 for intensity of display. These scores corresponded to the weakest recorded for observed males. DISCUSSION
Differences Between the Sexes Males and females differed in how the risk-determining variables minimum distance and intensity of display changed with nestling age (Table 4, Fig. 1). Males, but not females, changed their behaviour from watching the nest and predator from a distance when the nestlings were 2–4 days old to direct attacks on the predator when they were 7–12 days old. This finding is in accordance with predictions of the feedback hypothesis (McLean & Rhodes 1992). Because the female meadow pipit builds the nest and incubates the eggs her level of interaction with the nest is high before hatching, and she should defend it relatively intensively just after hatching. The male has few interactions with the nest before hatching, but feeds
the nestlings regularly; as his feeding effort increases, so should his nest defence activities. Similar results have been obtained with other species. Breitwisch (1988) observed that males, but not females, of the northern mockingbird, Mimus polyglottos, approached intruders more closely as the nestlings grew. Rytko ¨ nen et al. (1995) found that the nest defence intensity of female willow tits, Parus montanus, was related to their provisioning effort; although in males this relationship was nonsignificant, those males that visited the nest most frequently (but with the smallest load size and hence lower provisioning effort) were the most aggressive nest defenders. Such findings are consistent with our results and could be explained proximately by the feedback hypothesis as well as ultimately by the reproductive value hypothesis. Our results may also be affected by the occurrence of extrapair paternity in our populations. If males have any doubt about their paternity, their level of parental investment should be lower than that of females. If the probability of extrapair paternity changes during the breeding season, their parental investment should
Minimum distance (m)
PAVEL & BURES { : NEST DEFENCE IN PIPITS
30 25 20 15 10 5 0
Intensity of calling
3
2
1
0
Intensity of display
5 4 3 2 1
Young
Old Male
Young
Old Female
Figure 1. Minimum distance (m) from stuffed stoat, intensity of calling and intensity of display (ranked responses, see Table 1) of males and females with young and old nestlings. Bars indicate median values, error bars indicate interquartile range (all N=32).
change correspondingly. Although this prediction was not confirmed when confidence of paternity was experimentally decreased (e.g. in males of the eastern bluebird, Sialia sialis, MacDougall-Shackleton & Robertson 1998) it can add to the variation in male defence behaviour. However, as far as we know the level of extrapair paternity, if any, for meadow pipits has not been measured.
Do Alarm Calls Reflect Parental Risk? The strong increase in the intensity of calling with nestling age for both sexes corresponds with the results of other studies (Greig-Smith 1980; East 1981; Weatherhead 1989). However, the level of nest defence increased only for males. Furthermore, minimum distance was strongly negatively correlated with the intensity of display in both sexes at both nestling ages, while intensity of calling was related only to the intensity of display, not to the minimum distance, when the effect of other factors was partialled out. If the minimum distance to the predator is
an important factor that determines the risk taken by bird parents in the presence of a mammalian predator close to their nest, our findings cast serious doubt on the hypothesis that vocalizations are a relevant estimator of parental risk in birds (see also Welling et al. 1997). The risk to the parent will clearly increase as it displays more intensively, closer to the predator (Curio & Regelmann 1985; Redondo & Carranza 1989). But is this risk related to the intensity of calling? Although in some studies alarm calls have been associated with risk taken by parents, when they were distracting the predator from the nest (Anderson et al. 1980; Curio & Regelmann 1985; Westneat 1989), other studies have considered alarm calls to be a warning to the mate or offspring (Greig-Smith 1980; Knight & Temple 1986b). If the function of calling is to warn the nestlings, a parent may sometimes call more intensively, not because the predator is close by, but because the nestlings are further away. There was no such confounding effect of distance of parents from their nest on the intensity of calling in our study, as parent–nest distance did not differ between the two nestling ages. Furthermore, parents attended to the predator at both nestling ages from the far side of the nest (compare parent–nest distance in Results and minimum distance of parents from predator; Fig. 1). The most intensive calls, when a bird is confronted with a predator (distress calls), can also be explained by the calling for help hypothesis (distress calls are requests for aid from kin or reciprocal altruists, Rohwer et al. 1976), or by the predator attraction hypothesis (distress calls may attract a secondary predator that is capable of distracting the original predator, Ho ¨ gstedt 1983). These hypotheses, and our own results, suggest that calling rates are not necessarily directly associated with parental risk but are often related to the intensity of threat to the nest contents. This intensity can depend on the value of the nest contents, which increases during the breeding cycle.
Number of Mobbing Birds Some authors have suggested that the intensity of nest defence is influenced by the number of mobbers (e.g. Halupka & Halupka 1997). When the size of the mobbing group increases, the risk to individuals decreases, while overall nest defence is more effective. In many bird species the probability and intensity of mobbing and also the size of mobbing groups increase during the nesting period (e.g. Shields 1984) and thus simultaneously with offspring age. This effect can be explained by an increase in the number of males that respond at their nests (Gill & Sealy 1996), but also by a greater willingness of birds to mob the predator during the breeding than the nonbreeding period (Shedd 1983), or by an increasing number of birds that are available to mob (e.g. independent fledglings, finished or unsuccessful breeders; Francis et al. 1989). In our study the number of extrapair mobbers was also markedly higher with old than with young nestlings. However, stronger nest defence and calling intensity with old nestlings was not the result of an increasing number of mobbers. Although the number of mobbers increased with offspring age for both parents,
301
302
ANIMAL BEHAVIOUR, 61, 2
only males increased their responses. In contrast, nest defence was dependent on the size of the mobbing group only for females, not for males. These findings suggest that nest defence intensity is affected primarily by the age of the nestlings, although the number of extrapair mobbers had a marginal effect.
Conclusion McLean & Rhodes (1992) suggested that the proximate explanatory factor for the nest defence behaviour of parent birds is feedback from the current contents of the nest. Continual interaction between parents and their offspring may be, from an evolutionary point of view, the best proximate mechanism to ensure an optimal pattern of responses to predators. In accordance with the reproductive value hypothesis, the intensity of nest defence could increase in response to any developmental changes of offspring (perceived by parents as a proximate stimulus) that, together with passing time, indicate an increase in the nestlings’ probability of survival. Our results suggest that for the meadow pipit the level of interactions between parents and their offspring is probably the most important factor influencing parental responses. The male’s nest defence rapidly increased with increasing interactions with the nest when he started to feed the nestlings. In conclusion, interactions between parents and their nest could be the proximate mechanism that determines parental responses; therefore it should be considered in studies of nest defence in birds. Acknowledgments We are very grateful to Arne Moksnes and Eivin Røskaft for their help in arranging our stay in Norway. We also thank Karel Weidinger, Vladimi´r Remesˇ and anonymous referees for helpful comments on the manuscript. This work was supported by grants from the Grant Agency of the Czech Republic (206/97/1322) and the Czech Ministry of Education (VS 96019). References Anderson, M., Wiklund, C. G. & Rundgren, H. 1980. Parental defence of offspring: a model and an example. Animal Behaviour, 28, 536–542. Barash, D. P. 1975. Evolutionary aspects of parental behavior: distraction behavior of alpine accentor. Wilson Bulletin, 87, 367–373. Breitwisch, R. 1988. Sex differences in defence of eggs and nestlings by the northern mockingbird, Mimus polyglottos. Animal Behaviour, 36, 62–72. Brunton, D. H. 1990. The effects of nesting stage, sex, and type of predator on parental defense by killdeer (Charadrius vociferus): testing models of avian parental defense. Behavioral Ecology and Sociobiology, 26, 181–190. Coulson, J. C. 1956. Mortality and egg production of the meadow pipit with special reference to altitude. Bird Study, 3, 119–132. Curio, E. & Regelmann, K. 1985. The behaviour dynamics of great tits (Parus major) approaching a predator. Zeitschrift fu¨r Tierpsychologie, 69, 3–18.
East, M. 1981. Alarm calling and parental investment in the robin Erithacus rubecula. Ibis, 123, 223–230. Francis, A. M., Hailman, J. P. & Woolfenden, G. E. 1989. Mobbing by Florida scrub jays: behaviour, sexual asymmetry, role of helpers and ontogeny. Animal Behaviour, 38, 795–816. Gill, S. A. & Sealy, S. G. 1996. Nest defence by yellow warblers: recognition of a brood parasite and avian predator. Behaviour, 133, 263–282. Greig-Smith, P. W. 1980. Parental investment in nest defence by stonechat (Saxicola torquata). Animal Behaviour, 28, 604–619. Halupka, K. 1994. Incubation feeding in meadow pipit Anthus pratensis affects female time budget. Journal of Avian Biology, 25, 251–253. Halupka, K. & Halupka, L. 1997. The influence of reproductive season stage on nest defence by meadow pipits (Anthus pratensis). Ethology, Ecology and Evolution, 9, 89–98. Hayes, P. A. & Robertson, R. J. 1989. The impact of male parental care on female eastern kingbird reproductive success. Wilson Bulletin, 101, 462–467. Ho ¨ gstedt, G. 1983. Adaptation unto death: function of fear screams. American Naturalist, 121, 562–570. Hudson, P. J. & Newborn, D. 1990. Brood defence in precocial species: variations in the distraction displays of red grouse, Lagopus lagopus scoticus. Animal Behaviour, 40, 254–261. Knight, R. L. & Temple, S. A. 1986a. Methodological problems in studies of avian nest defence. Animal Behaviour, 34, 561–566. Knight, R. L. & Temple, S. A. 1986b. Nest defence in the American goldfinch. Animal Behaviour, 34, 887–897. MacDougall-Shackleton, E. A. & Robertson, R. J. 1998. Confidence of paternity and parental care by eastern bluebirds. Behavioral Ecology, 9, 201–205. McLean, I. G. & Rhodes, G. 1992. Enemy recognition and response in birds. In: Current Ornithology (Ed. by D. M. Power), pp. 173–211. New York: Plenum. Martin, K. & Horn, A. G. 1993. Clutch defense by male and female willow ptarmigan Lagopus lagopus. Ornis Scandinavica, 24, 261–266. Maynard Smith, J. 1977. Parental investment: a prospective analysis. Animal Behaviour, 25, 1–9. Montgomerie, R. D. & Weatherhead, P. J. 1988. Risk and rewards of nest defence by parent birds. Quarterly Review of Biology, 63, 167–187. Onnebrink, H. & Curio, E. 1991. Brood defense and age of young: a test of the Vulnerability Hypothesis. Behavioral Ecology and Sociobiology, 29, 61–68. Pedroli, J. C. 1978. Breeding success of the meadow pipit Anthus pratensis in the Swiss Jura. Ornis Scandinavica, 9, 168–171. Redondo, T. & Carranza, J. 1989. Offspring reproductive value and nest defense in the magpie (Pica pica). Behavioral Ecology and Sociobiology, 25, 369–378. Regelmann, K. & Curio, E. 1983. Determinants of brood defence in the great tit Parus major L. Behavioral Ecology and Sociobiology, 13, 131–145. Regelmann, K. & Curio, E. 1986. How do great tit (Parus major) pair mates cooperate in brood defence? Behaviour, 97, 10–36. Rohwer, S., Fretwell, S. D. & Tuckfield, R. C. 1976. Distress screams as a measure of kinship in birds. American Midland Naturalist, 96, 418–430. Rytko ¨ nen, S., Koivula, K. & Orell, M. 1990. Temporal increase in nest defence intensity of the willow tit (Parus montanus): parental investment or methodological artifact? Behavioral Ecology and Sociobiology, 27, 283–286. Rytko ¨ nen, S., Orell, M. & Koivula, K. 1993. Sex-role reversal in willow tit nest defence. Behavioral Ecology and Sociobiology, 33, 275–282. Rytko ¨ nen, S., Orell, M., Koivula, K. & Soppela, M. 1995. Correlation between two components of parental investment:
PAVEL & BURES { : NEST DEFENCE IN PIPITS
nest defence intensity and nestling provisioning effort of willow tits. Oecologia, 104, 386–393. Shedd, D. H. 1983. Seasonal variation in mobbing intensity in the black-capped chickadee. Wilson Bulletin, 95, 343–348. Shields, W. M. 1984. Barn swallow mobbing: self-defence, collateral kin defence, group defence or parental care? Animal Behaviour, 32, 132–148. StatSoft. 1998. STATISTICA for Windows (Computer program manual). Tulsa, Oklahoma: Statsoft.
Weatherhead, P. J. 1989. Nest defence by song sparrows: methodological and life history considerations. Behavioral Ecology and Sociobiology, 25, 129–136. Welling, P., Rytko ¨ nen, S., Koivula, K. & Orell, M. 1997. Song rate correlates with paternal care and survival in willow tits: advertisement of male quality? Behaviour, 134, 891–904. Westneat, D. F. 1989. Intensity of nest defence in indigo buntings increases with stage and not number of visits. Auk, 106, 747– 749.
303