AGGRESSION TO DUMMY CUCKOOS BY POTENTIAL EUROPEAN CUCKOO HOSTS by EIVIN RØSKAFT 1,2) , ARNE MOKSNES 1) , BÅRD G. STOKKE 1) , VÍTEZSLAV BICÍK3) and CSABA MOSKÁT4,5) Zoology, Norwegian University of Science and Technology, NTNU, Realfagbygget, N-7491 Trondheim, Norway; 3 Department of Zoology and Anthropology, Palacký University, Tr. Svobody 26, CZ-771 46 Olomouc, Czech Republic; 4 Animal Ecology Research Group of the Hungarian Academy of Sciences, c/o Hungarian Natural History Museum, H-1083 Budapest, Ludovika ter 2, Hungary) (1 Department of
(Acc. 4-II-2002)
Summary Aggression directed by 53 potential host species towards a dummy of the parasitic common cuckoo, Cuculus canorus, was tested in relation to their breeding habitat, their suitability as a host and whether they were breeding in sympatry or not with the cuckoo. Host habitats were divided into three categories: (1) always breeding near trees, (2) some populations breeding near trees, others in open areas, and (3) always breeding in open areas. Each species was also placed in one of ve categories according to their suitability as a cuckoo host. Strong support was found for predictions derived from the ‘spatial habitat structure hypothesis’, which argues that common cuckoos only breed in areas where they have access to vantage points in trees. Thus, species which have some populations breeding near trees and others breeding further from trees have a different cuckoo-host population dynamics than species that always breed near trees, or always breed in open areas. Aggression levels were highest among species regarded as being always suitable as hosts, and species which always breed near trees. However, populations breeding in sympatry with the cuckoo were more aggressive than allopatric populations, indicating the plasticity of aggressive behaviour. Adaptive behaviour in cuckoo hosts can be predicted from the ‘spatial habitat structure hypothesis’.
2)
Corresponding author’s e-mail address:
[email protected] We thank the Norwegian Research Council (grant no. 125805/410), the Nansen Foundation, the EU and the Hungarian Scienti c Research Fund (OTKA) (grant no. T29570) for providing funds for this project. Mojmír Bernatik and Miroslav Král helped with the eldwork. 5)
© Koninklijke Brill NV, Leiden, 2002
Behaviour 139, 613-628 Also available online -
614
RØSKAFT, MOKSNES, STOKKE, BICÍK & MOSKÁT
Introduction About 100 of the 9672 bird species described worldwide have evolved into obligate brood parasites (Johnsgard, 1997; Davies, 2000). One enigma in the study of co-evolution is why avian brood parasites are so obviously successful in parasitising their hosts. Brood parasites may specialise on a single or very few host species (e.g. the great spotted cuckoo Clamator glandarius), or be generalists where each female parasitises several host species (e.g. the brown-headed cowbird Molothrus ater, and the shiny cowbird M. bonariensis; Friedmann, 1963; Friedmann et al., 1977). The common cuckoo, Cuculus canorus, falls between these two extremes. Each individual female is believed to be a host specialist, but the species as a whole has been known to parasitise more than 100 hosts in Europe, although only 16 are normally utilised (Alvarez, 1994; Moksnes & Røskaft, 1995). Each specialist cuckoo strain (called gens) parasitises a single host species, or a group of hosts (Baker, 1942; Moksnes & Røskaft, 1987; Davies, 2000). A number of adaptations in both the cuckoo and its hosts are believed to be the result of a co-evolutionary arms race (Dawkins & Krebs, 1979; Davies & Brooke, 1988, 1989a, b; Moksnes et al., 1990). Such adaptations in the cuckoo include cryptic laying, egg mimicry and removal of host eggs when laying (Chance, 1940; Baker, 1942; Wyllie, 1981; Davies, 2000). Adaptations in the hosts include rejection of the cuckoo egg (Davies & Brooke, 1988, 1989a, b; Moksnes & Røskaft, 1989, 1992; Moksnes et al., 1990, 1991, 1993a), aggression towards the cuckoo (Moksnes et al., 1990; Duckworth, 1991; Røskaft et al., 2002b), and evolution of low intraclutch and high interclutch variation in the eggs (Øien et al., 1995; Soler & Møller, 1996; Stokke et al., 1999, 2002). Because the cost of acceptance is very high, the obvious response of a host should be to reject the parasitic egg. However, many hosts accept a large proportion of them. This is true for both North American cowbird (Rothstein, 1982, 1990) and European cuckoo (Davies & Brooke, 1988, 1989a, b; Moksnes et al., 1990; Røskaft & Moksnes, 1998) hosts. Several hypotheses have been proposed to explain this intermediate rejection pattern of many brood parasite hosts, but it should be stressed that all of them are consistent with the overall hypothesis that the interaction between parasites and their hosts can be described as a co-evolutionary arms race. Firstly, the ‘time lag hypothesis’ (Rothstein, 1982, 1990) proposes that the rejecter allele has not yet appeared
AGGRE SSION OF CUCKOO HOSTS
615
in, or spread through, the population. Secondly, the ‘equilibrium hypothesis’ (Lotem et al., 1992, 1995; Takasu et al., 1993; Lotem & Nakamura, 1998; Takasu, 1998a, b) states that a population should have a balance between acceptors and rejecters due to an age-dependent process or a parity between the bene ts of egg rejection and the costs of making recognition errors in relation to the level of parasitism (but see also Rohwer & Spaw, 1988; Røskaft et al., 1990, 1993; Røskaft & Moksnes, 1998). Thirdly, a hypothesis has been put forward recently which argues that the best behaviour is to reject all parasitic eggs. However, individuals in a non-parasitised population may be selected to be acceptors for different reasons. In species where the spatial structure of the habitats leads to the existence of both non-parasitised and parasitised populations, migration between these populations will in uence the level of rejection. These arguments are based on the importance of geographically separated populations for co-evolutionary processes (Thompson, 1994). The importance of a metapopulation approach for understanding the structures of host-parasite adaptations was rst acknowledged by Soler et al. (1998, 1999a), Lindholm (1999) and Lindholm & Thomas (2000) and the approach was further developed by Røskaft et al. (2002a). The metapopulation approach is based on models where patterns of gene ow among different local populations may affect micro-adaptations (Via et al., 1995; Gandon et al., 1996; Grenfell & Harwood, 1997; Schlichting & Pigliucci, 1998). According to this hypothesis, a species that has both non-parasitised and parasitised populations within a speci c geographical area will have a lower average level of rejection of cuckoo eggs than one whose populations are all, in principle, utilised by the cuckoo (Røskaft et al., 2002a). The behaviour of hosts towards a cuckoo at their nest is an antiparasite adaptation that remains virtually unstudied with reference to European cuckoos (Moksnes et al., 1990, 2000; Duckworth, 1991; Røskaft et al., 2002b). Moksnes et al. (1990) found a close relationship between aggressive behaviour towards a dummy cuckoo and the host rejection rate, in that a host species that rejects cuckoo eggs at a high rate is also more aggressive towards a dummy cuckoo, indicating that the co-evolutionary arms race (Davies & Brooke, 1989a) affects aggression and rejection at similar levels (but see also Briskie et al., 1992; Soler et al., 1999b). In this paper, we report data on how different European cuckoo hosts behave towards a dummy cuckoo near their nest and analyse them in the light of the ‘co-evolutionary arms race’ and the ‘spatial habitat structure’ hypotheses. The analyses are performed at different
616
RØSKAFT, MOKSNES, STOKKE, BICÍK & MOSKÁT
levels. Host species were divided according to which habitat the majority of populations were breeding in, their degree of suitability as a cuckoo host, and whether they are breeding in sympatry or allopatry with the cuckoo. Material and methods In this study, we have de ned a host population as sympatric with the cuckoo if cuckoos were breeding in the area in which we performed an experiment. If they were not breeding within the area, even though they were breeding only a few kilometres away, the host population was de ned as allopatric with the cuckoo. None of the allopatric populations bred further away than a few kilometres from a cuckoo population, indicating that experiments in allopatric populations are comparable. Data from central Norway were collected in the Tydal mountains (63± N, 12± E), where host species breed in sympatry with the cuckoo, and a lowland area in Stjørdal (63± N, 11± E), where they breed in allopatry (see Moksnes & Røskaft, 1987, 1989, 1992; Moksnes et al., 1990; for a more detailed description of these areas). Some experiments were also carried out near Lund in southern Sweden (55± N, 13± E) and in northern Zealand in Denmark (55± N, 12± E), where the hosts were sympatric with cuckoos in both cases. Data from the Czech Republic were collected near Luzice and Lednice (48± N, 16± E; see Moksnes et al., 1993b; Øien et al., 1996; Honza et al., 1998; for a description of the study area), and near Olomouc (49± N, 17± E) and in the Jeseniky Highlands (50± N, 17± E) in central and northern Moravia, respectively; in all these areas the hosts are in sympatry with cuckoos. In Hungary, data were collected from Budapest (47± N, 19± E), where the hosts are allopatric with cuckoos, and Kiskunlacháza, Bugyi and Apaj (47± N, 18± E), where they are sympatric with cuckoos (Moskát & Honza, 2000). All these sympatric or allopatric relationships with cuckoos are known to have existed for at least 50 years. The experiments were performed by placing a dummy cuckoo less than 1 m from a nest. The observer then moved about 40-50 metres away and observed the behaviour of the host for 15 minutes. The reaction of the host to the cuckoo was noted on a scale from 1 to 4. 1 is ‘No reaction’ — the host was observed near the nest, but paid no attention to the cuckoo; in some cases, the host even returned to the nest and sat on the eggs. 2 is ‘Distress calling’ — the host uttered distress calls, alarm calls, and sometimes also sang, the sounds obviously being directed at the cuckoo. 3 is ‘Mobbing’ — the host reacted to the dummy by ying around it or diving close to it one or several times, but never touching it. 4 is ‘Attacking’ — the host vigorously attacked the dummy, touching its body (Moksnes et al., 1990). The experiment was terminated as soon as the host started to attack the dummy. ‘No reaction’ and ‘distress calling’ were pooled and termed ‘no aggression’, and ‘mobbing’ and ‘attacking’ were pooled and termed ‘aggression’. When performing the statistical analyses, percentage aggression, i.e. the percentage of the experiments where the host either mobbed or attacked the dummy, was used as a unit. All the experiments were performed during the egg-laying or incubation periods of the host. The aim was to perform the experiments as early as possible in the breeding cycle of the host, but they were still performed if a nest was found during the incubation stage. No control experiments were performed, but cuckoo hosts have previously been shown to behave differently towards a cuckoo and a control dummy of another less threatening species (Duckworth, 1991; Moksnes et al., 1993a).
AGGRE SSION OF CUCKOO HOSTS
617
We performed a total of 621 experiments on 53 species (Table 1) during May and June in 1986-2000. The number of experiments in the various species varied between 1 and 97. However, each nest was tested only once. The species were divided into ve groups with regard to their suitability as a cuckoo host. (1) Twenty-eight of the species can be regarded as ‘always suitable’, a suitable host being de ned as having a nest that is accessible to the cuckoo, providing food that the cuckoo chick can digest, and having eggs and chicks that the cuckoo chick is able to eject (Davies & Brooke, 1989a, 1989b; Moksnes et al., 1990). (2) ‘Partially hole nesters’ are species (or individuals in some populations) that breed in inaccessible cavities in some areas, but otherwise in open nests that are accessible to the cuckoo. These data were taken from (Snow & Perrins, 1998). Six species are in this category (redstart Phoenicurus phoenicurus, black redstart P. ochruros, wren Troglodytes troglodytes, robin Erithacus rubecula, pied/white wagtail Motacilla alba and grey wagtail M. cinerea; Table 1). (3) ‘Large nest and eggs’ means it is dif cult for the young cuckoo to eject the nest content. Four Turdus species are in this category (Table 1) (Moksnes et al., 1990). (4) ‘Hole nesters’ are nine species that normally breed in nest boxes or cavities (Snow & Perrins, 1998) which are inaccessible to the cuckoo (Table 1). (5) ‘Seed eaters’ are six species (Table 1) that normally feed their chicks on seeds that the young cuckoo is unable to digest (Snow & Perrins, 1998). Finally, the hosts were divided into three categories regarding the habitat in which they normally breed (Røskaft et al., 2002a). (1) ‘Always breeding near trees’ — species whose populations breed in habitats where they are, in principle, always near trees that are suitable as cuckoo vantage points. This group comprises 36 species (Table 1). (2) ‘Breeding both near and far from trees’ — some populations of these species breed near trees and others in open areas far from trees where it is assumed they are not parasitised because cuckoos do not nd their nests. This group contains 12 species (Table 1). (3) ‘Breeding in open areas’ — in principle, such species always breed far from trees and are therefore not parasitised (or are only infrequently parasitised, as shown by Moksnes & Røskaft, 1995) even though they may otherwise be suitable as cuckoo hosts. This category comprises ve species (Table 1). The habitats were classi ed according to Snow & Perrins (1998). The rejection rates of host species to non-mimetic cuckoo eggs derive from published papers and our own unpublished experiments in Norway, the Czech Republic and Hungary (von Haartmann, 1976; Gärtner, 1982; Järvinen, 1984; Davies & Brooke, 1989a; Brown et al., 1990; Moksnes et al., 1990; 1994; Moksnes & Røskaft, 1992; Brooke et al., 1998; Alvarez, 1999; Moskát & Fuisz, 1999). The rejection rate is de ned as the number of cases where the parasitic egg was rejected (ejected or deserted) in relation to the total number of experiments involving the adding of arti cial, non-mimetic cuckoo eggs to the clutch. In this paper, experiments from different populations are pooled. Treating each species as an independent data point may lead to overestimation of the true number of degrees of freedom in statistical analyses (Felsenstein, 1985; Harvey & Pagel, 1991). To control for possible effects of common descent, the species used in the analysis were organised in a phylogenetic tree. We produced one tree based on molecular data .DNA hybridisation; Sibley & Ahlquist, 1991) and another on morphology (Howard & Moore, 1991). In the latter tree, we assumed polytomies between species within a genus, and genera within a family, etc. To obtain a normal distribution, the rejection rate had to be arcsin transformed before the analysis. We used the computer program package PDAP (Garland et al., 1993, 1999; Phenotypic Diversity Analysis Programs) version 5.0 to make the tree and load variable data. This package also contains Felsenstein’s (1985) independent
618
RØSKAFT, MOKSNES, STOKKE, BICÍK & MOSKÁT
comparison method, which allowed us to obtain pairwise contrasts of the variables between nodes in the phylogenetic trees that were independent of each other. The branch lengths were assigned by the methods of Grafen (1989) or Pagel (1992), or were set as a constant (D1). The branch length assignments that were used varied for each trait and also among the trees. We selected the branch lengths that yielded absolute values of contrasts that were not related to their standard deviations (p < 0:05) for any of the traits analysed (Garland et al., 1992). The relationship between the variables was analysed by General Linear Models (GLM). All the tests are two-tailed.
Results Aggression in relation to rejection rates The correlation between the rejection rate of different hosts and the ‘percentage aggression’ was highly signi cant (r36 D 0:718, p D 0:000). A linear regression of contrasts where phylogeny was based upon DNA hybridisation, and where ‘percentage aggression’ was the dependent variable and rejection rate was the independent variable was statistically signi cant 2 (r2;34 D 0:571, p < 0:001). A linear regression based on contrasts estimated under the phylogenetic tree based on morphology gave similar results 2 (r2;34 D 0:320, p D 0:002). Aggression in relation to whether hosts were breeding in sympatry or allopatry with the cuckoo For 14 species, we have data from both sympatric and allopatric populations (Table 1). Even though the sample size for some species is small, 11 species were more aggressive in sympatric than allopatric areas, and one species was more aggressive in allopatric areas; two species were equally aggressive. The difference in aggression between allopatric and sympatric areas was statistically signi cant (sign test, p < 0:003). We have conclusive data for two suitable, closely related species, the brambling Fringilla montifringilla and the chaf nch F. coelebs. In both cases, individuals in sympatric populations were more aggressive than those in allopatric populations, but the gures are only statistically signi cant for chaf nches (Â12 D 6:94, p D 0:008; Table 1). However, in the case of the blackcap Sylvia atricapilla, another suitable host, there was no difference between the sympatric and allopatric populations (Fisher’s exact probabilities test, p D 0:686; Table 1). Unfortunately, samples are too small for the other species, except the blackbird
619
AGGRE SSION OF CUCKOO HOSTS
TABLE 1. Experimental setup Host species
Fringilla montifringilla F. coelebs Sylvia atricapilla S. borin S. nisoria Emberiza citrinella E. schoenobaenus Muscicapa striata Phylloscopus trochilus P. collybita P. sibilatrix Hippolais icterina Luscinia svecica L. megarhynchos Lanius collurio Prunella modularis Anthus pratensis Acrocephalus scirpaceus A. arundinaceus A. schoenobaenus A. palustris Sylvia communis S. curruca Calcarius lapponicus Locustella naevia Anthus spinoletta Alauda arvensis Eremophila alpestris Phoenicurus phoenicurus Erithacus rubecula Troglodytes troglodytes Motacilla alba M. cinerea Phoenicurus ochruros Turdus merula T. iliacus T. pilaris T. philomelos Ficedula hypoleuca Parus major
Suit.
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 4 4
Habitat
T T T T T T T T T T T T T T T T T&O T&O T&O T&O T&O T&O T&O T&O O O O O T T T T&O T&O T&O T T T T T T
Aggr.
No aggr.
S=A
S=A
13 = 4 10 = 42 21 = 11 1=2 2=¡ 5=1 6=1 ¡=5 10 = ¡ 2=5 5=¡ ¡=4 4=¡ 1=¡ 11 = ¡ ¡=0 34 = ¡ 15 = ¡ 34 = ¡ 1=¡ 6=¡ 2=¡ 0=¡ 1=¡ 0=¡ 2=¡ 0=¡ 0=¡ 2=0 0=¡ 0=¡ 2=1 2=¡ 2=¡ 8=2 1=1 0=¡ 2=¡ 2=0 0=0
4=4 1 = 44 4=3 1=1 0=¡ 1=3 0=0 ¡=4 0=¡ 0=2 0=¡ 0=¡ 3=¡ 0=¡ 3=¡ ¡ = 13 19 = ¡ 42 = ¡ 11 = ¡ 5=¡ 0=¡ 0=¡ 1=¡ 0=¡ 1=¡ 1=¡ 4=¡ 1=¡ 4=1 2=¡ 2=¡ 0=1 1=¡ 2=¡ 3 = 19 2 = 23 12 = ¡ 14 = ¡ 7=3 5=2
p
0.359 0.008 0.686
0.114
0.006 0.205
0.545
620
RØSKAFT, MOKSNES, STOKKE, BICÍK & MOSKÁT
TABLE 1. (Continued) Host species
P. caeruleus P. palustris Ficedula albicollis Sturnus vulgaris Sitta europaea Saxicola torquata Riparia riparia Chloris chloris Coccothraustes coccothraustes Carduelis ammea C. carduelis Acanthis cannabina Serinus serinus
Suit.
4 4 4 4 4 4 4 5 5 5 5 5 5
Habitat
T T T T T T&O O T T T T T T
Aggr.
No aggr.
S=A
S=A
¡=0 ¡=0 0=¡ ¡=0 1=¡ 2=¡ 0=¡ 1=2 0=¡ ¡=0 0=¡ 1=¡ 0=¡
¡=5 ¡=1 8=¡ ¡=7 2=¡ 1=¡ 1=¡ 2=7 1=¡ ¡=1 3=¡ 3=¡ 2=¡
p
Number of experiments where passerine individuals behaved aggressively (Aggr.) or nonaggressively (No aggr.) in sympatric (S) and allopatric (A) populations when confronted with a dummy cuckoo in relation to their suitability (Suit.) as hosts (1 D always suitable, 2 D partially hole nester, 3 D large nest or eggs, 4 D hole nester, 5 D seed eater) and their breeding habitat (T D among trees, T & O D among trees or in open areas, O D in open areas). Differences between allopatric and sympatric populations were tested by Fisher’s exact probabilities test.
Turdus merula, which breeds near trees but has large nests and eggs. Blackbirds from the sympatric Czech and Swedish populations were more aggressive towards the dummy cuckoo than those from the allopatric populations in Hungary and Norway (Fisher’s exact probabilities test, p D 0:006; Table 1). Aggression in relation to breeding habitat As data for hosts that are not always suitable chie y come from a single habitat type, we have only analysed the effect of breeding habitat for species that are always suitable as cuckoo hosts (Table 2). Such species that breed close to trees were signi cantly more aggressive than those breeding further from trees, or in both types of habitat (F -test, p D 0:013). The difference was statistically signi cant between species always breeding near trees and those breeding in open areas (F -test, p D 0:002). The mean aggression value for species always breeding near trees was also higher than for species breeding both near trees and in open areas. However, the difference was not signi cant (F -test, p D 0:318). The difference between species some of
621
AGGRE SSION OF CUCKOO HOSTS
TABLE 2. Aggression of suitable European hosts to a dummy cuckoo in relation to their breeding habitat Breeding habitat
Percentage aggression Mean
SD
Nspecies
76.4 60.4 16.8
27.1 40.8 33.5
16 8 4
Always near trees Near trees and in open areas Always in open areas
The average of each species is used as a unit. Differences between groups: F D 5:2, p D 0:013.
TABLE 3. Aggression among European passerines to a dummy cuckoo in relation to their suitability as hosts Suitability as hosts
Always suitable Partially hole nesters Large nest and eggs Hole nesters Seed eaters
Percentage aggression Mean
SD
Nspecies
62:3 36:8 11:8 13:0 8:3
36.9 32.6 12.4 23.3 12.9
28 6 4 9 6
The average of each species is used as a unit. Differences between groups: F D 7:5, p D 0:000.
whose populations breed near trees and others in open areas, and species breeding only in open areas was not statistically signi cant (F -test, p D 0:096). In open areas, the level of aggression was similar to the values for the three categories regarded as not always suitable as hosts (Tables 2 and 3). Aggression in relation to host suitability The ‘percentage of experiments where the hosts behaved aggressively towards the cuckoo’ was in most cases signi cantly higher for the suitable hosts than for any of the other groups of potential hosts (Table 3; versus partially hole nesters: F -test, p D 0:129; versus species with large nests: F -test, p D 0:012; versus hole nesters: F -test, p D 0:001; versus seed eaters: F -test, p D 0:001). The values were more or less equal among the four categories of hosts regarded as ‘not always suitable’, and differences were in most cases not statistically signi cant (F -tests; 0:050 < p < 0:929;
622
RØSKAFT, MOKSNES, STOKKE, BICÍK & MOSKÁT
Table 3). Thus, suitable hosts behaved more aggressively towards the dummy cuckoo than other species. Aggression in relation to suitability and habitat when controlled for phylogeny A GLM controlling for phylogeny, based upon morphological classi cation and where ‘percentage aggression’ was the dependent variable, was statis2 tically signi cant (r2;50 D 0:159, F D 4:53, p D 0:016). Both the independent variables suitability (p D 0:041) and habitat (p D 0:030) gave a signi cant effect. The phylogeny test based on DNA hybridisation with ‘per2 centage aggression’ as the dependent variable was signi cant (r2;50 D 0:141, F D 3:93, p D 0:026), but only suitability as a host had a signi cant effect (p D 0:018), habitat (p D 0:161) giving no signi cant additional effect. Discussion We have shown that when allowance is made for phylogenetic effects, there is a strong correlation between the average aggression level of a species towards a dummy cuckoo and the rate at which it rejects nonmimetic cuckoo eggs. This nding supports Moksnes et al. (1990), who concluded that the cuckoo host arms race would lead to similar effects on aggression and on rejection of non-mimetic eggs. Thus, experiments on the aggressive behaviour of cuckoo hosts can be used as an alternative to egg recognition experiments for two reasons. (1) The disturbance of the nest is relatively short (maximum 15 minutes) and there is no reason to visit the nest repeatedly after the experiment ends. (2) Egg experiments may cause birds to either desert the nest or destroy the eggs (Davies & Brooke, 1989a; Moksnes et al., 1990). Since there is much less risk of desertion when performing aggression experiments, such experiments are preferable for ethical reasons, too. Hosts reported as always suitable were more aggressive to the dummy cuckoo than any of the other groups of hosts. This result can be predicted from the arms race hypothesis. We have made a distinction from previous studies because we have divided the potential hosts into several groups. Two of the groups (‘hole nesters in some areas’ and ‘hosts with large eggs and nests’) have previously been categorised as suitable hosts (Davies & Brooke,
AGGRE SSION OF CUCKOO HOSTS
623
1989a; Moksnes et al., 1990). Interestingly enough, both groups behaved in the same manner towards the dummy cuckoo as the ‘seed eater’ and ‘hole nester’ groups, although, the ‘hole nesters in some areas’ group was almost signi cantly more aggressive than the ‘seed eater’ group. However, their aggression level was far below that shown by the ‘always suitable’ hosts. We have shown that the spatial structure of the breeding habitat is a very important predictor of host behaviour. The ‘spatial habitat structure hypothesis’ was developed by Røskaft et al. (2002a). However, Soler et al. (1998) developed similar ideas when considering non-parasitised and parasitised populations of magpies, Pica pica, which are hosts for the great spotted cuckoo in southern Europe. In the GLM analyses, controlling for common descent, habitat was in one case (morphological classi cation) a signi cant predictor of suitability as a host in explaining the variation in host aggression levels. When the three habitat categories alone are considered, the ‘always suitable’ hosts breeding only in ‘open areas’ had the same level of aggression as the various groups not regarded as ‘always suitable’ hosts. This is not surprising because we assume that birds breeding only in open habitats are normally not used by the cuckoo and have therefore been selected on the same level as the groups of hosts that are not always suitable as cuckoo hosts. We would normally expect hosts with intermediate rejection rates to be less aggressive in populations allopatric with the cuckoo than in sympatric ones (Øien et al., 1999). Unfortunately, our data are insuf cient to test this prediction properly. However, chaf nches behaved less aggressively in allopatric than in sympatric populations. The chaf nch shows a high level of rejection (Braa et al., 1992; Moksnes, 1992) and, according to Øien et al. (1999), such species should not show such patterns. The brambling, which also has a high rejection rate, showed the same trend as the chaf nch, but the difference in aggression between allopatric and sympatric populations was not statistically signi cant, probably due to a low sample size. The blackcap behaved more in accordance with the hypothesis of Øien et al. (1999), because it rejected cuckoo eggs at a high rate (Moksnes et al. 1990) and would be expected to behave as aggressively in allopatric as in sympatric populations. The most surprising result was, however, found for the blackbird which, according to our classi cation, is not a particularly suitable host because of its large nest and eggs. It hardly ever gures as a host (Moksnes & Røskaft, 1995), although it is one of the most abundant passerines in Europe (Hagemeijer & Blair, 1997). However, in the Czech
624
RØSKAFT, MOKSNES, STOKKE, BICÍK & MOSKÁT
population, where blackbirds breed in sympatry with cuckoos, they were very aggressive towards dummy cuckoos. On the other hand, blackbirds in the allopatric populations in Hungary and Norway behaved less aggressively. Although, data are inadequate to draw rm conclusions (we did not test whether this could be anti-predator behaviour), they indicate that aggressive behaviour of hosts is more exible than rejection behaviour, and that birds breeding in sympatry with the cuckoo on average behave more aggressively, whether their level of rejection is high or low. We have shown that the level of aggression can be predicted from the rejection rate of a host, the habitat in which the host is breeding and the level of suitability of the host. All tests gave similar results, irrespective of whether or not they were tested in a phylogenetic framework. In further research, the metapopulation approach needs to be taken into consideration (Soler et al., 1998; Lindholm, 1999; Lindholm & Thomas, 2000) and different hosts have to be considered in relation to the habitat in which they are breeding. Variation in parasitism rates will affect the level of selection differently in populations that are sympatric and allopatric with the cuckoo. Gene ow between populations will determine the level of rejection and aggression. Even suitable hosts may never experience a cuckoo near their nests and there will be no selection for antiparasite behaviour. They will therefore behave in a similar manner towards the cuckoo as ‘unsuitable’ hosts. Even though the ‘equilibrium hypothesis’ may explain why there should be a balance between acceptors and rejecters within a population, the basic difference between the ‘equilibrium hypothesis’ and the ‘spatial habitat structure hypothesis’ is that the latter can predict the overall levels of rejection or aggression of a particular species. The levels for parasitism rate and costs of possible recognition errors may also in uence the host responses, although it is not possible to precisely predict their level of in uence (Røskaft et al., 2002a). The ‘spatial habitat structure hypothesis’ has been supported by the experiments carried out in the present study, although we used only one dummy species, and no controls. We feel there is no reason to add any results of control experiments in this study because hosts mostly behaved in accordance with most of the predictions, and previous studies have shown that cuckoo hosts behave differently towards dummy cuckoos and dummies of less threatening species (Moksnes & Røskaft, 1989; Duckworth, 1991).
AGGRE SSION OF CUCKOO HOSTS
625
References Alvarez, F. (1994). A gens of cuckoo Cuculus canorus parasitizing rufous bush chat Cercotrichas galactotes. — J. Avian Biol. 25, p. 239-243. — — (1999). Attractive non-mimetic stimuli in cuckoo Cuculus canorus eggs. — Ibis 141, p. 142-144. Baker, E.C.S. (1942). Cuckoo Problems. — Witherby Ltd, London. Briskie, J.V., Sealy, S.G. & Hobson, K.A. (1992). Behavioral defences against avian brood parasitism in sympatric and allopatric host populations. — Evolution 46, p. 334-340. Brooke, M. de L., Davies, N.B. & Noble, D.G. (1998). Rapid decline of host defences in response to reduced cuckoo parasitism: behavioural exibility of reed warblers in a changing world. — Proc. R. Soc., London B 265, p. 1277-1282. Brown, R.J., Brown, M.N., Brooke, M. de L. & Davies, N.B. (1990). Reactions of parasitized and unparasitized populations of Acrocephalus warblers to model cuckoo eggs. — Ibis 132, p. 109-111. Braa, A.T., Moksnes, A. & Røskaft, E. (1992). Adaptations of bramblings and chaf nches towards parasitism by the common cuckoo. — Anim. Behav. 43, p. 67-78. Chance, E. (1940). The truth about the cuckoo. — Country Life Ltd, London. Davies, N.B. (2000). Cuckoos, cowbirds and other cheats. — T&AD Poyser, London. — — & Brooke, M. de L. (1988). Cuckoos versus reed warblers: adaptations and counteradaptations. — Anim. Behav. 36, p. 262-284. — — & — — (1989a). An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. I. Host egg discrimination. — J. Anim. Ecol. 58, p. 207-224. — — & — — (1989b). An experimental study of co-evolution between the cuckoo, Cuculus canorus, and its hosts. II. Host egg markings, chick discrimination and general discussion. — J. Anim. Ecol. 58, p. 225-236. Dawkins, R. & Krebs, J.R. (1979). Arms races between and within species. — Proc. R. Soc., London B 205, p. 489-511. Duckworth, J.W. (1991). Responses of breeding reed warblers Acrocephalus scirpaceus to mounts of sparrowhawk Accipiter nisus, cuckoo Cuculus canorus and jay Garrulus glandarius. — Ibis 133, p. 68-74. Felsenstein, J. (1985). Phylogenies and the comparative method. — Am. Nat. 125, p. 1-15. Friedmann, H. (1963). Host relations of the parasitic cowbirds. — Smithsonian Institute, United State National Museum, Washington DC. Friedmann, H., Kiff, L.F. & Rothstein, S.I. (1977). A further contribution to knowledge of host relations of the parasitic cowbirds. — Smiths. Contr. Zool. 235, p. 1-75. Gandon, S., Capowiez, Y., Dubois, Y., Michalakis, Y. & Olivieri, I. (1996). Local adaptations and gene-for-gene coevolution in a metapopulation model. — Proc. R. Soc., London B 263, p. 1003-1009. Garland, T.J., Dickerman, A.W., Janis, C.M. & Jones, J.A. (1993). Phylogenetic analysis of covariance by computer simulation. — Syst. Biol. 42, p. 265-292. — —, Harvey, P.H. & Ives, A.R. (1992). Procedures for the analysis of comparative data using phylogenetically independent contrasts. — Syst. Biol. 41, p. 18-32. — —, Midford, P.E. & Ives, A.R. (1999). An introduction to phylogeneticallybased statistical methods, with a new method for con dence intervals on ancestral states. — Am. Zool. 39, p. 374-388.
626
RØSKAFT, MOKSNES, STOKKE, BICÍK & MOSKÁT
Grafen, A. (1989). The phylogenetic regression. — Phil. Trans. R. Soc., London B 326, p. 119-157. Grenfell, B. & Harwood, J. (1997). (Meta)population dynamics of interacting diseases. — Trends Ecol. Evol. 12, p. 395-399. Gärtner, K. (1982). Zur Ablehnung von Eiern und Jungen des Kuckucks (Cuculus canorus) durch die Wirtsvögel: Beobachtungen und Experimentelle Untersuchungen am Sumpfrohrsänger. — Vogelwelt 103, p. 210-224. von Haartmann, L. (1976). The reaction of a regular cuckoo host to foreign eggs. — Ornis Fenn. 53, p. 96-98. Hagemeijer, W.J.M. & Blair, M.S., eds. (1997). The EBCC atlas of European breeding birds. Their distribution and abundance. — T&AD Poyser, London. Harvey, P.H. & Pagel, M.D. (1991). The comparative method in evolutionary biology. — Oxford University Press, Oxford. Honza, M., Moksnes, A., Røskaft, E. & Øien, I.J. (1998). Survival of reed warblers Acrocephalus scirpaceus clutches in relation to nest position. — Bird Study 45, p. 104108. Howard, R. & Moore, A. (1991). A complete checklist of the birds in the world. — Academic Press, San Diego. Johnsgard, P.A. (1997). The Avian Brood Parasites. Deception at the Nest. — Oxford University Press, New York, Oxford. Järvinen, A. (1984). Relationship between the common cuckoo Cuculus canorus and its host, the redstart Phoenicurus phoenicurus. — Ornis Fenn. 61, p. 84-88. Lindholm, A.K. (1999). Brood parasitism by the cuckoo on patchy reed warbler populations in Britain. — J. Anim. Ecol. 68, p. 293-309. — — & Thomas, R.J. (2000). Differences between populations of reed warblers in defences against brood parasitism. — Behaviour 137, p. 25-42. Lotem, A. & Nakamura, H. (1998). Evolutionary equilibria in avian brood parasitism. — In: Parasitic birds and their hosts; Studies in coevolution (S.I. Rothstein & S.K. Robinson, eds). Oxford University Press, New York, Oxford, p. 223-235. — —, — — & Zahavi, A. (1992). Rejection of cuckoo eggs in relation to host age: a possible evolutionary equilibrium. — Behav. Ecol. 3, p. 128-132. — —, — — & — — (1995). Constraints on egg discriminationand cuckoo host co-evolution. — Anim. Behav. 49, p. 1185-1209. Moksnes, A. (1992). Egg recognition in chaf nches and bramblings. — Anim. Behav. 44, p. 993-995. — — & Røskaft, E. (1987). Cuckoo host interactions in Norwegian mountain areas. — Ornis Scand. 18, p. 168-172. — — & — — (1989). Adaptations of meadow pipits to parasitism by the common cuckoo. — Behav. Ecol. Sociobiol. 24, p. 25-30. — — & — — (1992). Responses of some rare cuckoo hosts to mimetic model cuckoo eggs and to foreign conspeci c eggs. — Ornis Scand. 23, p. 17-23. — — & — — (1995). Egg-morphs and host preference in the common cuckoo (Cuculus canorus): an analysis of cuckoo and host eggs from European museum collections. — J. Zool., London 236, p. 625-648. — —, — —, Bicík, V., Honza, M. & Øien, I.J. (1993b). Cuckoo Cuculus canorus parasitism on Acrocephalus warblers in Southern Moravia in the Czech Republic. — J. Ornithol. 134, p. 425-434.
AGGRE SSION OF CUCKOO HOSTS
627
— —, — — & Braa, A.T. (1991). Rejection behavior by common cuckoo hosts towards arti cial brood parasite eggs. — Auk 108, p. 348-354. — —, — —, — —, Korsnes, L., Lampe, H.M. & Pedersen, H.C. (1990). Behavioural responses of potential hosts towards arti cial cuckoo eggs and dummies. — Behaviour 116, p. 64-69. — —, — —, Hagen, L.G., Honza, M., Mørk, C. & Olsen, P.H. (2000). Common cuckoo Cuculus canorus and host behaviour at reed warbler Acrocephalus scirpaceus nests. — Ibis 142, p. 247-258. — —, — — & Korsnes, L. (1993a). Rejection of cuckoo (Cuculus canorus) eggs by meadow pipits (Anthus pratensis). — Behav. Ecol. 4, p. 120-127. — —, — — & Solli, M.M. (1994). Documenting puncture ejection of parasitic eggs by chaf nches Fringilla coelebs and blackcaps Sylvia atricapilla. — Fauna norvegica Series C, Cinclus 17, p. 115-118. Moskát, C. & Fuisz, T.I. (1999). Reactions of red-backed shrikes Lanius collurio to arti cial cuckoo Cuculus canorus eggs. — J. Avian Biol. 30, p. 175-181. — — & Honza, M. (2000). Effect of nest and nest site characteristics on the risk of cuckoo Cuculus canorus parasitism in great reed warbler Acrocephalus arundinaceus. — Ecography 23, p. 335-341. Øien, I.J., Honza, M., Moksnes, A. & Røskaft, E. (1996). The risk of parasitism in relation to the distance from reed warbler nests to cuckoo perches. — J. Anim. Ecol. 65, p. 147153. — —, Moksnes, A. & Røskaft, E. (1995). Evolution of variation in egg color and marking in European passerines: adaptations in a coevolutionary arms race with the cuckoo (Cuculus canorus). — Behav. Ecol. 6, p. 166-174. — —, — —, — —, Edvardsen, E., Honza, M., Kleven, O. & Rudolfsen, G. (1999). Conditional host responses to cuckoo Cuculus canorus parasitism. — In: Proceedings of the 22 International Ornithological Congress, Durban (N.J. Adams & R.H. Slotow, eds). Bird Life South Africa, Johannesburg, p. 3125-3145. Pagel, M.D. (1992). A method for the analysis of comparative data. — J. theor. Biol. 156, p. 431-442. Rohwer, S. & Spaw, C.D. (1988). Evolutionary lag versus bill-size constraints: a comparative study of acceptance of cowbird eggs by old hosts. — Evol. Ecol. 2, p. 27-36. Rothstein, S.I. (1982). Mechanisms of avian egg recognition: which egg parameters elicit responses by rejector species? — Behav. Ecol. Sociobiol. 11, p. 229-239. — — (1990). A model system for coevolution: avian brood parasitism. — Ann. Rev. Ecol. Syst. 21, p. 481-508. Røskaft, E. & Moksnes, A. (1998). Coevolution between brood parasites and their hosts: An optimality theory approach. — In: Parasitic birds and their hosts; Studies in coevolution (S.I. Rothstein & S.K. Robinson, eds). Oxford Univeristy Press, New York, Oxford, p. 236-254. — —, — —, Meilvang, D., Bicík, V., Jemelíková, J. & Honza, M. (2002b). No evidence for recognition errors in Acrocephalus warblers. — J. Avian Biol. 33, p. 31-38. — —, — —, Stokke, B.G., Moskát, C. & Honza, M. (2002a). The spatial habitat structure of host populations explains the pattern of rejection behavior in hosts and parasitic adaptations. — Behav. Ecol. 13, p. 163-168. — —, Orians, G.H. & Beletsky, D.L. (1990). Why do red-winged blackbirds accept eggs of brown-headed cowbirds? — Evol. Ecol. 4, p. 35-42.
628
RØSKAFT, MOKSNES, STOKKE, BICÍK & MOSKÁT
— —, Rohwer, S. & Spaw, C.D. (1993). Costs of puncture ejection compared with costs of rearing cowbird chicks for northern orioles. — Ornis Scand. 24, p. 28-32. Schlichting, C.D. & Pigliucci, M. (1998). Phenotypic evolution: a reaction norm perspective. — Sinauer, Sunderland. Sibley, C.G. & Ahlquist, J.E. (1991). Phylogeny and classi cation of birds. — Yale University Press, New Haven, Connecticut. Snow, D.W. & Perrins, C.M., eds. (1998). The birds of the Western Palearctic. — Oxford University Press, Oxford. Soler, J.J., Martinez, J.G., Soler, M. & Møller, A.P. (1999a). Genetic and geographic variation in rejection behavior of cuckoo eggs by European magpie populations: An experimental test of rejecter-gene ow. — Evolution 53, p. 947-956. — — & Møller, A.P. (1996). A comparative analysis of the evolution of variation in appearance of eggs of European passerines in relation to brood parasitism. — Behav. Ecol. 7, p. 89-94. — —, Soler, M., Pérez-Contreras, T., Aragón, S. & Møller, A.P. (1999b). Antagonistic antiparasite defences: nest defence and egg rejection in the magpie host of the great spotted cuckoo. — Behav. Ecol. 10, p. 707-713. Soler, M., Soler, J.J., Martinez, J.G., Pèrez-Contreras, T. & Møller, A.P. (1998). Microevolutionary change and population dynamics of a brood parasite and its primary host: the intermittent arms race hypothesis. — Oecologia 117, p. 381-390. Stokke, B.G., Moksnes, A. & Røskaft, E. (2002). Obligate brood parasites as selective agents for evolution of egg appearance in passerine birds. — Evolution 56, p. 199-205. — —, — —, — —, Rudolfsen, G. & Honza, M. (1999). Rejection of arti cial cuckoo (Cuculus canorus) eggs in relation to variation in egg appearance among reed warblers (Acrocephalus scirpaceus). — Proc. R. Soc., London B 266, p. 1483-1488. Takasu, F. (1998a). Why do all host species not show defence against avian brood parasitism: Evolutionary lag or equilibrium? — Am. Nat. 151, p. 193-205. — — (1998b). Modelling the arms race in avian brood parasitism. — Evol. Ecol. 12, p. 969987. — —, Kawasaki, K., Nakamura, H., Cohen, J.E. & Shigesada, N. (1993). Modeling the population dynamics of a cuckoo-host association and the evolution of host defences. — Am. Nat. 142, p. 819-839. Thompson, J.N. (1994). The Coevolutionary Process. — The University of Chicago Press, Chicago and London. Via, S., Gomiulkiewicz, R., De Jong, G., Scheiner, S.M., Schlichting, C.D. & van Tienderen, P.H. (1995). Adaptive phenotypic plasticity: consensus and controversy.— Trends Ecol. Evol. 10, p. 212-217. Wyllie, I. (1981). The cuckoo. — Batsford, London.
