Is shared male assistance with antiparasitic nest defence costly in the polygynous great reed warbler?

Animal Behaviour 85 (2013) 615e621

Contents lists available at SciVerse ScienceDirect

Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav

Is shared male assistance with antiparasitic nest defence costly in the polygynous great reed warbler? Milica Po
zgayová*, Petr Procházka, Marcel Honza Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic, v.v.i., Brno, Czech Republic

a r t i c l e i n f o Article history: Received 25 January 2012 Initial acceptance 30 March 2012 Final acceptance 17 December 2012 Available online 16 January 2013 MS. number: 12-00064R Keywords: Acrocephalus arundinaceus aggressive behaviour brood parasitism common cuckoo Cuculus canorus great reed warbler mating status

Polygyny typically has negative fitness consequences for secondary females, but may equally impose costs on primary females or even on polygynous males. We investigated how polygynous and monogamous great reed warbler, Acrocephalus arundinaceus, males assist their mates with aggressive nest defence against the common cuckoo, Cuculus canorus, and whether the females adjust their nest defence intensity according to male investment in aggression. Additionally, we investigated whether host social mating status affects host vulnerability to parasitism. We presented taxidermic cuckoo mounts at nests of primary, secondary and monogamous females, and recorded aggressive responses of nest owners. We found that monogamous males defended their nests most aggressively while polygynous males allocated their nest protection effort unevenly between their two mates, responding more vigorously on the primary than secondary nests. In contrast, nest defence intensity of females did not differ with respect to their social status, indicating that females of polygynous males did not compensate for low levels of male aggression. Similarly, we found no differences in natural cuckoo parasitism rates between monogamous, primary and secondary nests. Our results thus suggest that while monogamous females receive more assistance with nest defence than females of polygynous males, this has no effect on the probability of parasitism. Ó 2012 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

Social polygyny, the bond of one male with two or more females at a time, is the most prevalent and perhaps also the most intensively studied polygamous mating system in birds (Møller 1986; Davies 1991; Ligon 1999). Adoption of this strategy is regarded as beneficial to males, since by having multiple broods, they may significantly enhance their reproductive success (Hannon & Dobush 1997; Pearson et al. 2006; Ferretti & Winkler 2009). In contrast, females sharing one male’s territory are expected to compete with each other for resources or for male assistance with parental care and such competition may negatively affect their reproductive output (Verner 1964). It is then natural to expect that females confronted with the cost of polygyny may prefer to breed monogamously. The polygyny threshold model (Verner 1964; Verner & Willson 1966; Orians 1969) is the most widely accepted theoretical explanation of territorial polygyny in birds, although no less important additions and alternatives to this model have been suggested (reviewed by Searcy & Yasukawa 1989; Slagsvold & Lifjeld 1994; Ligon 1999). According to this model, there is a selective advantage for females to mate with a polygynous male rather than with


gayová, Institute of Vertebrate Biology AS CR, v.v.i., * Correspondence: M. Poz
tná 8, CZ-60365 Brno, Czech Republic. Kve E-mail address: [email protected] (M. Po
zgayová).

a bachelor, if the cost of polygyny is compensated for, for example through access to a territory or a male of superior quality. However, polygyny cannot be viewed only as an outcome of female decisions in relation to the variation in male genetic, phenotypic or territorial quality, but also in the context of sexual conflict (Davies 1989; Kempenaers 1995; Smith & Sandell 2005) or as a result of variation in female qualities or condition (Forstmeier et al. 2001; Griggio et al. 2003). It is generally known that polygynous males provide less parental care per nest than monogamous males and often invest more in offspring of their primary (i.e. first mated) than secondary (i.e. second mated) females (e.g. Johnson et al. 1993; Sandell et al. 1996; Forstmeier et al. 2001). However, owing to a trade-off between male sexual advertisement to the secondary females and paternal behaviour at the primary nests, sometimes it is the primary females that are left with a smaller contribution from their mates (Muldal et al. 1986; Pinxten & Eens 1994; Slagsvold & Lifjeld 1994). Females of polygynous males are thus expected to compensate for the reduced paternal help by increasing their own workload (Pinxten et al. 1993; Pinxten & Eens 1994; Sejberg et al. 2000; Redpath et al. 2006), but may not be able to compensate for it completely and therefore fledge fewer or lower quality offspring than monogamously paired females. Such fitness costs are typically reported for the less assisted secondary females (Pinxten

0003-3472/$38.00 Ó 2012 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.anbehav.2012.12.024

616

M. Po
zgayová et al. / Animal Behaviour 85 (2013) 615e621

& Eens 1990; Johnson & Kermott 1993; Pribil 2000). Less attention has been paid to the issue of how mate sharing affects primary females (Hansson et al. 1997; Czapka & Johnson 2000; Trnka et al. 2010) or whether polygyny imposes some costs also on males (e.g. Dunn & Robertson 1993; Lubjuhn et al. 2000; Pilastro et al. 2002). The predominantly investigated form of male parental care in polygynous systems is nestling provisioning (e.g. Alatalo et al. 1981; Urano 1990; Yasukawa et al. 1990). Other forms, such as male incubation behaviour (Pinxten et al. 1993; Smith et al. 1995; Grønstøl 2003) or male feeding of the incubating female (Altenburg et al. 1982; Lifjeld et al. 1987), are less intensely studied in this context. In addition, only a few studies have examined how polygynous males allocate their nest defence effort among the mates of different social status (Knight & Temple 1988; Weatherhead 1990; Yasukawa et al. 1992; Johnson & Albrecht 1993; Trnka & Prokop 2010). Of these studies, none have investigated male assistance with nest defence against brood parasites, although in some host populations, brood parasitism may be as detrimental to host fitness as nest predation if not more so (Rothstein 1990). Moreover, the male role in antiparasitic aggression is thought to be of importance because two defending parents may be more likely to prevent the nest from being parasitized than one parent alone. Indeed, there is some evidence that a higher number of nest defenders decreases the probability of brood parasitism. In colonial nesters, larger groups of hosts may be parasitized at lower rates than smaller groups, which is most probably caused by collective nest defence by the host (Brown & Lawes 2007). In cooperatively breeding birds, an incubating female fed by helpers may spend more time on the nest. Increased host nest attendance may then significantly reduce the likelihood of parasitism (Canestrari et al. 2009). Therefore, lower assistance with nest defence by polygynous males or their lower nest attentiveness might sometimes explain significantly higher parasitism rates in polygynous than monogamous territories (Trnka & Prokop 2011). We investigated whether the social mating system of the great reed warbler, Acrocephalus arundinaceus, affects its aggressive behaviour towards the cuckoo, Cuculus canorus, and whether it is costly in terms of increased parasitism rate. More specifically, we explored how polygynous and monogamous males assist their mates with nest defence against the cuckoo and whether females of particular mating status adjust their nest defence intensity according to that of their mates. Additionally, we compared natural cuckoo parasitism rates among monogamous, primary and secondary nests. We predicted primary females would get less assistance from males than secondary or monogamous females, because during egg laying of primary females, polygynous males may be distracted by activities connected with the acquisition of secondary females. As a consequence, the competing activities of the polygynous males may result in lower male attentiveness at the primary nests and thus these nests may be more vulnerable to parasitism. According to this scenario, we expected a higher parasitism rate on primary than secondary or monogamous nests (see also Trnka & Prokop 2011). We assumed that shared male assistance with antiparasitic nest defence will be costly for primary females and, eventually, also for polygynous males. However, this cost may be prevented if females increase their aggressive behaviour to compensate for the lower male contribution. METHODS Study Species The facultatively polygynous great reed warbler is an important cuckoo host (Moskát & Honza 2002; Kleven et al. 2004; Campobello

& Sealy 2009) known to behave aggressively towards nest intruders, including parasitic cuckoo females (Bártol et al. 2002; Røskaft et al. 2002; Po
zgayová et al. 2009; Trnka & Prokop 2012). In this passerine, only females are responsible for incubation while males guard the nests by watching for enemies, participate in nest defence and help females with caring for the chicks (Cramp 1992). The rate of male polygyny in the great reed warbler varies between 8% and 43% (Dyrcz 1986; Hasselquist 1998; Leisler & Wink 2000; Trnka et al. 2010), depending on the particular year and population studied. Fieldwork We carried out the study on a colour-ringed great reed warbler population, from late April to mid-July 2009 and 2010. During that period, the population consisted of about 100 breeding pairs and, in the 2 years, exhibited a 29% and 21% male polygyny rate, respectively. Polygynous males were almost exclusively bigamous; only two males (one in each year) mated with three females. We systematically searched for nests in littoral vegetation. The nests were found during the nest-building stage or at the beginning of egg laying and were checked daily until clutch completion. At each visit, we numbered a newly laid egg with a waterproof pen according to its laying order (to detect discrepancies in the laying sequence and to ascertain the clutch size) and checked the nest contents for the presence of a parasitic egg. Only clutches containing a cuckoo egg during our nest checks were considered parasitized; otherwise they were considered nonparasitized. Birds were mist-netted and colour-ringed soon after their territory establishment (males) or during the nest-building, egglaying or incubation stages (females). To make the mist-netting more effective, we used mp3 recordings of conspecific song to attract the birds. However, as many birds were colour-marked from previous years, we avoided unnecessary disturbance and confirmed their identity and social mating status based on the resightings of the colour rings. As the mating status of both males and females may change because of the settlement of new females or because of nest failure, we checked the status of each individual several times over the breeding season. Accordingly, a monogamous female was the only female of a monogamous male. A primary female was the first mated female of a polygynous male, sharing the territory simultaneously with a secondary female. Analogously, the secondary female was the second mated female of a polygynous male, sharing the same territory simultaneously with the primary female. At the end of egg laying (mean
SD ¼ 4.7
1.0 days after clutch initiation; mean clutch size
SD ¼ 4.1
0.8 eggs, N ¼ 67), nesting pairs were presented with a taxidermic cuckoo mount, randomly chosen from three specimens. The experimenter attached the mount to a pole <1 m from the focal nest, levelled it with the nest rim, and retreated to a distance of 15e20 m. From there she observed the behaviour of the nest owners with binoculars, allowing each pair member to respond for 2 min from its first arrival within a 5 m diameter around the dummy. If there was no reaction and no bird(s) visible in the vicinity of the nest for 15 min from when the dummy was attached to the pole, the experiment was stopped and the dummy was removed. All experiments were carried out between 0800 and 1900 hours CET. We chose the cuckoo for the experiments because it is a natural enemy of great reed warblers at our study site, where it parasitizes them at a rate of 30e40% (our own data). The cuckoo is only dangerous for host clutches, but not for the adults, which should not bias host aggressive behaviour. Indeed, many previous studies demonstrated that hosts (including the species used in this study) are aggressive towards brood parasites (e.g. Røskaft et al. 2002,

M. Po
zgayová et al. / Animal Behaviour 85 (2013) 615e621

Po
zgayová et al. 2009; Langmore et al. 2012), confirming that they perceive them as serious threats to their eggs. In total, we conducted the experiments on 67 nests (39 in 2009, 28 in 2010) belonging to 44 males and 61 females. Of these nests, 25 were of monogamous, 19 of primary and 23 of secondary females. The two tertiary nests of trigynous males were omitted from the analyses. Five nests were abandoned after the experiment; however, three of them were naturally parasitized. As nest desertion is a common host response towards brood parasitism (Rothstein 1974; Davies & Brooke 1988), this was most probably true also for the three cases of host nest abandonment. Nevertheless, the desertion rate between the nonparasitized experimental nests (2/33) and nonparasitized control nests (4/46) did not differ (Fisher’s exact probability test: P ¼ 1.00). The permission to perform the research at the study site was issued by the local conservation authorities (permit numbers 00312/PA/2008/AOPK and JMK20189/2010). Bird catching and ringing followed the licences (numbers 906 and 1058) and current rules given by the Czech Bird Ringing Centre. All manipulations with the birds adhered to the Animal Care Protocol of the Academy of Sciences of the Czech Republic (numbers 173/2008 and 128/ 2010) and were in compliance with the current Czech Law on the Protection of Animals against Mistreatment (licence numbers V/1/ 2005/28 and 0008/98-M103). Data Analysis As a measure of host aggression, we used the number of contact attacks on the dummy (see also Trnka & Prokop 2012; Trnka et al. 2012a), counted separately for each sex during a 2 min period. We chose this characteristic from those recorded during the experiments, because it showed the highest level of individual variation (range 0e61) and expressed the aggressive nature of the great reed warbler in the best way. As several instances of cuckoos killed by great reed warblers during their nest defence are known (Molnár 1944; Janisch 1948), we considered this behaviour to be the most effective way to prevent parasitic cuckoo females from egg laying. We conducted all statistical analyses in R version 2.12.0 (R Development Core Team 2010). To examine the variation in host antiparasitic aggression, we fitted two generalized linear mixed models (GLMMs) with Poisson error distribution and log link function. In the first model, the number of male contact attacks (count data) was entered as a response variable, nest status (categorical predictor with three levels: monogamous, primary, secondary), cuckoo parasitism (parasitized, nonparasitized), year, laying date (centred around the mean in each nest status category) and time of day when the experiment started were entered as fixed effects, and male identity was entered as a random effect. The second model included number of female contact attacks as a response, the same fixed effects as in the previous model and a random effect of female identity. To investigate the variation in cuckoo parasitism rate, we fitted a third GLMM (with binomial error distribution and logit link function), in which cuckoo parasitism was entered as a binary response variable (yes/no), nest status, year and laying date were entered as fixed effects, and male and female identity were entered as random effects. Female identity was used in the models because a certain proportion of females returned to nest at our study site in the second year of study. Male identity was used because polygynous males attended more than one nest and, similarly to females, some males also returned in the following year. In the analyses we also checked for interaction terms that could be biologically relevant to the goal of our study. The only significant interaction was between nest status and cuckoo parasitism (in the

617

models of host aggression). As both polygyny and cuckoo parasitism are rare and we did not manipulate them experimentally, the data included only four primary parasitized nests. This was much lower than the number of nests in the other nest status  parasitism groups. Consequently, such an unbalanced design together with a few outliers in host aggression on primary parasitized nests seemed to have contributed to the interaction. Although we cannot fully exclude the possibility that this may be a biologically relevant interaction, we have chosen not to include it in the analyses because the direction of the effect appeared unreasonable and probably reflected the effect of chance. On the basis of the three models mentioned above, we specified three sets of candidate models with all combinations of fixed effects. To determine the best-fit model(s) from each set, we adopted the model selection approach based on Akaike’s information criterion (AIC) corrected for low sample size (AICc; Burnham & Anderson 2002). We ranked all competing models according to their Di values, that is, the differences between the ith model and the model with minimum AICc (Di ¼ AICc(i)  AICc(min)). The models were further assigned Akaike weights (wi), providing a quantitative measure of support for each model relative to the others. When more than one model provided similar levels of support, we applied model averaging (Burnham & Anderson 2002) to calculate 95% confidence intervals (CI) and unconditional standard errors (SE) of parameter estimates. This procedure enables to identify the relative importance of each model term in predicting the response variable and to make robust parameter estimates across the set of most likely models. Following the recommendations of Burnham & Anderson (2002), we averaged the models with Di  2, as such models should provide substantial support from the data. All GLMMs were fitted in the package lme4 (Bates et al. 2008), using Laplace approximation for parameter estimations (Bolker et al. 2009). For model ranking and model averaging procedures  2011). As a measure of we employed the package MuMIn (Barton goodness of model fit we used explained deviance expressed as 100  (deviance of a null model  residual deviance of an actual model)/deviance of a null model. RESULTS The model including number of male contact attacks (dependent variable), nest status, parasitism, laying date, year, time (fixed effects) and male identity (random effect) was the only top candidate model (from a total of 32 alternative models) providing substantial support (Di  2) for the variation in male aggressiveness towards the cuckoo (Table 1). According to this model, male aggression was highest at monogamous, lower at primary and lowest at secondary nests (for treatment contrasts see Table 2). Males contact-attacked the cuckoo more if their nests were naturally parasitized, later in a season and in the second year of the study. Male aggression decreased over the course of the day (Table 2). Model selection from the set of 32 candidate models of female aggression resulted in three most likely models with Di  2 (Table 1). All three models included laying date and its importance as the major predictor of female aggression was confirmed also by model averaging (Table 2). Similarly to males, female aggression increased over a breeding season. Other predictors, namely time and year, were not important as the 95% CIs of their modelaveraged estimates contained zero (Table 2). In addition, females were overall less aggressive than males (GLMM: number of contact attacks w sex þ individual identity, treatment contrast males versus females
SE ¼ 1.32
0.36, 95% CI 2.03 to 0.62; Fig. 1). Two models with Di  2 explaining the variation in natural cuckoo parasitism were selected from the set of eight alternative

M. Po
zgayová et al. / Animal Behaviour 85 (2013) 615e621

618

Table 1 Three sets of candidate models explaining the variation in number of male/female contact attacks and cuckoo parasitism rates

Di

wi

Explained deviance (%)

8

576.0

0.0

0.98

12.3

7

585.3

9.3

0.01

10.4

Number of female contact attacks Laying date D Time D Fid Laying date D Time D Year D Fid Laying date D Fid Parasitism þ Laying date þ Time þ Fid .

4 5 3 5

192.4 192.8 193.9 193.9

0.0 0.4 1.5 2.1

0.25 0.20 0.12 0.09

43.8 44.4 42.7 43.9

Cuckoo parasitism rates Nest status D Mid D Fid Fid D Mid Nest status þ year þ Mid þ Fid .

5 3 6

96.7 96.8 98.7

0.0 0.1 2.1

0.29 0.28 0.11

5.2 0.0 5.6

50 40 30 20 10 0

Top candidate models (those with Di  2) are in bold. K: number of estimated parameters; AICc: Akaike information criterion corrected for low sample size; Di: delta AICc (Di ¼ AICc(i)  AICc(min)); wi: Akaike weight showing the relative support of a given model compared to the others; explained model deviance ¼ 100  (null deviance  residual deviance)/null deviance. Mid, Fid: random factors of male and female identity, respectively. For information about the other variables see the Methods.

models (Table 1). One of them contained nest status and random effects; the other included only random effects. After we averaged the two models, however, nest status was no longer important (Table 2). The frequency of parasitism thus did not differ with nest status (Fig. 2).

DISCUSSION Contrary to our prediction, polygynous great reed warbler males protected their primary nests significantly more vigorously than

Table 2 Model-averaged estimates, SEs and 95% confidence intervals (CIs) for the fixed effects across the top candidate models predicting the number of male/female contact attacks and cuckoo parasitism rates Estimate

SE

Lower CI

Upper CI

Number of male contact attacks Intercept Nest status * Primary Secondary Parasitism Yes Laying date y Year 2010 Time z

1.85 1.07 1.31 0.53 0.10 0.46 0.11

0.32 0.25 0.26 0.12 0.02 0.13 0.03

1.22 1.56 1.82 0.29 0.06 0.19 0.17

2.48 0.58 0.80 0.76 0.14 0.72 0.05

Number of female contact attacks Intercept Laying date y Year 2010 Time z

0.38 0.11 0.16 0.12

0.26 0.03 0.26 0.08

0.13 0.05 0.35 0.28

0.89 0.18 0.67 0.05

0.15 0.71 0.17

0.41 0.85 0.40

0.96 2.38 0.96

0.65 0.96 0.61

Cuckoo parasitism rates Intercept Nest status x Primary Secondary

Number of contact attacks

AICc

Monogamous

Primary Nest status

Secondary

Figure 1. Great reed warbler sex-specific aggression towards the cuckoo with respect to nest status. The central bar of the box indicates the median, the box delimits the 25th and 75th percentiles and the whiskers show the data range.

secondary nests; however, they defended the former less intensively than monogamous males. Males were also more aggressive at naturally parasitized than at nonparasitized nests, later in a season and in the second year of the study. Females were less aggressive than males and the level of their nest defence increased over a breeding season. However, females did not compensate for the lowered aggression of polygynous males: their nest defence intensity did not differ with respect to their mating status. Despite the fact that the secondary nests were the least defended by males, we found no differences in cuckoo parasitism rates between the primary, secondary and monogamous nests. Similar allocation of nest defence in polygynous males was also found in a study conducted by Knight & Temple (1988). Therein, the authors showed that when polygynous red-winged blackbird, Agelaius phoeniceus, males were simultaneously presented with two models of their natural nest predator, the American crow, Corvus brachyrhynchos, they defended primary nests more intensively than secondary nests. Thus, when forced to choose, they

100

80 % Parasitized nests

Number of male contact attacks Nest status D Parasitism D Laying date D Year D Time D Mid Nest status þ Parasitism þ Laying date þ Time þ Mid .

K

Males Females

60

60 25 40

23 19

20

The estimates of categorical predictors are presented as treatment contrasts. * Primary versus secondary nests: 0.24
0.11 (SE), 95% CI 0.47 to 0.02. y Clutch initiation date centred around zero for each nest status category. z Time of day when the experiment was conducted. x Primary versus secondary nests: 0.54
0.72 (SE), 95% CI 0.86 to 1.94.

0

Monogamous

Primary Nest status

Secondary

Figure 2. Natural cuckoo parasitism rate in relation to nest status of great reed warbler hosts. Total numbers of nests are given above the bars.

M. Po
zgayová et al. / Animal Behaviour 85 (2013) 615e621

invested in the nests of the primary females more than in those of the secondary females. However, further studies on this issue brought largely different results. Weatherhead (1990) showed that male red-winged blackbirds did not defend primary nests more aggressively against a human intruder than secondary nests when tested one at a time. The lack of male preferences for primary females in the same species was also supported indirectly by Yasukawa et al. (1992), who showed that the time the polygynous males spent in antipredator vigilance did not vary with nest status. Likewise, Johnson & Albrecht (1993) found little evidence that house wren, Troglodytes aedon, females settled with already mated males would obtain less aid in nest defence against model snakes than females settled with unmated males. Despite being presented at secondary nests when males were already feeding their primary broods and thus might be less attentive to secondary nests, the snakes were attacked at similar rates by bigamous and monogamous males (Johnson & Albrecht 1993). In the only nest defence study on polygynous great reed warblers, Trnka & Prokop (2010) found that monogamous males defended their nestlings against a human more aggressively than polygynous males on their primary nests, which corresponds to one of our results. Unfortunately, the authors did not provide any information about the level of male aggression at secondary nests, making a comparison between the primary and the secondary nests impossible. Nevertheless, our findings, together with those on great reed warbler provisioning rates to nestlings (Catchpole et al. 1985; Dyrcz 1986; Bensch & Hasselquist 1991, 1994; Sejberg et al. 2000), indicate that in this species, polygynous males may prefer their primary nests both at the egg and nestling stages, presumably because of higher quality, reproductive value or survival chances of the primary offspring. Apart from the effect of nest status, we also found that males were more aggressive at naturally parasitized than at nonparasitized nests, later in a season and in the second year of the study. Such patterns of responsiveness imply the presence of certain male experience with the cuckoo or positive reinforcement of male aggression induced by loss of fear (Knight & Temple 1986a,
b; Capek et al. 2010). Regarding increased host responsiveness with experience, recent studies showed that some cuckoo or cowbird hosts may indeed be capable of learning to respond to a brood
parasite (Capek et al. 2010; Campobello & Sealy 2011; Langmore et al. 2012). Moreover, decreased male nest defence intensity over the course of the day suggests either lower male propensity to mob the cuckoo or lower male nest attendance later in the day. In general, with a low attentiveness, hosts may be less likely to detect the cuckoo in time and thus the number of birds responding may not suffice to deter it. This probably explains why the cuckoo often parasitizes host nests at that time of day (Davies & Brooke 1988; Honza et al. 2002), since laying in the afternoon is thought to reduce the risk of being spotted by the hosts (Davies 2000). Compared with males, great reed warbler females behaved significantly less aggressively, suggesting the importance of the male role in great reed warbler nest defence; at least against the brood parasite (see also Po
zgayová et al. 2009). However, as we were interested only in clutch protection, we cannot rule out the possibility that the intensity of female responses will increase or even exceed that of males later in the nestling stage (cf. Trnka & Prokop 2010). Despite females being the less responsive sex, their aggressive behaviour increased over a breeding season, perhaps because of a lower renesting potential of late breeding females or greater value of replacement clutches. It is also possible that a proportion of females tested later in a season had already encountered the cuckoo during previous nesting attempts, and were therefore more aggressive to a cuckoo dummy at the end of the season (see also Langmore et al. 2012).

619

Females of a particular social mating status did not compensate for reduced male assistance with nest protection: monogamous, primary and secondary females did not differ in the strength of their response. Similarly, Trnka & Prokop (2010) also found that great reed warbler females defended their nests with the same intensity regardless of their status. This seems somewhat surprising, because to maintain a proper level of nest protection, one would expect more aggressive behaviour in females of polygynous males, especially the secondary ones. At least partial behavioural compensation by the less assisted females, for example through increased nestling provisioning rates or larger/higher quality food delivered may be beneficial to female reproductive success and was reported both in this and other polygynous species (e.g. Pinxten & Eens 1994; Sejberg et al. 2000; Redpath et al. 2006). In our study, however, the females possibly reached an asymptotic level of their nest defence intensity and thus could not respond
more (see Capek et al. 2010 for a similar pattern of aggression in the reed warbler, Acrocephalus scirpaceus). It is also possible that instead of trying to compensate for male aggression, females might rather match the level of male nest defence. Alternatively, they might not need to increase the intensity of their nest defence against the cuckoo because they are able to recognize and reject parasitic eggs (Po
zgayová et al. 2009). However, exploring the relationship between the level of female aggression and egg rejection rates was not the purpose of our study and the relationship is unknown. Despite the fact that the secondary females received the least male assistance with antiparasitic nest defence, we found no evidence that their nests would be the most frequently parasitized. Instead, there were no differences in cuckoo parasitism rates between the primary, secondary and monogamous nests. Our results thus indicate that the host’s social mating system has no effect on the incidence of brood parasitism. Surprisingly, Trnka & Prokop (2011) detected higher levels of cuckoo parasitism in nests of polygynous than monogamous great reed warbler males, suggesting either a lack of male assistance with antiparasitic nest defence or higher activity of hosts in polygynous territories. However, Trnka & Prokop (2011) considered only host nests containing a cuckoo chick (i.e. nests of acceptors) and thus their results did not reflect the original parasitism rates. In contrast, we took into account the initial levels of cuckoo parasitism (i.e. before egg rejection); which most probably explains the discrepancy between the results of the two studies. Yet, our results need not necessarily contradict those of Trnka & Prokop (2011). Although male assistance with cuckoo deterrence does not protect the nests from parasitism, it may still promote further lines of host antiparasitic defences. Increased male aggression might alert the host female to the cuckoo’s presence and thus facilitate rejection of parasitic eggs or chicks (sensu Davies & Brooke 1988; Moksnes et al. 1993; Langmore et al. 2003). However, this is unlikely because egg rejection rates in the great reed warbler do not differ between monogamous, primary and secondary nests (Trnka et al. 2012b) and chick discrimination in this hosteparasite system has not yet been documented. In general, conspicuous sexual behaviour, increased nestbuilding activity or higher density of nests may attract brood parasites (see Parejo & Avilés 2007 and references therein). Polygynous breeding may therefore be disadvantageous for hosts in terms of the elevated risk of parasitism (Trnka & Prokop 2011). In contrast, harems of polygynous males may possess better protection if, theoretically, individual females cooperate in nest guarding or aggressive nest defence. Such group defence by females in polygynous species, however, may be rarely recorded, because the females are known rather to compete with each other (Slagsvold & Lifjeld 1994; Hansson et al. 1997; Trnka et al.

620

M. Po
zgayová et al. / Animal Behaviour 85 (2013) 615e621

2010). The effect of host group size on the incidence of parasitism has been demonstrated only rarely. In the colonial southern red bishop, Euplectes orix, the larger and denser the colony, the lower was the probability of brood parasitism (Lawes & Kirkman 1996; Brown & Lawes 2007). Individuals breeding in large colonies apparently benefit from an increased capacity for group vigilance, which reduces the chances of parasitism through collective detection and deterrence of brood parasites (Brown & Lawes 2007). In cooperatively breeding species, however, there is mixed evidence about the effect of host group size (Brooker & Brooker 1996; Langmore & Kilner 2007; Canestrari et al. 2009; Svagelj et al. 2009). No study has shown that increased group size in cooperative breeders would reduce parasitism risk through communal nest defence. However, Canestrari et al. (2009) suggested that increased host nest attendance may contribute to lower parasitism rates in cooperatively breeding carrion crows, Corvus corone. In our study system, we found relatively high and similar cuckoo parasitism rates in territories of both monogamous and polygynous great reed warbler males. Thus, the issue of whether host group defence against brood parasites is effective in polygynous species deserves further investigation. To sum up, our results showed that polygynous great reed warbler males invest less in antiparasitic nest defence per female than monogamous males, defending their secondary nests less intensively than the primary nests. Females of polygynous males did not compensate for lowered male assistance with nest protection, defending their nests with the same intensity as monogamous females. Reduced male aggression towards the cuckoo at the primary and secondary nests, however, did not result in higher cuckoo parasitism rates at these nests in comparison with their monogamous counterparts. Sharing the antiparasitic nest defence of one mate thus apparently does not represent a cost of polygyny in terms of high parasitism risk on nests of polygynous males. Yet, to explore the general validity of our results, similar studies in other polygynous hosts of brood parasites are highly desirable. Acknowledgments

We are greatly indebted to V. Jelínek, M. Capek, Z. Sebelíková, K.
Morongová, T. Bolcková and M. Sulc for their invaluable assistance in the field. We are also obliged to the management of the Fish Farm Hodonín for permission to conduct the field work in their private area and to local ornithologists for their tolerant approach to our research. Our cordial thanks go to Sami Merilaita and two anonymous referees whose valuable comments and advice substantially improved the manuscript. The study was supported by the Grant Agency of the Academy of Sciences of the Czech Republic (grant numbers IAA600930605 and IAA600930903) and the Institutional Research Plan (RVO: 68081766). References Alatalo, R. V., Carlson, A., Lundberg, A. & Ulfstrand, S. 1981. The conflict between male polygamy and female monogamy: the case of the pied flycatcher Ficedula hypoleuca. American Naturalist, 117, 738e753. Altenburg, W., Daan, S., Starkenburg, J. & Zijlstra, M. 1982. Polygamy in the marsh harrier, Circus aeruginosus: individual variation in hunting performance and number of mates. Behaviour, 79, 272e312. Bártol, I., Karcza, Z., Moskát, C., Røskaft, E. & Kisbenedek, T. 2002. Responses of great reed warblers Acrocephalus arundinaceus to experimental brood parasitism: the effects of a cuckoo Cuculus canorus dummy and egg mimicry. Journal of Avian Biology, 33, 420e425.  , K. 2011. MuMIn: Multi-model inference. R package version 1.3.6. http:// Barton CRAN.R-project.org/package¼MuMIn. Bates, D., Maechler, M. & Dai, B. 2008. Lme4: linear mixed-effects models using S4 classes. R package version 0.999375-28. http://lme4.r-forge.r-project.org/.

Bensch, S. & Hasselquist, D. 1991. Nest predation lowers the polygyny threshold: a new compensation model. American Naturalist, 138, 1297e1306. Bensch, S. & Hasselquist, D. 1994. Higher rate of nest loss among primary than secondary females: infanticide in the great reed warbler? Behavioral Ecology and Sociobiology, 35, 309e317. Bolker, B. M., Brooks, M. E., Clark, C. J., Geange, S. W., Poulsen, J. R., Stevens, M. H. H. & White, J. S. S. 2009. Generalized linear mixed models: a practical guide for ecology and evolution. Trends in Ecology & Evolution, 24, 127e135. Brooker, M. & Brooker, L. 1996. Acceptance by the splendid fairy-wren of parasitism by Horsfield’s bronze-cuckoo: further evidence for evolutionary equilibrium in brood parasitism. Behavioral Ecology, 7, 395e407. Brown, M. & Lawes, M. J. 2007. Colony size and nest density predict the likelihood of parasitism in the colonial southern red bishop Euplectes orixediderick cuckoo Chrysococcyx caprius system. Ibis, 149, 321e327. Burnham, K. P. & Anderson, D. R. 2002. Model Selection and Multimodel Inference: a Practical Information-theoretic Approach. Berlin: Springer-Verlag. Campobello, D. & Sealy, S. G. 2009. Avian brood parasitism in a Mediterranean region: hosts and habitat preferences of common cuckoos Cuculus canorus. Bird Study, 56, 389e400. Campobello, D. & Sealy, S. G. 2011. Nest defence against avian brood parasites is promoted by egg-removal events in a cowbirdehost system. Animal Behaviour, 82, 885e891. Canestrari, D., Marcos, J. M. & Baglione, V. 2009. Cooperative breeding in carrion crows reduces the rate of brood parasitism by great spotted cuckoos. Animal Behaviour, 77, 1337e1344.
Capek, M., Po
zgayová, M., Procházka, P. & Honza, M. 2010. Repeated presentations of the common cuckoo increase nest defense by the Eurasian reed warbler but do not induce it to make recognition errors. Condor, 112, 763e769. Catchpole, C. K., Leisler, B. & Winkler, H. 1985. Polygyny in the great reed warbler, Acrocephalus arundinaceus: a possible case of deception. Behavioral Ecology and Sociobiology, 16, 285e291. Cramp, S. (Ed.). 1992. The Birds of the Western Palearctic, Vol. 6. Oxford: Oxford University Press. Czapka, S. J. & Johnson, L. S. 2000. Consequences of mate sharing for first-mated females in a polygynous songbird, the house wren. Wilson Bulletin, 112, 72e81. Davies, N. B. 1989. Sexual conflict and the polygamy threshold. Animal Behaviour, 38, 226e234. Davies, N. B. 1991. Mating systems. In: Behavioural Ecology: an Evolutionary Approach (Ed. by J. R. Krebs & N. B. Davies), pp. 263e294. 3rd edn. Oxford: Blackwell Scientific. Davies, N. B. 2000. Cuckoos, Cowbirds and other Cheats. London: T. & A. D. Poyser. Davies, N. B. & Brooke, M. de L. 1988. Cuckoos versus reed warblers: adaptations and counteradaptations. Animal Behaviour, 36, 262e284. Dunn, P. O. & Robertson, R. 1993. Extra-pair paternity in polygynous tree swallows. Animal Behaviour, 45, 231e239. Dyrcz, A. 1986. Factors affecting facultative polygyny and breeding results in the great reed warbler (Acrocephalus arundinaceus). Journal of Ornithology, 127, 447e461. Ferretti, V. & Winkler, D. W. 2009. Polygyny in the tree swallow Tachycineta bicolor: a result of the cost of searching for an unmated male. Journal of Avian Biology, 40, 290e295. Forstmeier, W., Kuijper, D. & Leisler, B. 2001. Polygyny in the dusky warbler (Phylloscopus fuscatus): the importance of female qualities. Animal Behaviour, 62, 1097e1108. Griggio, M., Tavecchia, G., Biddau, L. & Mingozzi, T. 2003. Mating strategies in the rock sparrow Petronia petronia: the role of female quality. Ethology Ecology & Evolution, 15, 389e398. Grønstøl, G. B. 2003. Mate-sharing costs in polygynous northern lapwings Vanellus vanellus. Ibis, 145, 203e211. Hannon, S. J. & Dobush, G. 1997. Pairing status of male willow ptarmigan: is polygyny costly to males? Animal Behaviour, 53, 369e380. Hansson, B., Bensch, S. & Hasselquist, D. 1997. Infanticide in great reed warblers: secondary females destroy eggs of primary females. Animal Behaviour, 54, 297e304. Hasselquist, D. 1998. Polygyny in great reed warblers: a long-term study of factors contributing to male fitness. Ecology, 79, 2376e2390. Honza, M., Taborsky, B., Taborsky, M., Teuschl, Y., Vogl, W., Moksnes, A. & Røskaft, E. 2002. Behaviour of female common cuckoos, Cuculus canorus, in the vicinity of host nests before and during egg laying: a radiotelemetry study. Animal Behaviour, 64, 861e868. Janisch, M. 1948. Fight between Cuculus c. canorus L. e cuckoo e and Acrocephalus a. arundinaceus L. e great reed warbler. Aquila, 55, 291. Johnson, L. S. & Albrecht, D. J. 1993. Does the cost of polygyny in house wrens include reduced male assistance in defending offspring? Behavioral Ecology and Sociobiology, 33, 131e136. Johnson, L. S. & Kermott, L. H. 1993. Why is reduced male parental assistance detrimental to the reproductive success of secondary females in the house wren (Troglodytes aedon)? Animal Behaviour, 46, 1111e1120. Johnson, L. S., Kermott, L. H. & Lein, M. R. 1993. The cost of polygyny in the house wren. Animal Ecology, 62, 669e682. Kempenaers, B. 1995. Polygyny in the blue tit: intra- and inter-sexual conflicts. Animal Behaviour, 49, 1047e1064. Kleven, O., Moksnes, A., Røskaft, E., Rudolfsen, G., Stokke, R. G. & Honza, M. 2004. Breeding success of common cuckoos Cuculus canorus parasitizing four sympatric species of Acrocephalus warblers. Journal of Avian Biology, 35, 394e398.

M. Po
zgayová et al. / Animal Behaviour 85 (2013) 615e621 Knight, R. L. & Temple, S. A. 1986a. Why does intensity of avian nest defense increase during the nesting cycle? Auk, 103, 318e327. Knight, R. L. & Temple, S. A. 1986b. Methodological problems in studies of avian nest defence. Animal Behaviour, 34, 561e566. Knight, R. L. & Temple, S. A. 1988. Nest-defense behavior in the red-winged blackbird. Condor, 90, 193e200. Langmore, N. E. & Kilner, R. M. 2007. Breeding site and host selection by Horsfield’s bronze-cuckoos, Chalcites basalis. Animal Behaviour, 74, 995e1004. Langmore, N. E., Hunt, S. & Kilner, R. M. 2003. Escalation of a coevolutionary arms race through host rejection of brood parasitic young. Nature, 422, 157e160. Langmore, N. E., Feeney, W. E., Crowe-Riddell, J., Luan, H., Louwrens, K. M. & Cockburn, A. 2012. Learned recognition of brood parasitic cuckoos in the superb fairy-wren Malurus cyaneus. Behavioral Ecology, 23, 798e805. Lawes, M. J. & Kirkman, S. 1996. Egg recognition and interspecific brood parasitism rates in red bishops (Aves: Ploceidae). Animal Behaviour, 52, 553e563. Leisler, B. & Wink, M. 2000. Frequencies of multiple paternity in three Acrocephalus species (Aves, Sylviidae) with different mating systems (A. palustris, A. arundinaceus, A. paludicola). Ethology Ecology & Evolution, 12, 237e249. Lifjeld, J. T., Slagsvold, T. & Stenmark, G. 1987. Allocation of incubation feeding in a polygynous mating system: a study on pied flycatchers Ficedula hypoleuca. Animal Behaviour, 35, 1663e1669. Ligon, J. D. 1999. The Evolution of Avian Breeding Systems. Oxford: Oxford University Press. Lubjuhn, T., Winkel, W., Epplen, J. T. & Brün, J. 2000. Reproductive success of monogamous and polygynous pied flycatchers (Ficedula hypoleuca). Behavioral Ecology and Sociobiology, 48, 12e17. Moksnes, A., Røskaft, E. & Korsnes, L. 1993. Rejection of cuckoo Cuculus canorus eggs by meadow pipits Anthus pratensis. Behavioral Ecology, 4, 120e127. Møller, A. P. 1986. Mating systems among European passerines: a review. Ibis, 128, 234e250. Molnár, B. 1944. The cuckoo in the Hungarian plain. Aquila, 51, 100e112. Moskát, C. & Honza, M. 2002. European cuckoo Cuculus canorus parasitism and host’s rejection behaviour in a heavily parasitized great reed warbler Acrocephalus arundinaceus population. Ibis, 144, 614e622. Muldal, A. M., Moffatt, J. D. & Robertson, R. J. 1986. Parental care of nestlings by male red-winged blackbirds. Behavioral Ecology and Sociobiology, 19, 105e114. Orians, G. H. 1969. On the evolution of mating systems in birds and mammals. American Naturalist, 103, 589e603. Parejo, D. & Avilés, J. M. 2007. Do avian brood parasites eavesdrop on heterospecific sexual signals revealing host quality? A review of the evidence. Animal Cognition, 10, 81e88. Pearson, T., Whitfield, M. J., Theimer, T. C. & Keim, P. 2006. Polygyny and extrapair paternity in a population of southwestern willow flycatchers. Condor, 108, 571e578. Pilastro, A., Griggio, M., Biddau, L. & Mingozzi, T. 2002. Extrapair paternity as a cost of polygyny in the rock sparrow: behavioural and genetic evidence of the ‘trade-off’ hypothesis. Animal Behaviour, 63, 967e974. Pinxten, R. & Eens, M. 1990. Polygyny in the European starling: effect on female reproductive success. Animal Behaviour, 40, 1035e1047. Pinxten, R. & Eens, M. 1994. Male feeding of nestlings in the facultatively polygynous European starling: allocation patterns and effect on female reproductive success. Behaviour, 129, 113e140. Pinxten, R., Eens, M. & Verheyen, R. F. 1993. Male and female nest attendance during incubation in the facultatively polygynous European starling. Ardea, 81, 125e133. Po
zgayová, M., Procházka, P. & Honza, M. 2009. Sex-specific defence behaviour against brood parasitism in a host with female-only incubation. Behavioural Processes, 81, 34e38. Pribil, S. 2000. Experimental evidence for the cost of polygyny in the red-winged blackbird Agelaius phoeniceus. Behaviour, 137, 1153e1173.

621

R Development Core Team. 2010. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing. http://www.Rproject.org/. Redpath, S. M., Leckie, F. M., Arroyo, B., Amar, A. & Thirgood, S. J. 2006. Compensating for the cost of polygyny in hen harriers Circus cyaneus. Behavioral Ecology and Sociobiology, 60, 386e391. Røskaft, E., Moksnes, A., Stokke, B. G., Bi
cík, V. & Moskát, C. 2002. Aggression towards cuckoo mounts by potential European cuckoo hosts. Behaviour, 139, 613e628. Rothstein, S. I. 1974. Mechanisms of avian egg recognition: possible learned and innate factors. Auk, 91, 796e807. Rothstein, S. I. 1990. A model system for coevolution: avian brood parasitism. Annual Review of Ecology and Systematics, 21, 481e508. Sandell, M. I., Smith, H. G. & Bruun, M. 1996. Paternal care in the European starling, Sturnus vulgaris: nestling provisioning. Behavioral Ecology and Sociobiology, 39, 301e309. Searcy, W. A. & Yasukawa, K. 1989. Alternative models of territorial polygyny in birds. American Naturalist, 134, 323e343. Sejberg, D., Bensch, S. & Hasselquist, D. 2000. Nestling provisioning in polygynous great reed warblers (Acrocephalus arundinaceus): do males bring larger prey to compensate for fewer nest visits? Behavioral Ecology and Sociobiology, 47, 213e219. Slagsvold, T. & Lifjeld, J. T. 1994. Polygyny in birds: the role of competition between females for male parental care. American Naturalist, 143, 59e94. Smith, H. G. & Sandell, M. I. 2005. The starling mating system as an outcome of the sexual conflict. Evolutionary Ecology, 19, 151e165. Smith, H. G., Sandell, M. I. & Bruun, M. 1995. Paternal care in the European starling, Sturnus vulgaris: incubation. Animal Behaviour, 50, 323e331. Svagelj, W. S., Fernández, G. J. & Mermoz, M. E. 2009. Effects of nest-site characteristics and parental activity on cowbird parasitism and nest predation in brown-and-yellow marshbirds. Journal of Field Ornithology, 80, 9e18. Trnka, A. & Prokop, P. 2010. Does social mating system influence nest defence behaviour in great reed warbler Acrocephalus arundinaceus males? Ethology, 116, 1075e1083. Trnka, A. & Prokop, P. 2011. Polygynous great reed warblers Acrocephalus arundinaceus suffer more cuckoo Cuculus canorus parasitism than monogamous pairs. Journal of Avian Biology, 42, 192e195. Trnka, A. & Prokop, P. 2012. The effectiveness of hawk mimicry in protecting cuckoos from aggressive hosts. Animal Behaviour, 83, 263e268. Trnka, A., Prokop, P. & Batáry, P. 2010. Infanticide or interference: does the great reed warbler selectively destroy eggs? Annales Zoologici Fennici, 47, 272e277. Trnka, A., Prokop, P. & Grim, T. 2012a. Uncovering dangerous cheats: how do avian hosts recognize adult brood parasites? PLoS ONE, 7, e37445. Trnka, A., Po
zgayová, M., Procházka, P., Prokop, P. & Honza, M. 2012b. Breeding success of a brood parasite is associated with social mating system of its host. Behavioral Ecology and Sociobiology, 66, 1187e1194. Urano, E. 1990. Factors affecting the cost of polygynous breeding for female great reed warblers Acrocephalus arundinaceus. Ibis, 132, 584e594. Verner, J. 1964. Evolution of polygamy in the long-billed marsh wren. Evolution, 18, 252e261. Verner, J. & Willson, M. F. 1966. The influence of habitats on mating systems of the North American passerine birds. Ecology, 47, 143e147. Weatherhead, P. J. 1990. Nest defence as shareable paternal care in red-winged blackbirds. Animal Behaviour, 39, 1173e1178. Yasukawa, K., McClure, J. L., Boley, R. A. & Zanocco, J. 1990. Provisioning of nestlings by male and female red-winged blackbirds, Agelaius phoeniceus. Animal Behaviour, 40, 153e166. Yasukawa, K., Whittenberger, L. K. & Nielson, T. A. 1992. Anti-predator vigilance in the red-winged blackbird, Agelaius phoeniceus: do males act as sentinels? Animal Behaviour, 43, 961e969.