Egg rejection and egg recognition mechanism in a Chinese Azure-winged Magpie (Cyanopica cyanus) population

1.   Introduction

Rejection of parasite eggs is an effective defense mechanism of bird hosts against brood parasitism (Davies, 2000; Soler, 2014; Ma and Liang, 2021; Ye et al., 2022a, 2022b). Accurate rejection of parasitic eggs requires recognition of the host eggs. Visual cues (Ruiz-Raya and Soler, 2020; Samaš et al., 2021), tactile cues (Meshcheryagina et al., 2016; Ye et al., 2022b) and olfactory cues (Soler et al., 2014; Golüke et al., 2016; Leclaire et al., 2017) are all used by hosts to identify foreign eggs. Recognition may occur through two different mechanisms (Feeney et al., 2014; Manna et al., 2017): true recognition, that is, the ability to compare the eggs in the nest to an internal template of the appearance of the host's own eggs, and discordancy, that is, the ability to identify an odd egg, or eggs, in the clutch. If the egg recognition mechanism of the hosts is true recognition, they reject the foreign eggs whether these are as numerous as their own eggs, outnumber their own eggs, or are the only egg type present (Rothstein, 1975; Lyon, 2007; Yi et al., 2020; Yang et al., 2022). If the egg recognition mechanism of the hosts is discordancy recognition, the small number of eggs that are different from the most eggs are rejected (Moskát et al., 2010, 2014; Bán et al., 2013; Yang et al., 2014; Tosi-Germán et al., 2020). However, only some hosts can readily recognize their own eggs.

Differences in selection pressures in the coevolutionary process between hosts and cuckoos in the same region have resulted in varied egg recognition abilities among host species (Begum et al., 2012; Liu et al., 2020). For example, there were significant differences in the egg recognition ability among the four hosts of different cuckoo species: Common Mynas (Acridotheres tristis), Jungle Babblers (Turdoides striata), Long-tailed Shrikes (Lanius schach) and House Crows (Corvus splendens). Common Mynas and Jungle Babblers do not possess egg recognition ability and accept all mimetic eggs. House Crows show low egg recognition ability, rejecting 9.1% of the mimetic eggs, while Long-tailed Shrikes have egg recognition ability and reject 75% of the mimetic eggs (Begum et al., 2012). The differences in coevolution and interaction pressure between a host species and its parasitic cuckoos may also lead to significant differences in egg rejection among geographic populations of the same host species (Yang et al., 2015; Liang et al., 2016). In southern Spain where there is a high risk of parasitism by Great Spotted Cuckoos (Clamator glandarius), the rejection rate of the Eurasian Magpie (Pica pica) toward non-mimetic eggs is close to 90% (Soler et al., 2015). In contrast, in China, where the Eurasian Magpie is not parasitized by any cuckoo species, the acceptance rate of experimental non-mimetic parasite eggs was 100% (Yang et al., 2021). Great Tits (Parus major) in Europe do not have a coevolutionary relationship with cuckoos (Davies, 2000), and they do not have the ability to recognize eggs. In China, Great Tits and Green-backed Tits (Parus monticolus) may have a strong coevolutionary interaction with cuckoos, because they all have a high egg recognition capability (Liang et al., 2016; Liu et al., 2019, 2020).

The Azure-winged Magpie (Cyanopica cyanus) is a passerine species (Zheng, 2021) commonly found in the eastern Palearctic (Madge, 2020; Madge and Juana, 2020). Other than Great Spotted Cuckoos of Spain (Erritzøe et al., 2012), the Common Cuckoo (Cuculus canorus) is the only parasite species in most regions of Europe. In the past, Azure-winged Magpies in Iberia and China were thought to be the same species, but because they have had an independent evolutionary history for a long time, they are now considered as two species, Azure-winged Magpie (Cyanopica cyanus) and Iberian Magpie (C. cooki) (Zheng, 2021). Although Iberian Magpies are not parasitized by Great Spotted Cuckoos in Europe, it has developed strong egg recognition ability. This suggests that the egg recognition ability of Iberian Magpies has evolved from intra-specific (con-specific) and not inter-specific brood parasitism (Avilés, 2004). In Japan, no egg recognition ability has been observed in Azure-winged Magpies when they are not parasitized by Common Cuckoos. Within 10–20 years of the Common Cuckoo parasitism, they rapidly evolved the ability to recognize and reject foreign eggs (Takasu et al., 1993; Nakamura et al., 1998). In South Korea, Azure-winged Magpies are rarely hosts of Indian Cuckoos (Cuculus micropterus) but they possess strong egg recognition ability. Son et al. (2015) believe that Korean Azure-winged Magpies migrated from China, and retained the egg recognition ability of their founders. In China, Azure-winged Magpies are parasitized by both Indian Cuckoos and Asian Koels (Eudynamyss scolopaceus) (Yang et al., 2012; J. Liu, own observations), but it is unclear if the Azure-winged Magpie populations in a multiple cuckoo system in China have egg recognition ability and, if so, how they identify parasitic eggs. One earlier research suggested that Azure-winged Magpies are more sensitive to the number of eggs than to their size (Shang et al., 1994). Therefore, we wondered if changes in egg numbers would affect egg rejection behavior in Azure-winged Magpies. To address these questions, egg discrimination ability and recognition mechanism of the Azure-winged Magpie in China were investigated. We predicted that Fusong Azure-winged Magpies also have a high egg recognition ability. If the Azure-winged Magpies are more sensitive to egg numbers, directly increasing the number of eggs by adding two experimental eggs will result in an increased rejection rate of the experimental eggs compared to replacing two Azure-winged Magpie's eggs with two experimental eggs.

2.   Methods

2.1   Study area

This study was conducted in Fusong County, southeastern Jilin, China (42°19ʹ N, 127°15ʹ E), an area of secondary forest fragmented by Corn (Zea mays) crop farmlands and scattered plantations (dominated by larch Larix spp.), from May to June 2013. This region is in the temperate zone at an elevation of 481 ​m, with a continental monsoon climate characterized by cold and snowy winters with an average annual temperature of 4 ​℃ (Yang et al., 2019).

2.2   Egg recognition experiment

Field studies were conducted during the breeding season of Azure-winged Magpies in 2013. Breeding nests in the study area were searched and egg discriminate experiments were conducted on nests in the early stage of incubation, that were assigned to one of these four treatments. Experiment 1, in which two Japanese Quail (Coturnix japonica) eggs were introduced (Fig. 1A). Experiment 2, in which two host eggs were removed and two quail eggs were added (Fig. 1B). In Experiment 1 and Experiment 2, we wondered if Azure-winged Magpies were able to recognize foreign eggs and whether the increase in clutch size caused Azure-winged Magpies to become more sensitive to experimental eggs and reject the experimental eggs more often. Experiment 3, two Azure-winged Magpie eggs (conspecific eggs) were added to the host magpie nest. Experiment 4, two host eggs were removed and two conspecific eggs were added (Fig. 1C). In Experiment 3 and Experiment 4, we wondered if Azure-winged Magpies were able to recognize conspecific eggs and whether the increase in clutch size caused Azure-winged Magpies to become more sensitive to conspecific eggs and reject the conspecific eggs more often. Experiment 5, namely the "control group", refers to a group in which nests were visited using the same procedure as the experimental group, but without any experimental eggs added to control for human disturbance (n ​= ​10). Most Azure-winged Magpies can recognize foreign eggs and will reject quail eggs within three days after the experiment (own unpublished data). Therefore, after confirming that the Azure-winged Magpies recognized and discarded the experimental parasitic eggs, the egg recognition mechanism was further tested. For Azure-winged Magpies that could correctly recognize and reject the experimental parasitic eggs, Experiment 6, a '4 ​+ ​4' experiment, was subsequently performed as described by Wang et al. (2013) and Yi et al. (2020). In Experiment 6, four Azure-winged Magpie eggs and four quail eggs were simultaneously placed in a nest (Fig. 1D). If the egg recognition mechanism of the Azure-winged Magpie is true recognition mechanism, then all the individual Azure-winged Magpies will reject the quail eggs added to their nests. For all of the experimental nests, host response to the experimental eggs was observed for six days (Moksnes et al., 1991; Yang et al., 2019). If no egg rejection was observed within this period, the eggs were considered accepted. If the experimental eggs were rejected and the host eggs were not damaged, disappeared, or the host abandoned the nest, the eggs were considered rejected. If the host rejected its own eggs, it was considered as the cost of rejection. Experimental nests that were predated or destroyed by humans during the six-day period were eliminated from the experimental results.

Figure  1.  Egg recognition experiments in nests of the Azure-winged Magpie. (A) refers to adding two Japanese Quail eggs to the nest; (B) refers to replacing two host eggs with two quail eggs; (C) refers to replacing two host eggs with two conspecific eggs and (D) refers to experimental nests with four experimental eggs and four host eggs.

DownLoad: Full-Size Img PowerPoint

2.3   Statistical analyses

Statistical analysis was performed with IBM SPSS 20.0 (IBM Inc., USA) for Windows. Fisher's exact test or Pearson's Chi-squared test was used to compare the probabilities obtained in different experimental groups. All tests were two-tailed. The significance level was P ​ < ​0.05, and summary data are presented as mean ​± ​SD.

3.   Results

In Fusong, no Azure-winged Magpies were found to be parasitized by Indian Cuckoos and Asian Koels in 2013. There were no differences in rejection rate of quail eggs between treatments 1 and 2 (45.5%, n ​= ​22 vs. 71.4%, n ​= ​14, Pearson's Chi-squared test, χ² ​= ​2.414, df ​= ​2, P ​= ​0.299, Table 1). There were no differences in rejection rate of conspecific eggs between treatments 3 and 4: Azure-winged Magpies did not reject any eggs (0%, n ​= ​16 vs. 0%, n ​= ​12). In Experiment 6, Azure-winged Magpies accurately rejected all experimental eggs without incurring any rejection cost (100%, n ​= ​13). Video recordings showed that in Experiment 6, most Azure-winged Magpies rejected the experimental eggs within 2 ​h by pecking the eggs first and then seizing them and throwing them out of the nest (Appendix Video S1 and S2). Some individuals simply seized and threw the experimental eggs out of the nest without pecking. In all the experiments, we found no nest desertion in Azure-winged Magpies.

Table  1.  Rejection frequencies (%) of experimental Japanese Quail eggs by Azure-winged Magpies.

Experiment	Number of exp. nests	Rejecting two exp. eggs	Accepting two exp. eggs	Rejecting one exp. egg	Rejection cost
Adding two quail eggs	22	8 (36.4)	8 (36.4)	2 (9.0)	4 (18.2)
Replacing two host eggs with two quail eggs	14	9 (64.4)	3 (21.4)	1 (7.1)	1 (7.1)

 | Show Table

DownLoad: CSV

Supplementary video related to this article can be found at https://doi.org/10.1016/j.avrs.2023.100112

4.   Discussion

The present study showed that the Azure-winged Magpie population of Fusong, Jilin, China, had a moderate egg rejection rate toward non-mimetic quail eggs but was unable to recognize conspecific eggs that were highly mimetic to its own eggs. Some individuals displayed recognition errors in their rejection of experimental eggs. For example, three nests of Azure-winged Magpies mistakenly rejected one or two of their own eggs but accepted quail eggs. In addition, the species showed a true recognition mechanism (template-based recognition) and rejected foreign eggs by accurate recognition of their own eggs.

Although Azure-winged Magpies were found to be parasitized by Indian Cuckoos in another population 320 ​km away from Fusong, no Azure-winged Magpies were found to be parasitized by Indian Cuckoos and Asian Koels in our study area in 2013. The Fusong Azure-winged Magpie population studied showed a rejection rate of 55.6% toward non-mimetic quail eggs. This rejection rate is lower than the rejection rate displayed by a Korean population (96%) (Son et al., 2015), and the Iberian Magpies (73.7%) (Avilés, 2004). The difference in the egg rejection rate in the three studies may result from the differences in coevolutionary interactions between Azure-winged Magpies and cuckoos in these regions (Yang et al., 2015; Liang et al., 2016), or the differences among the experimental eggs used. The experimental eggs used by Avilés (2004) were highly similar to the eggs of Great Spotted Cuckoos, and Son et al. (2015) used blue model eggs made from soft clay, while our study used non-mimetic quail eggs, which was a bit larger than host eggs (quail eggs: 29.42–37.25 ​mm length ​× ​24.11–27.3 ​mm breadth vs. magpie eggs: 24.06–28.28 ​mm length ​× ​19.12–21.42 ​mm breadth). The three types of experimental eggs differ in material, color, and size and this possibly contributed to the different egg rejection rates observed in these studies (Roncalli et al., 2017; Luro et al., 2018). The Azure-winged Magpie population studied here showed a relatively low egg rejection rate, which could be due to the addition of two experimental eggs because some studies suggest a reduced host rejection probability toward parasitic eggs in the presence of multiple parasitism (Moskát et al., 2009; Stevens et al., 2013). Similar to the observations of Son et al. (2015), we also found no nest desertion in Azure-winged Magpies, suggesting that nest desertion would not be a behavioral strategy against cuckoo parasitism.

In this study, recognition errors and rejection costs were sometimes involved in the rejection of experimental eggs by the Azure-winged Magpie. Two nests of Azure-winged Magpies also rejected one of their own eggs at the same time when they rejected a quail egg. Three nests of Azure-winged Magpies accepted quail eggs and rejected one or two of their own eggs. This is similar to an Azure-winged Magpie population in Beijing (Shang et al., 1994), but differs from the results of Son et al. (2015), as the magpie population they studied did not suffer any rejection cost when they rejected blue model eggs. Avilés (2004) painted quail eggs with acrylic paint to simulate Great Spotted Cuckoo eggs and studied the egg rejection behavior of Azure-winged Magpies. They observed rejection costs in that population with many (28.6%) of the nests losing one or two of their own eggs when rejecting the parasite eggs. The rejection cost incurred by Azure-winged Magpies in rejecting quail eggs could be related to the thicker eggshell of the latter. Our video recordings show that Azure-winged Magpies usually puncture a hole in the experimental eggs before seizing them and throwing them out of the nest. Quail eggs, with a thicker eggshell, are not easily punctured (Swynnerton, 1918; Krüger, 2011). Azure-winged Magpies, in their effort to puncture quail eggs, could also destroy some of their own eggs and therefore incur a rejection cost.

Avilés (2004); Avilés and Parejo (2012), and Son et al. (2022) observed the ability of magpies to recognize conspecific eggs, with 42.8% in Avilés (2004), 35.5% in Avilés and Parejo (2012) and 30% in Son et al. (2022). Both Avilés and Parejo (2012) and Son et al. (2022) found that the recognition and rejection of conspecific eggs by Iberian Magpies and Azure-winged Magpies were strongly related to variation in egg size within the nests and less related to variation in egg color. However, in our study, Azure-winged Magpies accepted all the conspecific eggs whether two such eggs were added to the nests or the eggs were added as a replacement for two removed host eggs. This indicates that the Azure-winged Magpie of Fusong population is unable to recognize conspecific eggs. This difference may be related to differences in the coevolutionary history and the interactions between Azure-winged Magpies, Iberian Magpies and different cuckoo species in China, Korea and Iberia. For example, in Iberia, records show that Great Spotted Cuckoos and Common Cuckoos once co-evolved with Iberian Magpies but there are few field observations indicating the recent occurrence of inter-specific brood parasitism in the Iberian Magpie (Alonso et al., 1991; Valencia et al., 2003, 2005; Avilés, 2004). For this reason, Avilés (2004) suggested that the Iberian Magpies evolved egg recognition not as a defense against interspecific brood parasitism, but as a defense against conspecific brood parastisim. Indian Cuckoo parasitism is also rarely reported on Azure-winged Magpies in South Korea (Lee, 2014), and Son et al. (2015) suggested that the high egg recognition ability of Azure-winged Magpies may be due to the coevolutionary relationship between Azure-winged Magpies and Indian Cuckoo in the past, because interspecific brood parasite is the main selection pressure leading to the evolution of egg recognition ability in birds (Ruiz-Raya et al., 2016). However, Azure-winged Magpies are parasitized by both Indian Cuckoos and Asian Koels in China.

Interestingly, different from Wang et al. (2013), in which increase of clutch size triggered clutch destruction behavior in Common Moorhens (Gallinula chloropus), Azure-winged Magpies in this study accepted all additional two or four experimental eggs added to their nests in Experiment 1, Experiment 3, and Experiment 6 without nest desertion. In Experiment 6, all the individuals rejected quail eggs, and no one rejected their own eggs. Although we did not design an additional experiment by adding an extra set of experimental treatment in which the quail eggs were more numerous than the Azure-winged Magpie's own eggs, the egg recognition mechanism of Fusong Azure-winged Magpies is more likely to be true recognition mechanism, as the discordancy mechanism predicts that hosts typically eject the egg types that are most dissimilar (and thus, necessarily, in the minority) in clutches, irrespective of whether these are their own eggs or parasitic eggs. If egg discrimination in the Azure-winged Magpie occurs by the discordancy recognition mechanism, then there are two possible outcomes in Experiment 6. The first outcome is that all the individuals do not reject the eggs in nests, and the second outcome is that all the individuals randomly reject the eggs in the nests. If all the individuals randomly reject the eggs in the nests, it is highly unlikely that in Experiment 6 all individuals will reject only quail eggs.

In conclusion, our study shows that Azure-winged Magpies in a north Chinese population possess moderate egg recognition ability as some populations do in other geographical regions, and show a true recognition mechanism of rejecting foreign eggs by accurate recognition of their own eggs. However, they cannot recognize mimetic conspecific eggs. In the '4 ​+ ​4' experiment, all Azure-winged Magpies accurately rejected quail eggs added to their nests. However, more experiments may be necessary in order to completely eliminate the mechanism of recognition by discordancy. The egg characteristics used as recognition clues remain to be identified. There may be differences in parasitic cuckoo species and cuckoo parasitism across China (Yang et al., 2012). For example, in northern China (Palearctic realm), Azure-winged Magpies are parasitized by only one cuckoo species, while in southern China (Indomalayan realm), Azure-winged Magpies could be parasitized by several cuckoo species, including Common Cuckoos, Indian Cuckoos and Asian Koels. Geographical variations in egg recognition ability of Azure-winged Magpies under different cuckoo and/or even conspecific parasitism remain to be determined.

Ethics statements

The experiments comply with the current laws of China, where they were performed. Experimental procedures were in agreement with the Animal Research Ethics Committee of Hainan Provincial Education Centre for Ecology and Environment, Hainan Normal University (No. HNECEE-2014-005).

Authors' contribution

WL designed the study, LW carried out field experiments, JL performed the analyses and wrote the draft manuscript, WL edited and improved the manuscript. All authors read and approved the final manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

We are grateful to two anonymous reviewers' constructive comments which help improve this manuscript.