Research articlesHost nest defence does not act as selective agent against plumage polymorphism in brood parasites
Abstract
Batesian mimicry in brood parasites is often viewed as an evolutionary strategy to mitigate host aggression. Female common cuckoos (Cuculus canorus) exhibit two morphs: the hawk-like grey and the rufous one, potentially maintained by apostatic selection. It was hypothesized that the grey morph’s predator-like appearance deters host defences, while the rufous morph benefits from its rarity by evading host attention. Previous research predominantly utilized static cuckoo dummies, lacking insights into real-world interactions. We investigated the effectiveness of the cuckoo morphs in accessing great reed warbler (Acrocephalus arundinaceus) nests under natural conditions. Analysing video-recorded cuckoo attempts, we found no significant difference in nest-access success between the morphs. Both experienced a similar probability of physical attacks when hosts were present, and the rufous morph did not evade host detection more often compared with the grey morph. These results fail to support the assumptions of (a) Batesian mimicry, that hawk-like mimicry enhances nest access or reduces host aggression, and (b) apostatic selection, that the rarity of the rufous morph confers an advantage in successfully accessing the host nest. Future research should aim to identify stages in the cuckoo’s life cycle or host interactions where colour polymorphism provides an evolutionary benefit.
1. Introduction
The traditional view of parental care in the animal kingdom assumes that parents care for their own young, but this is challenged by interspecific brood parasitism, a phenomenon where one species raises the offspring of another species. In invertebrates, it manifests in various insect taxa through social parasitism or kleptoparasitism, notably among wasps [1], ants [2] and bees [3]. In vertebrates, such examples include one fish species [4] and some species of birds [5]. Among birds, obligate interspecific brood parasitism has evolved independently in seven different clades, representing approximately 1% of bird species globally [6].
Interestingly, some of the intensively studied obligate brood parasites belonging to the order Cuculiformes, along with Strigiformes, Galliformes and Ciconiiformes, exhibit significant plumage colour polymorphism [7]. These colour variations can be adaptive because the morphs differ in life-history traits like mating strategies, crypsis and avoidance of prey or predators [7,8]. Multiple hypotheses, including frequency-dependent selection, disruptive selection and non-random mating, have been proposed to explain the mechanisms, functions and benefits of this phenomenon [7]. In the case of Cuculus cuckoos, plumage colour polymorphism is mapped to the female-restricted genome and is maintained by apostatic selection as a special case of frequency-dependent selection and mimicry dynamics [9,10]. Moreover, it was proposed that it serves as a strategic adaptation to increase survival and reproductive success of cuckoos by deceiving host recognition and thus avoiding mobbing [11–13] or to escape the costly male harassment directed at the more common cuckoo female morph [14,15].
Brood parasitism has detrimental effects on host fitness, positioning it as a significant evolutionary pressure that drives the development of host defence strategies [5]. One of the most effective host defences against this negative effect is building nests in places that minimize or eliminate the risk of parasitism [16–19]. Additionally, shifts in the timing of breeding between parasites and hosts can lead to laying asynchrony, reducing the chances of successful parasitism [20,21]. On the other hand, synchronous breeding can dilute the risk of being parasitized [22–24]. Other effective defence mechanisms include blocking access to the nest cup by the obstruction of the nest entrance with the host body [21] and the host firmly sitting on the nest cup [25]. In addition, hosts may resort to aggressive nest protection, often characterized by loud alarm calls and direct physical attacks, as a means to prevent parasitism. Such a strategy is considered adaptive from the host perspective [26,27] but can pose significant risks to parasites, as evidenced by instances of dead parasites found near host nests [28–31]. The aggressive behaviour is a well documented self-defence mechanism across vertebrates, including fish [32], birds [33] and mammals [34,35]. In the context of brood parasitism, various studies have suggested that mobbing is a phenotypically plastic behaviour, although these studies were based mainly on model-presentation experiments [27]. Host aggression intensity may fluctuate in response to factors such as brood parasite density [27,36], the presence of specific cuckoo morphs at a host breeding site [12] and host social status [37].
The effectiveness of host aggressive behaviour as a frontline defence mechanism, particularly under natural conditions, has been little studied owing to its challenging nature [26]. To date, research on host–parasite interactions has predominantly focused on direct observations at the nests by using continuous video recording [29,38–41]. To our knowledge, only Gloag et al. [29] filmed host mobbing efficiency also around the nest, albeit based on a limited sample size.
This study aims to examine the effectiveness of the great reed warbler (Acrocephalus arundinaceus, hereafter: GRW) in defending against the two colour morphs of its brood parasite, the common cuckoo (Cuculus canorus, hereafter: cuckoo). Our study is novel in that it investigates host–parasite interactions under natural conditions, i.e. before the cuckoo reaches the host nest. Unlike previous studies that used static cuckoo dummies with varying degrees of similarity to cuckoos, we assess host reactions to real-life cuckoos in real-world settings. The grey morph of the cuckoo, an example of Batesian mimicry, mimics the harmful Eurasian sparrowhawk (Accipiter nisus, hereafter: sparrowhawk), which is hypothesized to reduce host mobbing intensity and facilitate cuckoo access to host nests [42,43]. The rarer rufous morph is thought to exploit its low frequency in the population, potentially evading host defences by deceiving host recognition mechanisms [10,12,44]. In our study plot, the rufous morph was rarer over the last two decades, with grey/rufous ratio ranging between 9:1 and 3:1 ([12,45], and in this study).
Under the Batesian mimicry hypothesis, we predict that the grey morph will experience lower probability of contact attacks compared with the rufous morph. Under the apostatic selection hypothesis, we predict that the relative probability of attempt events will shift in favour of the rufous morph at host nests where parents are absent compared with the nests with parents present. Additionally, we explored in more detail the effectivity of nest defence against the two cuckoo colour morphs, with respect to host presence and their number, daytime, date in the season and clutch size.
2. Methods
(a) Study species and area
GRW, weighing approximately 30 g, is an ideal candidate for this study owing to its size, making it one of the cuckoo's largest and most preferred hosts [46]. Parasitism rates in this species vary across its breeding range, for example, 0% in central Greece [47], 5−39% in the southeast of the Czech Republic [48,49], with a marked increase (up to 90%) in recent years [50], and 66% in Hungary [17,51]. GRW is known for its strong mobbing response to cuckoo dummies, with sex-specific variability and consistency across different breeding seasons [37,52–54]. Jelínek et al. [39] observed that if GRWs detected live cuckoos attempting to parasitize their nests, they almost invariably attacked them (91 of 104 cases; 88%).
Data for this study were collected during the breeding seasons 2020–2022 within the fishpond area situated between Hodonín (48°51′0″ N, 17°07′0″ E) and Mutěnice (48°54′0″ N, 17°02′0″ E), in the Czech Republic. The studied GRW population was individually marked, with 65, 55 and 30 breeding pairs in the respective years, experiencing high rates of brood parasitism (78% in 2020, 53% in 2021 and 79% in 2022). GRW nests were found through regular mapping of male territories and systematic searches of littoral vegetation.
(b) Video-recording of great reed warbler nests
We employed a video-recording setup of one camera focused on the nest from a close distance of <1 m and another camera installed on a tripod approximately 4−10 m from the nest, monitoring the nest surroundings. Most of the nests were video-recorded from the day the nest was completed throughout the egg-laying period. In a minority of nests, the filming started any time during the egg-laying period.
The closer-positioned camera was a Carmedien STO-IR rear-view camera (Carmedien, Tschernitz, Germany) with infrared illumination connected to a miniature Mini DVR CH-HD0065 digital video recorder (Shenzhen Chu-Tech Co., Shenzhen, China) placed in a water-resistant box, or a custom HD camera with inbuilt recorders without infrared illumination. The cameras were powered by 12 V/100 Ah gel batteries placed on the bank. In 2021, these cameras were replaced with motion-sensitive BUNATY micro 4K trail cameras powered by AA batteries and placed about 1 m from the nest.
The camera capturing the wider area around the nests was a Sony HDR-CX 240 powered by a power bank (GoGEN 20 000 mAh) with a scanning range of about 10 m from the nest on each side, again depending on reed density. Daily checks were conducted to record the nest status and replace the power bank and secure digital memory card. All the recording equipment was camouflaged by green cloth and reed stems to prevent disturbance of nesting birds. The use of simultaneous filming allowed us to analyse cuckoo–host interactions around the nest and at the nest, determining the outcome of cuckoo activity (unsuccessful access or successful access to the nest; and in the latter case, further parasitism, predation or only nest-checking event). We considered nest access successful when a cuckoo made physical contact with the host nest. The location of the nest in the reeds sometimes prevented us from detecting cuckoo activity using a more distant camera, so we used only recordings from a closer camera that captured the nest. This setup, initiated in 2016, has previously been assessed as having no negative effect on GRW nest success [39].
(c) Data analyses
Reviewing all the video recordings would require enormous effort. To speed up the search process for interactions between the host and cuckoo, we (i) converted the recordings from .mts video format to the lossless .flac mono channel audio format (using the FFmpeg multimedia framework) and (ii) created full-record length spectrograms with timestamp in .svg file format using Sonic Visualizer 4.3 software. The spectrograms were carefully examined for distinct patterns representing GRW alarm and distress calls, which were clearly recognizable from other environmental sounds (figure 1). The spectrograms were manually reviewed by three experienced researchers (P.S., G.Š., M.P.). Manual search proved to be the most effective and error-free method for recognizing GRW alarms.
Figure 1. Example of spectrogram patterns used for identification of great reed warbler (GRW) alarm and distress call events. The searched vocalizations were characterized by dense, vertical lines that are closely spaced. The lines are uniform in appearance, with little variation in note shape, and frequency being markedly distinct from any other sound detected at the research locality, such as GRW song.
After identifying the search patterns, we recorded the timestamp and video recording ID. We then revisited the specific moment in the original video to observe whether the cuckoo successfully accessed the host nests or not and whether it was physically attacked. For each event, we recorded the cuckoo morph (grey or rufous), the number of host parents present at the nest and whether they mobbed or also physically attacked the cuckoo, the time of day and the date within the breeding season. From the nest cards, we also recorded the clutch size at the time of the event.
When considering direct physical attacks by GRWs, we had to exclude some cases from the analysis owing to environmental obstructions, primarily dense reed coverage. In these cases, while we could confirm cuckoo mobbing based on GRW alarm calls, we could not definitively determine whether physical attacks occurred.
We further monitored the fate of eggs at all recorded nests by daily checks recorded in the nest cards—if any change occurred, e.g. one or more host eggs disappeared or a cuckoo egg appeared at a nest and no event was detected by spectrogram analysis (because the hosts were not present at the time of the event), we viewed all relevant videos between the two nest checks. If a cuckoo visit occurred, we added the event to the database.
(d) Statistical analyses
In 2020, 2021 and 2022, we filmed a total of 72 nests, representing 3163 h, and recorded 126 cases of cuckoos trying to access host nests. Six of these cases were discarded because we only recorded the cuckoo’s departure, not its arrival (cuckoos sometimes flew low from the bank and could not be seen through the reedbed) and in a further six cases, we assumed the intruder was a cuckoo based on host behaviour but were unable to reliably identify it. These 12 excluded cases represent only 10% of the total sample, demonstrating the high effectiveness of our video setup in capturing cuckoo activity. From the remaining 114 cases with confirmed cuckoo presence, we identified and omitted four instances of cuckoos flying above the reeds and passing the nest from a distance without any clear association or attempt to access a host nest and three cases where the cuckoo morph was indistinguishable because the direct sunlight overexposed the video. In six other cases, the camera failed to record the time of the day. Finally, in five cases, the nest was already abandoned at the time of the cuckoo’s visit, which was confirmed by video recordings from a camera close to the nest and daily nest checks by a researcher. All these cases were omitted from the final analysis. Consequently, 96 attempts (64 by grey and 32 by rufous morphs) to access 46 host nests were included in our final analysis (table 1). The number of cuckoo visits at each host nest recorded was 1 at 21 nests, 2 at 13 nests, 3 at 7 nests, 4 at 2 nests, 5 at 2 nests and 10 at 1 nest. When a cuckoo successfully accessed a GRW nest (n = 66), we further classified its activity at the nest thanks to cameras focused directly at the nest and daily nests checks as follows: parasitism (if it laid a cuckoo egg in the nest, n = 56), predation (if it only took one or more GRW eggs from the nest but did not lay its own, n = 3) and visit only (if it only visited the nest, but did not take or lay an egg, n = 7). However, we did not further explore the mentioned cuckoo activities at the host nests as we were not interested in cuckoo’s activity after reaching the host nest in this study, and it has already been investigated in our study population [39]. To address the possibility of overlooking instances where hosts might behave quietly, potentially causing us to miss cuckoo nest access attempts, we conducted a detailed visual analysis of a subset of full-length video recordings. We examined 16 nests, totalling 12 555 min of continuous footage, at standard playback speed. This thorough review confirmed that in all cases where hosts were present, their behaviour invariably included mobbing actions accompanied by distinct alarm calls.
| host presence | cuckoo morph | no. attempts | no. nests | no. successful attempts (%) |
|---|---|---|---|---|
| yes | grey | 44 | 21 | 21 (48) |
| yes | rufous | 23 | 19 | 16 (70) |
| no | grey | 20 | 16 | 20 (100) |
| no | rufous | 9 | 9 | 9 (100) |
First, we examined if the hosts feared the hawk-like grey cuckoo more than the rufous morph. We specifically tested if the probability of physical attack by host (binary response variable: yes or no) differed between cuckoo morphs (categorical: grey or rufous). In this model, we additionally tested the effect of the number of host parents present (1 or 2), the clutch size at the time of event (0−5), the time of day (hour) and the ordinal date within the breeding season (1 = 1 January). To account for repeated cuckoo visits to the same nest, we included a random intercept for nest ID in our model. We assessed potential multicollinearity using the variance inflation factor (VIF), which was satisfactory for all predictors (VIF < 2.0). We employed a Bayesian model with Bernoulli distribution and logit link, which effectively solved the problem by setting a weakly informative prior for fixed effects to be the normal distribution of the mean set to 0 and the standard deviation to 10. For this purpose, we used the package brms (v. 2.20.4) [55] in R software (v. 4.3.2) [56]. We ran 5 × 103 iterations with a burn-in phase of 1000 to obtain more than 3000 effective samples per parameter for posterior inference. After fitting, we evaluated the model using posterior predictive checks and inspecting the potential scale reduction factor (Ȓ) for all parameters. Ȓ in this fit was 1.00, suggesting a good convergence of the fit. We computed 89% credible intervals (CrI), which are assumed to be more stable than 95% CrI if the effective sample size for a parameter is <10 000 [57]. However, re-calculating with 95% CrI intervals did not change the conclusions. Predictor effects were tested using CrI intervals, where intervals not overlapping zero suggested a significant difference. For posterior inferences, we used the package emmeans (v. 1.9.0) [58].
Next, we tested whether the rarer rufous morph evades the host’s attention more often compared with the grey morph. This should be indicated by a higher proportion of rufous morph observed at the nests where hosts are absent, compared with the nests with hosts present. Again, we employed a Bayesian mixed-effects model with Bernoulli distribution with response variable of cuckoo morph and predictor of the presence of host parents during a cuckoo visit (at least one host present or hosts absent). The additional predictors and model structure were identical to the previous model. Furthermore, we aimed to identify the factors that influence the success of cuckoos in accessing host nests (binary response variable: yes or no). We primarily examined the potential effects of the presence of host parents during a cuckoo visit (yes or no), and effect of cuckoo morph (grey or rufous) and then included the same additional predictors as in the above models. In this model, there was a problem with a complete separation (all the cuckoos successfully reached the nest if no host was present; n = 29). We effectively solved this problem by employing a Bayesian model and setting a weakly informative prior for fixed effects to be the normal distribution of the mean set to 0 and the standard deviation to 2.
Finally, we used the same Bayesian mixed-effects model to explore the effect of the number of host parents present (1 or 2) on the success of cuckoos to access host nests (binary: yes or no). In this model, we excluded all cases where no parent was present during the cuckoo attempts and applied the same model structure and settings as in the previous model. The only exception was setting a weakly informative prior again to the mean of 0 and standard deviation of 10.
3. Results
Out of the 96 cuckoo attempts, we recorded 29 cases on 25 nests where hosts were absent (table 1). In all of these cases, the cuckoo reached the nest. When hosts were present (n = 67), they always mobbed the cuckoo and emitted alarm or distress calls, and in some cases physically attacked it. In 55% of these cases, the cuckoo successfully reached the nest. However, this summary did not consider other factors such as repeated approaching attempts at nests (table 1). We could confidently ascertain that a cuckoo was physically attacked by the hosts in 32 out of 50 cases (64%). The grey cuckoo morph was physically attacked in 20 out of 33 instances (61%), comparable to the rufous morph, which was attacked in 12 out of 17 instances (71%).
The Bayesian model with a Bernoulli distribution showed that hosts did not fear the hawk-like grey cuckoo more than the rufous morph and both morphs were physically attacked at similar rates (log odds [89% CrI]: 2.90 [−7.39 to 14.11]). Notably, the frequency of attacks on cuckoos significantly diminished as the breeding season progressed (log odds: −1.21 [−3.29 to −0.09]; figure 2).
Figure 2. Predicted probability of host attacks on cuckoos in relation to date within the breeding season and cuckoo morph. Points represent raw data, colour-coded by cuckoo morph, and lines are predicted from a Bayesian mixed-effects model with a Bernoulli distribution.
The model testing the apostatic selection hypothesis, which predicts that the rarer rufous morph should more often slip past the hosts' attention (i.e. higher proportion of rufous morph observed when hosts are absent), showed that morph ratios remained similar regardless of host presence or absence at the nest (log odds: −0.09 [−1.60 to 1.35]; figure 3). In the earlier hours, there was a higher probability of encountering the grey cuckoo morph. As the day progressed, this probability decreased, with both morphs being encountered with similar frequency in the evening (log odds: 0.69 [0.23 to 1.25]; figure 3).
Figure 3. Probability of detecting a cuckoo morph at host nests (1.00 = grey morph, 0 = rufous morph) in relation to time of day and host presence. Points represent raw data, colour-coded by host parents presence status, with lines predicted from a Bayesian mixed-effects model with a Bernoulli distribution.
The Bayesian model confirmed that the overall success of cuckoos in accessing host nests significantly decreased when hosts were present at the nest (log odds [89% CrI]: −3.50 [−5.78 to −1.30]), while the success increased with the advancing time of the day (log odds: 0.69 [0.23 to 1.25]; figure 4). Again, we did not find a significant effect of different cuckoo colour morphs on success in accessing host nests (log odds: 0.32 [−1.91 to 2.61]).
Figure 4. Predicted probability of cuckoos successfully accessing host nests in relation to time of day and host presence. Points represent raw data, colour-coded by host parents presence status, with lines predicted from a Bayesian mixed-effects model with a Bernoulli distribution.
Upon excluding events without host presence, the model detected no effect regarding the number of hosts (1 or 2) and cuckoo colour morphs on the success of cuckoos in accessing host nests. However, success rates again increased later in the day (log odds: 6.17 [1.46 to 5.85]; figure 5).
Figure 5. Predicted probability of cuckoos successfully accessing host nests in relation to time of day and number of hosts present. Points represent raw data, colour-coded by number of host parents present, and lines are predicted from a Bayesian mixed-effects model with a Bernoulli distribution.
4. Discussion
Our examination of host defence against real-life cuckoos did not support the hypothesis that Batesian mimicry or apostatic selection plays a significant role during the critical moment of parasitism, which is successful access to the host nest. Both cuckoo morphs faced similar probabilities of physical attack when hosts were present at the nest. Additionally, we did not observe an increase in proportion of rufous morphs at host nests when hosts were absent compared with when hosts were present.
(a) Is mobbing an effective component of host frontline defences?
Our study for the first time quantified the GRW effectivity of frontline defence [59] in natural conditions realized in the vicinity of the host nest and showed that host mobbing can, in some cases, deter cuckoos from accessing host nests. GRWs successfully drove cuckoos away in 45% of all cases. However, a closer examination reveals that 87% of the 46 nests were ultimately parasitized. If we consider only parasitism events when at least one host was present and could defend its nest, our results show approximately 21% effectiveness of individual GRW pairs in defending their nests.
When comparing these results with previous studies that assessed the effectiveness of host defence directly at the nests, we found that the host frontline defence is not effective. Jelínek et al. [39] observed in the same GRW population that when hosts detected a cuckoo at the nest, they almost always attacked it (91 of 104 cases; 88%); however, the attacks prevented the cuckoo from laying only in 10% of cases. Similarly, Mikulica & Trnka [60] observed that GRW hosts attacked cuckoo females in 74.3% of encounters. However, these attacks rarely deterred the cuckoos from laying their eggs in the host nests, with successful prevention occurring in only 11.5% of cases. A similarly low effectiveness of mobbing was reported by Gloag et al. [29], where the chalk-browed mockingbird (Mimus saturninus) managed to prevent parasitism by the shiny cowbird (Molothrus bonariensis) in at most 15% of cases. Likewise, attacks by the magpie (Pica pica) against the great spotted cuckoo (Clamator glandarius) females never effectively prevented parasitism [61]. Welbergen & Davies [27] on the other hand reported that mobbing of the cuckoo by the smaller Eurasian reed warbler (Acrocephalus scirpaceus; hereafter: reed warbler) was associated with a lower likelihood of parasitism, suggesting that mobbing can indeed be effective. Similarly, Fiorini et al. [62] observed that mockingbirds with unparasitized nests exhibited more aggression towards model cowbirds compared with those with parasitized nests, indicating that mobbing may reduce the risk of parasitism.
Previous studies on the effectiveness of host aggression against brood parasites have reported inconsistent conclusions. While our findings suggest that even high levels of aggression do not prevent cuckoos from successfully accessing host nests in GRWs, this may not be the case for all host species. The discrepancies in results could potentially be attributed to differences in methodological approaches, which may influence host behaviour and the interpretation of results. For instance, studies using dummies may yield different outcomes compared with observations in natural conditions. These inconsistencies highlight a significant knowledge gap in our understanding of host–parasite interactions. They underscore the need for further research in this field, with a particular emphasis on studies conducted under real-life conditions. Such studies could provide more accurate insights into the true effectiveness of host defences against brood parasites across different species and ecological contexts.
(b) Do hosts consider cuckoo hawk-like mimicry to be a threat?
Over a century ago, Wallace [63] suggested that mimicry of hawks by parasitic cuckoos may have evolved as a form of protective mimicry, potentially reducing attacks from hawks. Subsequent research has extensively explored this concept within the framework of Batesian mimicry, yielding mixed results. The Batesian mimicry hypothesis suggests that when cuckoos mimic hawks, nest defence is more costly for hosts because they need to extra-evaluate the potential risk to their survival [64]. Indeed, studies by Welbergen & Davies [42] and Davies & Welbergen [43] indicated that the artificial dummy of the cuckoo grey morph, which mimics the sparrowhawk, experienced reduced mobbing by the reed warbler hosts. However, experimental research on GRWs and other more aggressive hosts suggested that this trait offers cuckoos minimal protection from host contact attacks because the hosts are able to distinguish cuckoos from sparrowhawks [65,66] (see also [41,67] for similar results in Oriental reed warblers (Acrocephalus orientalis) and fork-tailed drongos (Dicrurus adsimilis), respectively) but GRWs cannot discriminate between kestrel and rufous cuckoo [66]. The hypothesis of rufous cuckoo mimicry of kestrels has been questioned by several studies, primarily owing to significant morphological differences between the two species [68]. Notably, kestrels lack the barred plumage and yellow eyes characteristic of rufous cuckoos, which are the most important features used by hosts in recognizing brood parasites [43,66,69]. However, even if rufous cuckoos would not mimic kestrels, they still should have an advantage because of being rarer and distinct from grey cuckoos. The hosts should then be less adept at recognizing the less common morph [68].
Honza et al. [12] also reported no difference in aggression towards grey and rufous artificial cuckoo dummies in our study area almost 20 years ago. Interestingly, the frequency of rufous morph seems to have slightly increased at our study site over the last decade, from 11 to 20% ([45] and this study). The dynamics in ratio of the two colour morphs in other parts of their breeding period or life aspects could potentially explain this slight increase in rufous morph frequency at our site. For example, the difference could be pronounced in the perfection of cuckoo egg mimicry and subsequent host recognition and rejection, or in variations in the survival rates of adults or fledglings. When considering contact attacks as the peak of aggression in mobbing escalation, Trnka & Grim [66] found that GRW vigorously attacked both the grey and rufous artificial cuckoo dummies. Host aggression intensity was 1.7 times higher towards the grey cuckoo than towards the rufous, and this ratio was similar to the ratio of the cuckoo colour morphs occurring in their study population. However, our findings did not support their conclusion that host aggression reflects the occurrence ratio of the cuckoo colour morphs in a population. Such a discrepancy could be caused by methodological differences like real-world versus experimental settings and evaluating aggressive behaviour using different metrics.
Similar aggression to both cuckoo morphs in our study could be alternatively explained by heavy parasitism rate in our population [50] leading to frequent encounters with the cuckoo. Such repeated stimulation can also lead to over-aggression towards any intruder, including brood parasites regardless of colour morph. Indeed, we observed that GRWs alarm-called if other species approached their nest, such as harmless coots (Fulica atra), common moorhens (Gallinula chloropus) and little grebes (Tachybaptus ruficollis), and more threatening species like the grey heron (Ardea cinerea) and little bittern (Ixobrychus minutus). Thus, the lack of enemy-specific recognition (similar attack ratios on grey versus rufous morphs) could be indicative of a generalized nest defence strategy, where GRW vehemently reacts to any sudden sound or movement around the nest.
(c) Which factors influence the success of reaching the nest?
Our analyses showed that the presence of hosts at the nest significantly reduced the cuckoo’s success in accessing the host nest regardless of the presence of one or both host parents. This is corroborated by Požgayová et al. [37], who indirectly supported our findings by reporting no differences in cuckoo parasitism rates between monogamous, primary and secondary GRW females, even though the monogamous ones received more assistance from males. Interestingly, Wang et al. [40] found that the presence of two or fewer individuals was largely ineffective in preventing cuckoo egg-laying in the nest of the closely related Oriental reed warbler. Unlike Wang et al. [40], we found no cooperative nest defence from neighbouring birds in our studied population.
We, however, showed that cuckoo success in reaching the nest increased as the day progressed, irrespective of parental presence, contradicting the notion that cuckoos lay eggs in the afternoon to avoid detection by hosts [70]. Moksnes et al. [38] also found little support for the hypothesis that cuckoos lay in the afternoon because they are less likely to be detected or attacked by the nest owners. This pattern could be explained for example by reduction of intensity of host aggression and, therefore, also effectivity of host defence in the evening. Alternatively, our results suggest that the timing of parasitism may be driven by factors other than host presence, possibly reflecting an intrinsic motivation of cuckoos to lay eggs as the day advances. Also, other environmental factors such as weather conditions or reed density (related to nest concealment) might theoretically affect the cuckoo’s success. The influence of these passive environmental factors on host defensive behaviour in our study site is likely limited because of relatively stable weather across the years and within a breeding season and similar vegetation structure around nests.
In conclusion, our results suggest that host mobbing in the nest vicinity can be effective in driving away individual cuckoos during specific encounters and the presence of hosts decreases the chance of cuckoo success. However, when considering the fate of individual nests throughout the entire breeding season, this frontline defence proves insufficient to fully protect host nests against parasitism, even in a highly aggressive host. Importantly, we showed that both cuckoo colour morphs were physically attacked with similar probability, and neither of them was more successful in accessing the host nests. This suggests neither that the hawk’s disguise of the grey morph increased protection from host attacks nor that being the rarer rufous morph helped them slip undetected, which does not support the hypotheses of Batesian mimicry and apostatic selection. It is important to emphasize that our results stem from a single population, potentially limiting the broader applicability of our conclusions. To comprehensively address the questions posed in our predictions, future research should extend beyond our study population and species, encompassing diverse host species experiencing varying levels of parasitism and cuckoo populations with different morph ratios. Such studies would offer a more nuanced understanding of the intricate dynamics between cuckoo colour polymorphism and host defences across varied ecological contexts. Notably, determining the specific phases of the cuckoo’s life cycle or host interactions when colour polymorphism confers an evolutionary advantage remains a formidable challenge in this field.
Ethics
This research has been done under permit number JMK 38506/2016 and complies with the ‘Association of Animal Behaviour’, the Animal Behavior Society guidelines for the treatment of animals in research and the current laws and ethical guidelines of the Czech Republic.
Data accessibility
Data are deposited in the Dryad repository [71].
Declaration of AI use
We have not used AI-assisted technologies in creating this article.
Authors’ contributions
M.H.: conceptualization, data curation, funding acquisition, methodology, supervision, writing—original draft, writing—review and editing; G.Š.: conceptualization, data curation, methodology, writing—review and editing; M.P.: conceptualization, data curation, methodology, writing—review and editing; P.S.: conceptualization, data curation, formal analysis, methodology, visualization, writing—review and editing.
All authors gave final approval for publication and agreed to be held accountable for the work performed herein.
Conflict of interest declaration
We declare we have no competing interests.
Funding
This work was funded by the Czech Science Foundation (grant no. 22-26812S).
Acknowledgements
We thank V. Audrlický, V. Brlík, P. Horák, P. Procházka and J. Studecký for their assistance in the field. Finally, we are grateful to the managers of the Fish Farm Hodonín and local conservation authorities for permission to conduct the fieldwork.
