Behavioral Ecology Vol. 14 No. 5: 724-729
© 2003 International Society for Behavioral Ecology
Increased vigilance of paired males in sexually dimorphic species: distinguishing between alternative explanations in wintering Eurasian wigeon
Centre for Ecology and Hydrology, CEH-Dorset, Winfrith Technology Centre, Winfrith Newburgh, Dorchester, Dorset DT2 8ZD, UK
Address correspondence to M. Guillemain, who is now at the Office National de la Chasse et de la Faune Sauvage, CNERA Avifaune Migratrice, La Tour du Valat, Le Sambuc, 13200 Arles, France. E-mail: m.guillemain{at}oncfs.gouv.fr.
Received 28 November 2001; revised 2 November 2002; accepted 5 January 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In animal pairs, males are often more vigilant than females. This is generally assumed to result from mate guarding (either against predators or other males). However, when males have conspicuous secondary sexual characteristics, they could be constrained to be more vigilant because of a higher predation risk than females. We attempted to distinguish between the "male constraint hypothesis" and two variations of the mate-guarding hypothesis by studying the vigilance behavior of the sexually dimorphic wigeon during early winter, when some males are in breeding plumage and some are not and when not all males are paired. The proportion of time spent vigilant by paired males in breeding plumage was five times higher than any other category of males or females. We found no significant differences between the vigilance levels of unpaired male wigeon in cryptic and in breeding plumage and therefore rejected the male constraint hypothesis. As vigilance levels of paired and unpaired females did not differ either, we rejected the hypothesis that paired males invest in vigilance to reduce their mate's need to be vigilant to predation risks. Paired females interacted less frequently with other wigeon than unpaired ones, and it is probably to protect their female from other males that paired male wigeon increase their vigilance times.
Key words: Anas penelope, Eurasian wigeon, mate guarding, predation risk, sexual differences, vigilance.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Many comparative studies of the time budgets of male and female animals within pairs have found that the proportion of time spent foraging is greater in females than in males, who spend relatively more time in vigilant behavior (e.g., Ashcroft, 1976
; Choudhury and Black, 1993; Dahlgren, 1990). This may arise as a consequence of males guarding their mate either from conspecifics or from predators (e.g. Hogstad, 1995), referred to as the "male investment hypothesis." As well as ensuring males continued access to and survival of their mate, this allows females to increase their food intake and in turn potentially increase the breeding output of the pair (Black, 1996; Hogstad, 1992; Teunissen et al., 1985). However, higher vigilance levels may also be characteristic of males and occur even in unpaired individuals if, for example, their secondary sexual characteristics make them more likely than females to be detected or caught by predators (e.g., Jacobsen and Ugelvik, 1994b; Lendrem, 1983). This phenomenon is referred to as the "male constraint hypothesis." Most dabbling duck species (Anas spp.) are highly dimorphic, with females having a cryptic brown plumage and males having breeding plumage that is brightly colored and highly conspicuous (e.g., Cramp and Simmons, 1977). Thus, the frequent observation of higher vigilance levels in paired male ducks in comparison with their mates (e.g., Jacobsen and Ugelvik, 1994a; Mayhew, 1987; Paulus, 1988) may be due either to the male constraint hypothesis, the male investment hypothesis, or both. In this study, we gathered data on the foraging and vigilance behavior of Eurasian wigeon (Anas penelope) to distinguish between these two hypotheses. Although the male constraint hypothesis and the two versions of the male investment hypothesis (i.e., toward predation risk and toward mate loss) are not necessarily mutually exclusive, they generate different testable predictions concerning the relative vigilance levels exhibited by various classes of bird when studied under identical conditions:
- Hypothesis 1—Paired males are more vigilant than their mates because they have a bright plumage that increases their own risk of predation relative to that of their cryptically colored mates (the male constraint hypothesis). This hypothesis would be supported if unpaired males in breeding plumage were more vigilant than unpaired males in cryptic eclipse plumage.
- Hypothesis 2—Paired males are more vigilant than their mates because they are protecting their mates from predators, providing the female with longer feeding times and thus potentially higher fecundity (the male investment hypothesis toward predation risk). This hypothesis would be supported if paired males in breeding plumage were more vigilant than unpaired males in breeding plumage and paired females were less vigilant than unpaired ones.
- Hypothesis 3—Paired males are more vigilant than their mates because they are protecting their mates from other males, either to protect their pair-bond or to prevent disruption of feeding of the female (the male investment hypothesis toward mate loss). This hypothesis would be supported if paired males in breeding plumage were more vigilant than unpaired males in breeding plumage, paired females exhibited little or no benefit from pairing in terms of reduced vigilance, and paired females were less often involved in intraspecific interactions than unpaired ones.
- Hypothesis 2—Paired males are more vigilant than their mates because they are protecting their mates from predators, providing the female with longer feeding times and thus potentially higher fecundity (the male investment hypothesis toward predation risk). This hypothesis would be supported if paired males in breeding plumage were more vigilant than unpaired males in breeding plumage and paired females were less vigilant than unpaired ones.
Like most dabbling ducks, male wigeon molt into a cryptic eclipse plumage just after breeding and then molt a second time, to their breeding plumage, during autumn/early winter (Cramp and Simmons, 1977). This second molting episode can either occur on the staging areas of the southward autumn migration or on the wintering grounds. Thus, throughout late autumn–early winter, any local population of wigeon is likely to contain a mixture of males that are either already in breeding plumage or still crypticically colored.
Unlike geese, male dabbling ducks often do not stay with their mate after hatching of the young, and pairs are generally considered to last for one season only, although in some species, including wigeon (Mitchell, 1997), pairs of the same individuals may sometimes reunite the following year. Pairing generally occurs during winter in dabbling ducks (Weller, 1965). Thus, throughout late autumn–early winter, any local population of wigeon is likely to contain a mixture of males that are either still single or recently paired.
Accordingly, we conducted our study during the first part of the winter. This maximized the probability of being able to compare the vigilance levels of individuals of differing plumage types, sex, and mating status in the same place and at the same time and so test the various hypotheses while avoiding any confounding effects of environmental variables that may vary in space and time.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Data collection
The source of the data for this study was videotapes of birds foraging on the Zostera spp. beds of the Exe estuary, England. These recordings were made between October and December in both 1996 and 1997. Although these recordings were primarily dedicated to the study of Brent goose (Branta bernicla) behavior (see Pettifor et al., 2000
), they provided a simultaneous record of the behavior of the wigeon that foraged alongside the geese. We recorded a total of approximately 100 h of footage over the two seasons and across all daylight hours. The same observer recorded data from all videotapes. We performed focal sampling (Altman, 1974) on each wigeon that appeared on the screen for more than 60 s (338 birds; Table 1). Each change in behavior of the focal individual was recorded on a portable computer that automatically recorded time to the nearest 0.5 s (continuous recording of time bout measurements; Martin and Bateson, 1986). Because wigeon were observed at a foraging site, there were only three types of behaviors: foraging, vigilance, and social interactions, the latter being direct aggression that could consequently be recorded from the behavior of the focal bird even when the opponent was not visible on the screen.
|
To test for a potential autocorrelation in vigilance behavior, we tested if the length of a scan bout was significantly correlated with the length of the subsequent one for individuals that performed more than 5 scan bouts during a focal observation period (73 birds). Spearman correlations between the lengths of successive scan bouts were significant (p <.05) for only 2 of these 73 individuals, suggesting that autocorrelation in vigilance behavior was very unlikely in this sample of birds. Focal samples lasted 60–329 s (median: 100 s). To ensure that 60 s provided comparable data to longer focal observations, we compared the proportion of time spent vigilant during the first 60 s and the average of subsequent 1-min blocks in those individuals that were observed for more than 2 min (115 birds). There was no significant difference between these two periods of time (bilateral paired t test, t = -0.58, df = 114, p =.56), suggesting that even the shortest of our focal samples provided reliable information.
We recorded the sex, plumage, and pairing status of each focal individual. Pairing status was determined from the birds' behavior (i.e., the proximity and coordination of movements of the two mates). Although this method has never been formally validated, it is obvious that these features enable one to identify with great confidence when two birds are paired (see also Wishart, 1983). This is especially the case when wigeon forage on Zostera, due to the high mobility of birds over Zostera beds. In males, we could also distinguish between juveniles (<1 year) and adults, as the greater upper wing coverts of the juveniles are gray, whereas these feathers are pure white in adults (Cramp and Simmons, 1977). As we only recorded one paired juvenile male, data from this individual were not included in the analyses (n = 337). We did not see a single male in cryptic plumage that was paired.
Because ducks were not individually marked, we may have sampled the same bird more than once. However, this is unlikely to have occurred frequently enough to cause an important problem for statistical analyses in view of (1) the fact that the turnover of individuals can be rapid in wintering dabbling duck populations (Pradel et al., 1997), (2) the very small number of focal observations per sex, age, plumage and pairing status class per day (1–13, median 2) relative to the average number of wigeon wintering on the Exe (ca. 2000; Fox, 1996), and (3) the duration of the study (3 months in each of 2 years).
Data analyses
To confirm that males of different plumage types coexisted at the study site throughout the study period, we used a MIXED procedure (SAS Institute, 1990) to test how the proportion of males in breeding plumage (i.e., that had molted from cryptic to breeding) varied across time (expressed in 10-day periods, "decades," after 1 October) and if a difference existed between adults and juveniles, considering year (1996 or 1997) as a random variable. In the same way, we sought to confirm that our study period spanned the period of pair formation by testing with a MIXED procedure (still considering year as a random variable) how the proportion of paired females and adult males varied across time and if a difference existed between sexes.
The continuous recording of time-bout behavior allowed us not only to compare levels of vigilance between birds of differing plumage types and pairing status, but also to assess if (and potentially how) the structure of vigilance behavior varied between these bird types. We analyzed three vigilance parameters: scan-bout length (the length of head-up vigilance bouts), scan-bout frequency (the number of these bouts per minute), and the percentage of feeding time spent vigilant (total time spent in head-up vigilance relative to total time spent feeding). The percentage of time spent vigilant was arcsine-transformed in all analyses (Sokal and Rohlf, 1995).
As a prerequisite to testing the three principal hypothesis, it was necessary to establish whether on the Exe estuary paired male wigeon in breeding plumage did indeed exhibit higher levels of vigilance than their mates. If this were not the case, there would be little merit in testing the predictions generated by hypotheses that seek to explain this phenomenon. For most paired individuals (38 out of 45 females and 38 out of 42 males), focal samples were recorded for both mates. This offered the opportunity to test directly if vigilance (both the proportion of time spent vigilant and the structure of vigilance; i.e., length and frequency of bouts) differed between a paired male and its mate (to date, most studies of sexual differences in vigilance levels have considered average individuals in the population rather than the two members of pairs; e.g., Mayhew, 1987). We used paired t tests in this analysis. Note that only 37 pairs could be analyzed for scan-bout length because one paired female did not scan at all during the focal observation.
We then tested whether the three vigilance parameters differed between unpaired males with cryptic and breeding plumage (hypothesis 1), taking into account the potential confounding effect of age (general linear model [GLM] analyses with age, plumage [breeding or cryptic] and age x plumage as potential explanatory factors). We then tested the effect of mate guarding on the vigilance of both sexes by studying the relative effect of sex, pairing (paired or unpaired) and sex x pairing on the vigilance parameters of individuals, again using GLMs and excluding cryptic males (hypotheses 2 and 3). The power of nonsignificant tests was assessed using the observed power option of SPSS GLMs. After ensuring that focal sample lengths did not differ significantly between sexes and birds of different pairing status (GLM with sex, pairing and sex x pairing, F3,333 = 0.27, p =.8483), we tested (chi-square test) if the number of females and adult males involved in at least one interaction (only four individuals were involved in two or more interactions) differed between paired and unpaired individuals (hypothesis 3). Intrapair interactions (i.e., interactions with mate) were not considered, as they do not represent a potential threat to the pair-bond.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Molting and pairing chronology
The proportion of males in breeding plumage increased with time over the course of the study period (time: p =.0162) and differed between adults and juveniles (age: p =.0345) but did not increase at a different rate in adults and juveniles (age x time: p =.3669). Although most males were in cryptic plumage at the beginning of the study period, all adult males were in breeding plumage during the last part of the study (Figure 1). The proportion of birds that was paired increased significantly with time (p =.0015). Sex and sex x time were not significant factors (p =.7106 and p =.8944, respectively). No birds were paired during the first decade of the study, and the proportion of focal birds that were paired increased gradually and at the same rate in females and adult males to reach at least half of the birds at the end of the period (Figure 2).
|
|
Vigilance parameters
Paired males versus paired females
Although the length of scan bouts did not differ between mates (paired t test: t = -1.46, df = 36, p = 0.1535), scan-bout frequency differed significantly between paired males and their mates (males: 3.18 scans/min ± 0.29 SE, n = 38; females: 1.72 scans/min ± 0.15, n = 38; t = 4.75, df = 37, p <.0001). This resulted in paired males spending a much higher percentage of time in vigilance than their mates (males: 14.72% ± 3.18, n = 38; females: 2.76% ± 0.36, n = 38; t = 4.36, df = 37, p <.0001).
Hypothesis 1: cryptic unpaired males versus unpaired males in breeding plumage
Neither the age of an individual, nor its plumage type, nor the interaction age x plumage had a significant effect on average scan-bout length, scan-bout frequency, or the percentage of time spent scanning by unpaired male wigeon (GLMs, all model F values < 0.62, all p >.60). No intrinsic differences in vigilance between unpaired males in different plumages could thus be detected, the power of tests varying between 9% and 18%.
Hypotheses 2 and 3: paired versus unpaired individuals
The sex and pair status of an individual had a highly significant effect on the length and frequency of scan bouts and on the percentage of time spent vigilant (Table 2). The interaction sex x pair status had a significant effect on all three vigilance parameters, but values from only one sex–pair status class (i.e., paired males) differed significantly from all other types of birds (Figure 3). Unpaired male wigeon (both ages combined) in either cryptic or breeding plumage had vigilance levels that were not significantly different from those of females and significantly lower than those exhibited by paired males in breeding plumage (Figure 3). Moreover, the vigilance levels of females did not differ significantly between paired and unpaired individuals (Figure 3). When paired males were excluded from the analysis, neither the length nor the frequency of scan bouts nor the percentage of time spent vigilant differed between sex and pair status classes (GLMs, all model F values < 0.53, all p >.59). The power of tests varied between 10% and 14 %.
|
|
Intraspecific interactions
Hypothesis 3
Paired and unpaired males were equally likely to be involved in at least one interaction (
2 = 1.23, df = 1, p =.267; Table 3). Paired females, however, tended to interact less often than unpaired ones (2 = 3.28, df = 1, p =.070; Table 3), although the small number of interactions led to the difference being just above the threshold for significance.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
During this study there was an obvious mixing of males in cryptic eclipse and breeding plumage, as well as paired and unpaired birds. The acquisition of breeding plumage by males and the pairing of birds both took place gradually over the 2-month study period. The fact that juvenile male wigeon acquired their breeding plumage later than adults on the Exe accords with previous studies on the molting chronology of dabbling ducks (e.g., Cramp and Simmons, 1977
) and may explain why adult males paired sooner than juveniles. Amat (1990) also reported that adult male Eurasian wigeon are more frequently paired than juveniles, and this is also true of American wigeon (Anas americana; Wishart, 1983), gadwall (Anas strepera; Blohm, 1982), and mottled duck (Anas fulvigula; Paulus, 1988). Observations on the Exe showed that male and female wigeon gradually paired across winter. In contrast to the findings of Campredon (1982), we could not detect a significantly higher level or a faster rate of pairing in females than in males, although Figure 2 suggests that a tendency for fewer males than females to be paired may have existed at the end of the study period. Sex-specific rates of pairing in wigeon are likely to be a consequence of biased sex ratio, as there are more males than female wigeon in northwest Europe (as is the case in American wigeon; Wishart, 1983). The sex-ratio is generally 1.32 males per female wigeon in England (Owen and Mitchell, 1988). On the assumption that our observations are an unbiased sample of the population, the ratio on the Exe estuary in 1996 and 1997 was 1.65 males per female (210 focal males and 127 focal females). The fact that molting and pairing occurred gradually, and that the acquisition of breeding plumage necessarily precedes mate acquisition by some time, provided us with a situation in which birds with different plumage types and pairing status were mixed, thus experiencing similar environmental conditions (which otherwise may have confounded the comparison of vigilance levels between individuals, see Elgar, 1989).
Consistent with several other studies of sexually dimorphic ducks (Jacobsen and Ugelvik, 1994a; Mayhew, 1987; Paulus, 1988), paired male wigeon in breeding plumage on the Exe estuary did indeed exhibit greater levels of vigilance than their cryptically colored mates. This was achieved through both longer and more frequent scan bouts (though the length of scan bouts only differed when considering all focal birds and not when comparing only the two mates within pairs). With this confirmation of the expected difference in vigilance behavior between male and female partners, it is then valid to test the predictions of the various hypotheses that might explain the phenomenon.
The vigilance parameters of unpaired males in breeding plumage did not differ significantly from males in cryptic plumage and were not affected by the age of the individual. It is true that the power of these tests was limited, so there may have been small differences in the vigilance parameters of unpaired males of different ages and plumage types that remained undetected. However, if such differences existed, they were negligible compared to the two- to fivefold differences in vigilance parameters between these birds and paired males. Thus, we conclude that our results lend no support to hypothesis 1, that the much higher vigilance we observed in paired males is a consequence of higher predation risk due to brightly colored plumage.
This result differs from those of Lendrem (1983), who showed that the vigilance levels of unpaired male mallard decreased when they molted from conspicuous breeding plumage to the cryptic eclipse one. This result was attributed to a higher predation risk in the former phase. It is possible that the lack of a significant difference on the Exe was due to a relatively low potential predation risk at this site. Mayhew (1987) has shown that the difference in vigilance levels of breeding male wigeon (conspicuous) and females (cryptic) decreases when birds experience safer conditions. The frequency of flyovers by raptors on the Exe has been estimated to be 0.3 flyovers per hour, on average, over the winter (Stillman R, personal communication), which is much lower than the frequency experienced by wigeon in, for example, western France (2.6–9.5 flyovers per hour; Fritz et al., 2000). Apart from paired males, wigeon on the Exe spent less than 4% of their feeding time in vigilance, compared to 15% in western France (Guillemain et al., unpublished data). This also suggests that conditions on the Exe were relatively safe and may help explain the lack of a significant difference in the vigilance of cryptically colored and brightly colored unpaired males. Regardless of the underlying explanation, the key point is that the possession of bright breeding plumage per se was not associated with much higher vigilance levels among male wigeon on the Exe.
Given the lack of a significant effect of plumage type on the vigilance levels of unpaired males, the difference between the vigilance levels of paired males and females cannot be attributed to different levels of mortality risk between sexes due to their plumage but should directly reflect the investment of males in either of the two components of mate guarding (i.e., male investment toward predation risk and/or male investment toward mate loss, hypotheses 2 and 3). There was no significant difference between the vigilance levels of paired and unpaired females, suggesting that females did not benefit from pairing in terms of being able to reduce their own vigilance, presumably for predators. Here again, the power of the tests was limited, but if a difference existed between paired and unpaired females, it could only have been small. The high investment in vigilance by paired males might reasonably be expected to have had more of an effect on the vigilance levels of their mate if this is the purpose of the males' vigilant behavior. Thus, we conclude that our results do not support hypothesis 2 (the male investment toward predation risk). However, paired females tended to be involved less often in intraspecific interactions than unpaired females. This supports hypothesis 3, that the investment in vigilance by paired males serves to protect their mates from interactions with conspecifics.
An alternative hypothesis that might explain the high vigilance of paired males in comparison with other birds is that male wigeon do not become more vigilant once they pair but that high-quality males that can afford to be highly vigilant (i.e., able to survive reduced feeding time because they have a good enough body condition) are also the most likely ones to acquire a mate. This pattern would be consistent with the fact that only one juvenile was observed to be paired during this study. However, if this were true, then the mean vigilance of paired males should have decreased across time as more and more males of somewhat lower quality became paired. This was not the case, as the proportion of time spent vigilant by paired males did not show a significant trend with time (Proc MIXED with year as a random variable; date: p =.80). We therefore conclude that this alternative hypothesis is unlikely to hold.
Mayhew (1987) compared the vigilance levels of male and female wigeon in conditions of varying safety (i.e., varying distances from water and different group sizes). The safest conditions Mayhew recorded can be compared with the conditions we observed on the Exe but, although in his study males appeared to be more vigilant than females (reflecting mate guarding), the differences he observed between the two sexes (2.41% when closest to water, 0.9% in the largest groups) were far lower than the values from this study (10%). However, Mayhew (1987) did not distinguish between paired and unpaired individuals. Our results suggest that the relatively low value recorded for males by Mayhew (1987) may be due to the fact that less vigilant unpaired males were included in the analysis. Our direct comparison of paired males and females shows that the percentage of time that paired males spent in head-up vigilance was almost five times higher than that of their mates and of unpaired males.
Investing so much time in mate guarding has undeniable short-term costs for male wdigeon on the Exe because this increase in vigilance translated to a major decrease in feeding time and more frequent interruptions of feeding activity. It could be argued that because this cost is incurred early in the winter when the environmental conditions are relatively favorable, it will be of little long-term consequence to the birds. However, at the beginning of the winter, when wigeon arrive from migration, they do have to restore their body condition (e.g., Tamisier et al., 1995). Moreover, studies on other species (e.g., oystercatcher Haematopus ostralegus) suggest that poor body condition in the early winter, as assessed by exhibiting arrested molt, is associated with elevated probability of mortality during the subsequent months (Clarke, 1993). The fact that male wigeon invest in pairing at the expense of feeding so soon after their return to their wintering grounds may reflect the strength of male–male competition for females in these areas where there is a biased sex-ratio in the population (Amat, 1990). Mate guarding cannot, however, be explained on the basis of preventing extrapair copulations because copulation generally does not occur during winter in wigeon (Cramp and Simmons, 1977). Indeed, no copulation attempt was observed during the study period, even between paired birds. Mate guarding in early winter is more likely to be related to the high incidence of divorce in ducks at the beginning of this season, as observed in mottled duck by Paulus (1988) and in American wigeon by Wishart (1983). This aspect of behavior has not been specifically studied in Eurasian wigeon, but despite the fact that pairs may sometimes reunite during successive years, strong competition between males for mates has been recorded in this species during winter, when most annual pairs form (Mitchell, 1997). The fact that paired females tended to be involved less often in intraspecific interactions than unpaired females is consistent with the hypothesis that male vigilance prevents paired females from simply leaving with another male.
In conclusion, we found no significant differences between the vigilance levels of unpaired male wigeon in cryptic and in breeding plumage and therefore rejected the male constraint hypothesis. As vigilance levels of paired and unpaired females did not differ either, we rejected the hypothesis that paired males invest in vigilance to reduce their mate's need to be vigilant to predation risks. Paired females interacted less frequently with other wigeon than unpaired ones, and it is probably to protect their female from other males that paired male wigeon increase their vigilance times.
|
ACKNOWLEDGEMENTS |
|---|
We thank Richard Stillman for providing data on the frequency of flyovers by raptors on the Exe estuary and Hervé Fritz for valuable discussions and help with the literature. The filming was conducted by R.W.G.C. and K.H.H. under Natural Environment Research Council/SERAD Thematic grants GST/02/2078 and GST/02/1202 as part of the Large-scale processes in Ecology and Hydrology Thematic Programme. M.G. was supported by a Marie Curie Fellowship of the European Community programme "Human Potential" under contract number HPMF-CT-2000–00945.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Altman J, 1974. Observational study of behaviour: sampling methods. Behaviour 49: 227-267. [Medline]
Amat JA, 1990. Age-related pair bonding by male Eurasian wigeons in relation to courtship activity. Auk 107: 197-198.
Ashcroft RE, 1976. A function of the pairbond in the common eider. Wildfowl 27: 101-105.
Black JM, 1996. Introduction: pair bonds and partnerships. In: Partnerships in birds. The study of monogamy (Black JM, ed). Oxford: Oxford University Press; 3–20.
Blohm RJ, 1982. Differential occurence of yearling and adult male gadwalls in pair bonds. Auk 99: 378-379.
Campredon P, 1982. Démographie et Ecologie du Canard siffleur Anas penelope L. pendant son hivernage en France (PhD dissertation). Montpellier, France: Université des Sciences et Techniques du Languedoc.
Choudhury S, Black JM, 1993. Mate-selection behaviour and sampling strategies in geese. Anim Behav 46: 747-757. [CrossRef]
Clark N, 1993. Wash oystercatchers starving. BTO News 185: 1.
Cramp S, Simmons KEL, 1977. Handbook of the birds of Europe, the Middle East and North Africa, vol. 1. Oxford: Oxford University Press.
Dahlgren J, 1990. Females choose vigilant males: an experiment with the monogamous grey partridge, Perdrix perdrix. Anim Behav 39: 646-651.
Elgar MA, 1989. Predator vigilance and group size in mammals and birds: a critical review of the empirical evidence. Biol Rev 64: 13-33. [Medline]
Fox AD, 1996. Zostera exploitation by Brent geese and wigeon on the Exe estuary, southern England. Bird Study 43: 257-268.
Fritz H, Guillemain M, Guérin S, 2000. Changes in the frequency of prospecting fly-overs by marsh harriers Circus aeruginosus in relation to short-term fluctuations in dabbling duck abundance. Ardea 88: 9-16.
Hogstad O, 1992. Mate protection in alpha pairs of wintering willow tits, Parus montanus. Anim Behav 43: 323-328.
Hogstad O, 1995. Alarm calling by willow tits, Parus montanus, as mate investment. Anim Behav 49: 221-225.
Jacobsen OW, Ugelvik M, 1994a. Effects of presence of waders on grazing and vigilance in breeding wigeon, Anas penelope. Anim Behav 47: 488-490. [CrossRef]
Jacobsen OW, Ugelvik M, 1994b. Grazing and vigilance behaviour of breeding Eurasian wigeon Anas penelope in relation to distance from water. Wildfowl 45: 119-123.
Lendrem DW, 1983. Sleeping and vigilance in birds. 1: Field observations of the mallard. Anim Behav 31: 532-538.
Martin P, Bateson P, 1986. Measuring behaviour. An introductory guide. Cambridge: Cambridge University Press.
Mayhew PW, 1987. Vigilance levels in European wigeon—sexual differences. Wildfowl 38: 77-81.
Mitchell C, 1997. Re-mating in migratory wigeon Anas penelope. Ardea 85: 275-277.
Owen M, Mitchell C, 1988. Movements and migrations of wigeon Anas penelope wintering in Britain and Ireland. Bird Study 35: 47-59.
Paulus SL, 1988. Social behavior and pairing chronology of mottled ducks during autumn and winter in Louisiana. In: Waterfowl in winter (Weller MW, ed). Minneapolis: University of Minnesota Press; 59–70.
Pettifor RA, Caldow RWG, Rowcliffe JM, Goss-Custard JD, Black JM, Hodder KH, Houston AI, Lang A, Webb J, 2000. Spatially explicit, individual-based, behavioural models of the annual cycle of two migratory goose populations. J Appl Ecol 37:(suppl. 1): 103-135. [CrossRef]
Pradel R, Rioux N, Tamisier A, Lebreton JD, 1997. Individual turnover among wintering teal in Camargue: a mark-recapture study. J Wildl Manage 61: 816-821.
SAS Institute,, 1990. SAS/STAT user's guide, version 6, 4th ed., vol. 2. Cary, North Carolina: SAS Institute.
Sokal RR, Rohlf FJ, 1995. Biometry: the principles and practice of statistics in biological research, 3rd ed. New York: Freeman.
Tamisier A, Allouche A, Aubry F, Dehorter O, 1995. Wintering strategies and breeding success: hypothesis for a trade-off in some waterfowl species. Wildfowl 46: 76-88.
Teunissen W, Spaans B, Drent R, 1985. Breeding success in relation to individual feeding opportunities during spring staging in the Wadden Sea. Ardea 73: 109-119.
Weller MW, 1965. Chronology of pair formation in some Nearctic Aythya (Anatidae). Auk 82: 227-235.
Wishart RA, 1983. Pairing chronology and mate selection in the American wigeon (Anas americana). Can J Zool 61: 1733-1743.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||