SEARCH

SEARCH BY CITATION

Keywords:

  • benefits of philopatry;
  • ecological constraints;
  • Corvus corone;
  • social behaviour

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    The ecological factors that promote delayed dispersal of offspring in cooperatively breeding bird species are poorly understood. While single population studies have supported the view that natal dispersal is delayed as a consequence of lack of suitable breeding vacancies (ecological constraints hypothesis), recent theoretical models claim that habitat saturation cannot be the main factor leading to kin sociality.
  • 2
    The carrion crow (Corvus corone ssp.) is an ideal model to investigate the ecology of delayed dispersal. The occurrence of kin sociality is highly variable among European populations, and it is known to be determined by environmental rather than genetic factors. Here we compare juvenile dispersal, territory turnover, habitat saturation (number of competitors per breeding vacancy), variability of territory quality and territoriality of two crow populations that differ in social organization (extensive kin sociality in Spain vs. social monogamy in Italy).
  • 3
    We found that philopatry occurred in the less competitive and less variable environment. A review of the information available in literature on other crow populations qualitatively supports this result.
  • 4
    In crows, juvenile philopatry seemed to be associated with breeders’ year-round residency in territories. This association is widespread among birds and it has been previously explained through an ecological constraints perspective, with residency slowing territory turnover and therefore augmenting saturation. Our data do not support this explanation.
  • 5
    We suggest that year-round residency might play a direct role in determining kin sociality in crows. In Spain, adults occupy and defend the territory year-round, providing the offspring with a place where they enjoy a preferential access to resources that may represent an incentive to stay. Conversely in Italy, where adults often abandon their territories after breeding, the natal site does not have any special value for the offspring that hence disperse.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Kin societies in birds form when offspring delay natal dispersal and prolong their association with the parents. In most species living in kin groups, non-dispersing offspring forgo reproduction and usually help their parents to rear new young. Therefore, delayed dispersal is a precondition for kin-based cooperative breeding to arise.

Although the ecological factors that promote delayed dispersal have been discussed for decades (for a historical perspective of the dispute between the habitat saturation hypothesis and the benefits of philopatry hypothesis see Emlen 1982, 1991, Stacey & Ligon 1991; Koenig et al. 1992; Cockburn 1996; Hatchwell & Komdeur 2000) the actual factors that drive the formation of families are still unclear. Most single-population studies on cooperative species, including elegant experiments, have detected severe constraints on independent breeding (e.g. Woolfenden & Fitzpatrick 1984; Emlen & Wrege 1989; Pruett-Jones & Lewis 1990; Komdeur 1992; Walters, Copeyon & Carter 1992) providing empirical support for the ecological constraints model. This model predicts that offspring will ‘stay at home’ when suitable opportunities for independent breeding are scarce either because of lack of vacant territories (habitat saturation) or because the variability in habitat quality is so high that it pays to wait for a good territory instead of attempting to breed on a low-quality vacancy (saturation of high-quality territories). However, the model lacks of predictive power, as many non-cooperative species are as constrained as the cooperative ones in term of breeding opportunities (Ekman et al. 2001), suggesting that habitat saturation might not be a key factor in determining the expression of cooperative breeding (Koenig et al. 1992; Cockburn 1996). Furthermore, the model cannot explain why, under conditions of intense competition for breeding, philopatry should be preferred to a floating strategy, where roving individuals could sample many territories and detect vacancies more efficiently (Ekman et al. 2001).

In recent years, the search for the ecological correlates of delayed dispersal and helping behaviour has moved from the generalization of single-species studies to the application of comparative analyses and mathematical modelling. Some comparative analyses seemed to confirm the essential role of habitat saturation in the expression of delayed dispersal. Arnold & Owens (1998, 1999) showed that cooperative breeding in birds is associated with low annual mortality (but see criticism in Cockburn 2003) and year-round residency. These factors would slow population turnover, leading to an overcrowded environment that facilitates the expression of cooperative breeding (Arnold & Owens 1999; but see Hatchwell & Komdeur 2000).

Recent theoretical models suggest instead that delayed dispersal might be disjoined from the saturation of the habitat. Kokko & Lundberg (2001) showed how habitat saturation can arise either when survival of year-round resident breeders is high (i.e. slow territory turnover) or when the survival of disperser increases, but the effects of these two traits in selecting the dispersal strategy are opposite. When a population saturates the habitat as a consequence of high dispersers’ survival, delayed dispersal is selected against and cooperative breeding is unlikely to evolve.

While theory has developed rapidly in recent years, providing many new insights and fuelling the debate, few empirical studies have attempted to test the new ideas. The carrion crow (Corvus corone L.) exhibits extreme interpopulations variation in the extent of natal philopatry and cooperative breeding. In northern Spain, crows breed cooperatively in about 75% of the territories, whereas elsewhere in Europe helpers at the nest are either rare (e.g. 6% of the territories in an urban population in Switzerland, Richner 1990) or totally absent (Baglione, Marcos & Canestrari 2002a). Despite the phylogenetic predisposition to cooperative breeding in some bird taxa, corvids in particular (Cockburn 1996), in the carrion crow it has been shown experimentally that current environmental conditions determine the expression of cooperative behaviour (Baglione et al. 2002b). This species therefore represents a good model species to investigate further the ecology of cooperative breeding. In this study we compare two populations of crows that differ in social organization and in patterns of natal dispersal and we review the information available on other well studied European populations. The aim is to test whether the observed between-populations variability is consistent with the ecological constraints model, which would predict higher habitat saturation and/or higher variability in the territory quality in the cooperative population. Subsequently, we examine the hypothesis put forward by comparative analyses and theoretical models that year-round residency is, compared to habitat saturation, a better predictor of the occurrence of delayed dispersal than is habitat saturation (Arnold & Owens 1998, 1999, Kokko & Lundberg 2001).

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

study areas

We studied two different crow populations, one in northern Italy (1989–91 and 2001, 2002) and one in northern Spain (1995–2003). In Italy we studied non cooperative crows at two different agricultural sites, 10 km apart (45°N, 9°E). The two areas (10 km2 each) were similar, with a flat landscape dominated by rice fields, maize and, to lesser extent, meadows. Agriculture was intensive, with fields irrigated regularly and natural vegetation confined to the shores of irrigation canals. Crows nested in poplar tree (Populus sp. L.) plantations or in small patches of relict oak (Quercus robur L.) forest, with similar breeding densities at the two sites (6·3 and 6·2 pairs km2). Crows in Italy are socially monogamous, and non-breeders live year-round in non-territorial flocks. Occasional observations (n = 2, 6% of observed territories) of a ‘third bird’ in a territory have been reported (Baglione et al. 2002a; V. Baglione unpublished data), but the helping role of those crows could not be confirmed.

The habitat in rural northern Spain (42°N, 5°W) was markedly different, with low-intensity agricultural use (no irrigation) and a mosaic of crops (mainly wheat and oat), meadows, uncultivated land for grazing of sheep and cattle and oak (Quercus pyrenaica Willdenow) forest patches. Here 75% of crow territories are held by cooperative groups of up to nine individuals (mode 3 birds). Extra birds are either non-dispersing offspring that stay with their parents on the natal territories or immigrants that form coalitions with resident territorial birds. Groups may have up to three helpers involved in feeding the nestlings (for details on the Spanish crow society see Baglione et al. 2002a).

dispersal

Nestlings were banded with wing tags and coloured rings when they were about to leave the nest (28–32 days old). The banded juveniles (n = 79 in Italy, n = 188 in Spain) were followed within the study areas and the surrounding landscape. We visited the territories at least once a week during the first 5 weeks after offspring had fledged, when they still depend on their parents (Cramp & Perrins 1994), and thereafter at least once a month until the following breeding season. We concentrated the search for dispersing young on communal feeding grounds and communal roosts. To determine whether the behaviour of the parents played a role in provoking offspring dispersal in Italy, we surveyed five crow families until the offspring dispersed in 2002. We observed families for 30–60 min at least once a week for 5 weeks after fledging in order to document any evidence of parental aggression towards offspring.

habitat saturation

We measured habitat saturation (HO) in Italy and Spain as number of competitors per breeding vacancy arisen in a year time span (Kokko & Lundberg 2001). To do so in the two populations, we collected data on the number of individuals competing for territory vacancies and the proportion of breeding vacancies available for each year of study.

Both in Italy and Spain, we searched all new nests at the beginning of the breeding season. Afterwards, regular surveys allowed the detection of any late nesting attempts. We used different methods to estimate the number of non-territorial crows of the two populations. In 1989, 2001 and 2002 we counted crows in the Italian study areas at least three times throughout the breeding season, following a standard itinerary with a car. The extreme structural simplicity of the landscape allowed a very accurate bird census. We also surveyed the surroundings of the study sites, where we commonly observed other stable flocks of floaters. As same-year bird counts within the study plots gave almost identical results (variation between 1 and 3%), we believe that we assessed the actual number of birds settled within the study sites and that our censuses were not inflated by individuals occasionally coming from elsewhere. We therefore regarded the maximum number of crows observed during the censuses as the number of crows living on the study areas. The number of potential competitors for territories was obtained by subtracting 2 × number of active nests from the total number of individuals.

In Spain, floaters were a negligible component of the studied population during the breeding season (Baglione et al. 2002a). Small flocks of floaters were observed only twice within the study area in 5 years (nine and 13 individuals, respectively). Almost all individuals that do not own a territory either lived on their natal territory or immigrated to a new territory, where they associated with the resident birds. The total number of potential competitors for territory vacancies was therefore obtained by counting the extra-birds in the territories. To do so, we have recorded the group size of every territory since the beginning of our study. For 64–79% of territories we assessed group size every year by observing the territories from a favourable spot for at least 3 h (for details see Baglione et al. 2002a). In the remaining territories we obtained group size when we visited them for routine data collection (egg-laying sequence, nestlings’ growth, fledgling dispersal). Because group members usually form a cohesive group, those short visits to the territories (15–30 min) provided reliable information on group size.

The population sex ratio was known both in Spain (Baglione et al. 2002a) and in Italy, where it was derived from a large sample of adult crows (n = 121) that had been caught as part of provincial administration's pest management programmes (see Acquarone, Cucco & Malacarne 2001). The sex ratio of non-breeders was derived from the population sex ratio, allowing assessment of habitat saturation for each sex separately in both populations (Table 1).

Table 1.  Territory turnover, percentage of individuals competing for territory vacancies in the population and habitat saturation (number of competitors per breeding vacancy) in Spain (2000–2002) and Italy (2001 and 2002)
 YearTurnover (%) (no of individuals)Percentage of individuals competing for vacanciesSex ratio population??://Habitat saturation (HO)1
??//??//
  • 1

    The proportion of sexes among non breeders was derived from the sex ratio in the population as: proportion of non-breeding males = (proportion of males in the population −0·5 × proportion of breeders)/proportion of non-breeders; proportion of non-breeding females = 1 – proportion of non-breeding males.

Spain200020·0 (10)28·6 (7)39·8 4·71·3
200120·0 (15)45·5 (11)41·11 : 0·675·00·9
200218·2 (11)16·7 (6)38·5 4·82·3
Italy200122·2 (9)37·5 (8)55·81 : 0·618·12·0
200230·8 (13)37·5 (16)58·4 6·52·2

To measure the proportion of breeding vacancies available each year, we caught free-flying crows in 1999 and 2000 in Spain (n = 61) and in 2001 and 2002 in Italy (n = 41; see Baglione et al. 2002a for details on trapping system). Crows were banded with wing tags and an individual code of coloured rings and blood samples were collected. P2/P8 DNA-based identification (Griffiths et al. 1998) provided the sex of the carrion crows (for details on blood samples storage and DNA extraction see Baglione et al. 2002a,c).

In Italy, during the first days after banding, we observed crows with binoculars and spotting scopes until we saw them providing food to the nestlings. We therefore regarded them as breeders of that particular territory. In Spain, captured crows were assigned more easily to a specific territory, due to the virtual lack of floaters and to the scattered distribution of the territories (Marcos & Baglione 2003), and breeding status was assigned according to a parentage analysis based on eight polymorphic microsatallite loci (see Baglione et al. 2002c for details). In both populations we considered that a breeding vacancy arose when a breeder disappeared from a territory before the commencement of the breeding season. In Spain, in groups where reproduction was shared, we considered that a vacancy arose when the alpha-breeder disappeared. As almost all vacancies were occupied later by new territorial crows in both sites (100% in Italy, n = 15; 87% in Spain, n = 15), the proportion of breeding vacancies approximated the actual turnover of individuals in the breeding territories. Therefore for simplicity, we will refer hereafter to this datum as ‘territory turnover’.

The fate of the breeders that disappeared from the territories was unknown in most cases. Two individuals, one in Italy and one in Spain, were found dead. One Spanish male immigrated to another territory, probably after usurpation of his own territory, and one breeding female in Italy joined the non-breeding flock after her mate had disappeared. The lack of resightings during our regular surveys of the study areas, roosts and communal feeding grounds suggests that most of the missing individuals had died.

variability of territory quality

We defined the variability of habitat quality as among-territories coefficient of variation (CV) of the number of fledglings produced in a number of territories over the study period (according to Stacey & Ligon 1991). We measured the productivity of 19 territories in Italy and 20 territories in Spain over 3 (1988, 1989, 1991) and 5 years (1995–99), respectively, considering only those territories that could be recognized unequivocally throughout the study period (Baglione et al. 2002c). The difference between the CVs was tested according to Sokal & Braumann (1980). To ensure that the result was not influenced by the different duration of the sampling periods at the two study sites, we also calculated the Spanish CV on the productivity of territories in periods of three consecutive years (1995–97, 1996–98 and 1997–99) and compared the results.

When helpers have a positive effect on reproductive success, the CV of productivity of the territories can be a biased index of the variation of habitat quality because the difference between territories might be overestimated. Better territories produce more young that will become helpers, which will enhance further productivity and therefore also the difference with less successful territories. This potential bias may inflate the calculation of the CV only in Spain, where crows are cooperative, but this is conservative with regard to the conclusions of this paper (see Discussion).

territoriality

We used two different methods to determine the shape and the overlap of the breeding territories in the two populations.

Method A

We assessed territory shape by following crows’ feedings trips to the nest and mapping the location where they landed to resume foraging activity. Because crows attend only one nest, this method guaranteed that we always followed the crows belonging to the territory in which we were interested, even when such birds were unbanded. Observations were collected for at least 9 h in three different bouts (3 h each) for each territory. Such a sampling effort guarantees an accurate measurement of territory size (Spray 1978) in crows. In this way, we sampled 22 territories in Spain (from 1995 to 1998) and 12 territories in Italy (1989).

Method B

In territories where adult crows were banded, we mapped their position each time we visited the territory. We selected for analysis only those territories where we collected a number of locations corresponding at least to the minimum number of locations per territory taken with method A (n = 27), i.e. 19 territories in Spain, sampled in 1999 and 2000, and 11 in Italy in 2000 and 2001.

Territories were defined as the minimum convex polygons among 95% localizations of territory holders (Kenward 1987). Within areas, territory size did not differ significantly between years (Spanish territories, Kruskal–Wallis non-parametric anova, H = 6·64, d.f. = 5, P = 0·25; Italian territories, Kruskal–Wallis non-parametric anova, H = 1·0, d.f. = 2, P = 0·32) nor between sampling methods (Spanish territories, Kruskal–Wallis non-parametric anova, H = 3·42, d.f. = 1, P = 0·06; Italian territories, H = 1·00, d.f. = 1, P = 0·32). For each pair of adjacent territories (a and b, for example) we calculated the overlap as: surface shared/(surface a+ surface b). We used this value as measure of territory overlap because we could not always assess for each territory the actual surface shared with all the neighbours (some of which had not been sampled in most cases). As some territories contributed to more than one datum, we based the statistical analysis of territory overlap on a randomization (Manly 1991). We randomized the samples 1000 times, resampling the data without replacement, and calculated the t-statistic at each step, using pop tools version 2·6 (available on the internet at http://www.cse.csiro.au/poptools). The proportion of times the randomized t-values exceeded the actual t-value represents the P-value of this test.

Thirty territories sampled in Spain were surveyed year-round to assess the degree of residency of territory holders and the rate of intrusion of conspecifics. We visited the territories at least twice a week during the breeding period (15 March−15 July) and once a month outside the breeding season, recording the presence of territory holders and intruders. Group size was known for all those territories, and all the groups contained at least one banded individual (one or more retained offspring in 22 territories, adults and retained offspring in 8 territories). Because crows in Spain form cohesive and stable groups and retained offspring always associate with their parents on their natal territory (Baglione et al. 2002a), the consistency of group size and the presence of retained offspring reliably indicated the presence of the territory holders on the territory. This made it possible to also survey territories where some group members were unbanded (22 territories). Conversely in Italy, reliable information on year-round residency of territory holders and rate of intrusion could be collected only on territories where at least one of the breeders was banded (11 territories).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

juvenile dispersal

During the 5 weeks after fledging, a conspicuous proportion of banded fledglings (40% and 45% in Spain and Italy, respectively) was not resighted on their natal territories, with no significant difference between the two populations (Fig. 1). At this stage, crow fledglings still depend on their parents (Baglione, Pieri & Bogliani 1994; Cramp & Perrins 1994), suggesting that those that were not resighted had probably died. By the age of 81–90 days and thereafter, the proportion of fledglings remaining on their natal territories was significantly higher in Spain than in Italy. In Spain, this proportion declined steadily until the next breeding season, when 28% of the banded fledglings were still ‘at home’, acting as helpers in most cases (on average 71% of non-dispersing yearlings help at the nest, Canestrari unpublished data). Conversely in Italy, already at the end of their first summer, no fledglings were longer associated with their natal territory. Dispersal was confirmed for 11 Italian juveniles, which were observed joining large flocks of non-breeders and foraging on communal feeding sites within the study area after dispersal. None of them returned to the natal territory thereafter.

image

Figure 1. Proportions of banded offspring remaining on their natal territory throughout their first year in Italy and Spain. Differences were tested with Yates's corrected χ2 adjusted to multiple tests by applying the sequential Bonferroni correction (Holm 1979). Asterisks indicate P < 0·05.

Download figure to PowerPoint

Our observations did not suggest parental aggression as a mechanism of offspring dispersal in Italy. In the five territories surveyed in 2002 (828 min in 31 bouts distributed among 45 days, covering a range of ages of the offspring between 30 and 77 days), the parents never showed any aggressiveness toward their offspring. After the first 4 weeks they ceased feeding them and thereafter merely ignored their begging. It must be noted that detecting offspring eviction in the field might be difficult. However, our data are supported by previous less standardized observations carried out in 1989–91 (Baglione and Bogliani, unpublished data), which also suggested that parental aggressiveness is negligible in determining the dispersal decision of the offspring in the Italian population.

habitat saturation and variability in territory quality

In Spain the percentage of individuals potentially competing for territory vacancies averaged 39·8% between 2000 and 2002 (range 38·5–41·4%, Table 1). In 1999, such a percentage was 31·9. In Italy 55·8% and 58·4% of the total population of the study area did not own a territory in 2000 and 2001, respectively (in 1989 the percentage was 65·9). Even though Spanish males seemed to be less likely to disappear from their territories than Italian males (Table 1), the turnover of breeders did not significantly differ between the two populations (Fisher's exact test; males, P = 0·5; females, P = 0·76; both sexes, P = 0·27). The number of competitors per breeding vacancy (habitat saturation) was consistently higher in Italy than in Spain for both sexes (Table 1), with the exception of females in 2002.

The two study sites did not differ significantly in term of variability of territory quality. The coefficients of variation of fledglings’ productivity of territories in Spain (76·1; n = 20) was lower than in Italy (98·4; n = 19), but the difference was not significant (F19,18 = 1·54; P = 0·18). The result did not change qualitatively when we equated the sampling periods of the two sites. The CVs obtained for three periods of 3 consecutive years in Spain (72·6 in 1995–97; 78·7 in 1996–98 and 105·1 in 1997–99) did not significantly differ from the Italian CV.

A review of crow studies carried out in Europe supports the view that high saturation might be disjointed from the occurrence of delayed dispersal and cooperative breeding. Table 2 shows how other non-cooperative crow populations in Europe are characterized by a high proportion of non-breeding individuals, which are comparable to that of the Italian population. Although estimates of HO in terms of competitors per breeding vacancies are not available for those populations, such high numbers of non-breeders suggest that co-occurrence of strong saturation and lack of cooperative breeding is widespread among European crows. Table 2 also suggests that the expression of delayed dispersal is not restricted to the subspecies Corvus corone corone. Offspring retained in the natal territory beyond the first winter have been described to occur occasionally in the Swedish C. c. cornix population (Loman 1985). Furthermore, C. c. cornix shows variability in the timing of dispersal comparable to that of non-cooperative populations of C. c. corone (typically from a few weeks after fledging to the end of the first winter; see Cramp & Perrins 1994 for a review). This indicates that C. c. cornix also posses the capacity to adjust dispersal to current conditions and that our results are not biased by considering both subspecies.

Table 2.  Proportion of non breeders, occurrence of cooperative breeding and territorial system in well-studied European crow populations
Study areaSubspeciesBreeding density (pp/km2)Percentage of non-breedersOccurrence of helper at nestTerritorial systemReference
  • 1

    ‘Third bird’ in two territories, helping role unclear.

  • 2

    Calculated from Richner (1990).

  • 3

    Calculated from Loman (1985) (referred to in Cramp & Perrins 1994).

  • 4

    Offspring occasionally retained at home for 1 year.

Northern SpainCorvus c. corone2·038·5–41·1%75% of territoriesYear-round territorialityThis study
Northern ItalyCorvus c. cornix6·555·8–75%0%1Undefended home rangesThis study Bogliani et al. (1994)
Switzerland, Zurich airport andsurroundingsCorvus c. corone2·0–3·0Unknown0%Estimated maximum occupancy rateduring non-breeding season = 0·15Griesser, unpublisheddata
Switzerland, rural areaCorvus c. corone6‘High’0%Territories defended only during the breeding seasonRichner (1989)
Switzerland, urbanareaCorvus c. corone3653–66%26%Year-round territorialityRichner (1989,1990)
Southern SwedenCorvus c. cornix1·9–2·577%30%4Territories abandoned in summer and reoccupied in springLoman (1985)
ScotlandBoth2·3–3·050–73%0%Territories defended only during the breeding seasonPicozzi (1975), Charles (1972)
GermanyCorvus c. corone333–42%0%Territories abandoned in summer and reoccupied in late autumn; undefended in winterWittemberg (1988)

territoriality

Crows’ mean territory size differed between the two study sites (Mann–Whitney U-test, U = 234, P = 0·001), being larger in Spain (average ± SE = 0·13 ± 0·01 km2, n = 41) than in Italy (0·08 ± 0·01 km2, n = 23). Adjacent breeding territories only marginally overlapped in Spain (average proportion of overlap ± SE = 0·01 ± 0·01, n = 14), while in Italy the proportion of surface that two neighbours share (0·1 ± 0·03) was significantly higher (randomization test, P = 0·001). By considering all the neighbours, where possible, we found that a pair in Italy could share up to 70% of its territory with other pairs.

Crows in Spain showed a significantly higher rate of year-round territory occupancy (i.e. for each territory sampled, the number of times that the territory owners were found ‘at home’ divided by the total number of surveys; Mann–Whitney test, U = 21, P < 0·01; Fig. 2a), which did not correlate significantly with the size of the group that held the territory (Spearman's R = 0·16, n = 30, P = 0·42). Conspecific intrusion rate (measured, for each territory, as the number of times that at least one crow intruder was found foraging on that territory divided by the total number of surveys) was lower in Spain than in Italy, both in the breeding and non-breeding period (Mann–Whitney test, breeding U = 38, P < 0·01; non-breeding U = 12, P < 0·01; Fig. 2b). In Spain, we found a marginally non-significant positive correlation between group size and intrusion rate in the breeding season (Spearman's R = 0·3, n = 40, P = 0·06) but not in the non-breeding period (Spearman's R = 0·18, n = 30, P = 0·36).

image

Figure 2. (a) Territory occupancy rate during the non-breeding period in the two populations studied; (b) territory intrusion rate during the breeding and non-breeding season.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Although the role of ecology in the expression of cooperative breeding has been a central issue in behavioural ecology for decades, it still represents an unresolved puzzle. The widespread acceptance of the ecological constraints model (Emlen 1994) has been attributed more to its broad and unspecific definition of constraints (Hatchwell & Komdeur 2000; Kokko & Ekman 2002) than to its actual predictive power, which is in fact very low (Cockburn 1996). The model identifies habitat saturation, lack of potential mates, costs of dispersal and lack of high-quality territories as factors that prevent natal dispersal. As most avian species face one or more of these ‘constraints’, easy post hoc explanations that apply to almost any cooperative society have created the illusion of a high explanatory power of the model. In this paper we restricted the definition of constraints to habitat saturation, which is thought to be the most widespread cause of kin sociality among birds, and is certainly the most debated. Hereafter, the term ‘constraints’ will refer to habitat saturation only.

habitat saturation, variability of territory quality and geographical patterns of natal dispersal in crows

Recent theoretical models have proposed that delayed dispersal is disjoint from habitat saturation and that cooperative breeding can evolve in absence of such a constraint (e.g. Kokko & Lundberg 2001; Kokko et al. 2001; Perrin & Lehmann 2001), but only a few studies, reporting on species living in apparently unsaturated habitats, seem to support this view (Caffrey 1992; Williams, Lawton & Lawton 1994; Macedo & Bianchi 1997). However, as the suitability for breeding of unfilled vacancies is difficult to prove, such data offer only a weak support to the new theory. On the other hand, most studies on the relationship between habitat saturation and delayed dispersal have failed to show causation (Hatchwell & Komdeur 2000), and experimental evidence is restricted to single population studies, which leave uncertainty about the role of constraints in determining differences across species or populations. In this respect, the environmentally determined geographical variation of the social behaviour of the carrion crow represents an ideal model to assess the actual role of habitat saturation in shaping avian social systems.

Our findings contradict the expectation of the ecological constraints model (Table 3). The habitat turned out to be consistently more saturated in Italy, where crows are not cooperative, than in Spain, especially for crow males, which are more likely to help at the nest (Baglione et al. 2002a). It might be argued that a comparison of only two populations might be randomly biased and therefore represents an inconclusive test of the model. However, the review of data published on other European crow populations supports our conclusions by showing qualitatively a widespread lack of association between high saturation and occurrence of cooperative breeding.

Table 3.  Ecological constraints model's predictions vs. observed patterns of territory turnover, habitat saturation and variability of territory quality in the two studied crow populations, according to the fact that crows delay natal dispersal and help at the nest only in Spain
 Ecological constraints hypothesis (according to Emlen 1994)Ecological constraints hypothesis (according to Arnold & Owens 1998)Observed
Territory turnoverNo specific predictionSlower in Spain than in ItalySlower in Spain than in Italy (NS)
Habitat saturationHigher in Spain than in ItalyHigher in Spain than in ItalyHigher in Italy than Spain
Variability in territory qualityHigher in Spain than in ItalyNo specific predictionHigher in Italy than in Spain (NS)

The ecological constraints model predicts lack of delayed dispersal in homogeneous environments, regardless of the saturation of the habitat. When territories do not differ in quality, ‘staying at home’ does not offer any advantage to the offspring in terms of future reproduction (i.e. by queuing, a juvenile does not gain a better territory) or can also be detrimental, if philopatric individuals miss a vacancy elsewhere (Stacey & Ligon 1991; Emlen 1994). A relative higher homogeneity of the habitat in Italy (i.e. lower CV, see Methods) might have therefore reconciled our results with the ecological constraints model. Nevertheless, our data again showed the opposite pattern, namely the quality of territories was more variable, although not significantly, in Italy than in Spain (Table 3). Furthermore, our data cannot be reconciled with the interpretation of ecological constraints suggested by Arnold & Owens (1998), where a slow turnover of breeders (high survival) represents the permissive factor to cooperation to arise, as it leads to the saturation of the habitat. In crows, the slowest territory turnover is associated with the lowest saturation, showing that the relationship between the two factors is not as straightforward as thought previously.

Kokko & Lundberg (2001) showed theoretically how the alternative demographic routes that lead to saturation, i.e. high survival of resident individuals or high survival or migrants/dispersers, have contrasting effects on the expression of delayed dispersal (the former favouring it, the latter selecting against). In crows, average residents’ survival seems to be higher in Spain than in Italy, especially in males (males: 80·6% vs. 72·7%, respectively; females: 66·7% vs. 64%; both sexes: 75% vs. 68%; derived from Table 1), although the difference was not statistically significant. More data are needed to assess how saturation is achieved in Italy and in Spain, and in particular to estimate survival of dispersers in the two populations. However, to the extent we could explore it, the crow system seems to fit into Kokko and Lundberg's theoretical framework.

the role of year-round territoriality

Brown (1987) already noted that cooperative breeding in birds is especially common among year-round territorial species. More recent comparative analyses have shown that year-round territoriality is correlated with the expression of delayed dispersal and cooperative breeding (Arnold & Owens 1998, 1999). Such a link seems to exist in the crow as well. The Spanish cooperative crows showed a significantly higher degree of year-round residency in the territories than Italian non-cooperative crows. Furthermore, unlike in Italy, territories in Spain are used exclusively by their owners both during the breeding and non-breeding season. Information on territorial behaviour of other European populations (Table 2) is not fully comparable to that presented in our study, as data were collected in different ways. However, at least qualitatively, the review presented in Table 2 suggests that year-round residency is peculiar of cooperative populations.

Arnold & Owens (1999) interpreted the correlation between year-round territoriality and the occurrence of delayed dispersal under an ecological constraints perspective. Year-round territoriality would act by slowing territory turnover, which in turn would limit further the opportunities for independent breeding (saturation), leading eventually to natal dispersal to be delayed. Our data do not support this view, as year-round residency of the Spanish population is not associated with a higher degree of saturation. Instead, the low residency of Italian crows might reflect the benefit of a floating strategy for both adults and juveniles during the non-breeding period in those areas. Under these conditions, Kokko and Lundberg's model predicts high degree of habitat saturation and lack of kin sociality to co-occur in the populations, which is what we have observed.

We suggest that year-round residency might have a more direct role in the expression of delayed dispersal in crows because it confers intrinsic benefits to natal philopatry. The ‘safe haven hypothesis’ (Kokko & Ekman 2002) suggests that the presence of parents makes the natal territory a special place for juveniles, which gain access to defended resources and sometimes protection against predators (Ekman, Sklepkovych & Tegelström 1994; Kraaijeveld & Dickinson 2001; Pravosudova, Grubb & Parker 2001; Griesser 2003). This hypothesis explains the correlation between year-round territoriality and delayed dispersal in birds without attributing any role to habitat saturation, and predicts that juveniles will disperse early in life either when parents cannot afford prolonging the care of their offspring or where they switch to non-territorial behaviour after the breeding season, allowing free competition for resources within their home range. Crow parents are tolerant towards their offspring both in Italy and in Spain but, as a consequence of lack of territoriality during the non-breeding season, young crows in Italy do not have any incentive to stay at home. Therefore, it might be the case that the ecological factors that promote year-round residency in crows (e.g. predictability and defensibility of trophic resources, which are currently under investigation), are the ones that ultimately shape the social organization.

An alternative explanation might be that living in kin groups allows the crows to defend the territories year-round, and that therefore the social system is the cause of the observed pattern of geographical variation in territoriality. However, this seems to be unlikely as Spanish unassisted pairs also live in the territories year-round and we found no effect of group size on either the occupancy rate of territories or the frequency of conspecific intrusions in Spain during the non breeding period.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We are grateful to Hanna Kokko, Carolee Caffrey, Nick Davies, Jan Ekman and two anonymous referees for reviewing the manuscript. Thanks to Monica Pieri, Antonio Mangiarotti, Annalisa Gratteri, Pili Zumel and José Luis Robles for help in the field, and to Camilla Acquarone and Giorgio Malacarne for data on sex ratio of the Italian crow population. This study was financially supported by Ax:son Johnsons Stieftelse (to V.B.), Ministero Italiano dell’Università e della Ricerca − COFIN Program (to G.B.), and Parco Lombardo del Ticino (to D.C.). Birds trapping, banding and blood sampling were authorized by Junta of Castilla y León, Regione Lombardia e Parco Lombardo del Ticino.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Acquarone, C., Cucco, M. & Malacarne, G. (2001) Short-term effects on body condition and size of immunocompetent organs in the hooded crow. Italian Journal of Zoology, 68, 195199.
  • Arnold, K.E. & Owens, I.P.F. (1999) Cooperative breeding in birds: the role of ecology. Behavioral Ecology, 10, 465471.
  • Arnold, K.E. & Owens. I.P.F. (1998) Cooperative breeding in birds: a comparative test of the life history hypothesis. Proceedings of the Royal Society of London, Series B, 265, 739745.
  • Baglione, V., Marcos, J.M. & Canestrari, D. (2002a) Cooperatively breeding groups of the carrion Crow Corvus corone corone in northern Spain. Auk, 119, 790799.
  • Baglione, V., Canestrari, D., Marcos, J.M., Griesser, M. & Ekman, J. (2002b) History, environment and social behaviour: experimentally induced cooperative breeding in the carrion crow. Proceeding of the Royal Society of London, Seies B, 269, 12471251.
  • Baglione, V., Marcos, J.M., Canestrari, D. & Ekman, J. (2002c) Direct fitness benefits of group living in a complex cooperative society of carrion crows Corvus corone corone. Animal Behaviour, 64, 887893.
  • Baglione, V., Pieri, M. & Bogliani, G. (1994) Maladaptive adoptions in the hooded crow. Avocetta, 18, 147149.
  • Bogliani, G., Pieri, M. & Baglione. V. (1994) Home range size, risk of nest predation and choice of foraging habitat in two hooded crow populations. Ethology, Ecology and Evolution, 6, 412.
  • Brown, J.L. (1987) Helping and Communal Breeding in Birds. Princeton University Press, Princeton.
  • Caffrey, C. (1992) Female-biased delayed dispersal and helping in American crows. Auk, 109, 609619.
  • Charles, J.K. (1972) Territorial behaviour and the limitation of population size in Crows Corvus corone and Corvus cornix. PhD Thesis, University of Aberdeen, Aberdeen.
  • Cockburn, A. (1996) Why do so many Australian birds cooperate: social evolution in the Corvida? Frontiers of Population Ecology (eds R.B.Floyd, A.W.Sheppard & P.J.De Barro), pp. 451472. CSIRO Publishing, Melbourne.
  • Cockburn, A. (2003) Cooperative breeding in oscine passerines: does sociality inhibit speciation? Proceedings of the Royal Society of London, Series B, 270, 22072214.
  • Cramp, S. & Perrins, C.M. (1994) The Birds of the Western Paleartic, Vol. VIII. Crows to Finches. Oxford University Press, Oxford.
  • Ekman, J., Baglione, V., Eggers, S. & Griesser, M. (2001) Delayed dispersal: living under the reign of nepotistic parents. Auk, 118, 110.
  • Ekman, J., Sklepkovych, B. & Tegelström. H. (1994) Offspring retention in the Siberian jay (Perisoreus infaustus): the prolonged brood care hypothesis. Behavioral Ecology, 5, 245253.
  • Emlen, S.T. (1982) The evolution of helping. I. An ecological constraints model. American Naturalist, 119, 2939.
  • Emlen, S.T. (1991) Evolution of cooperative breeding in birds and mammals. Behavioural Ecology. An Evolutionary Approach, 3rd end (eds J.R.Krebs & N.B.Davies), pp. 301335. Blackwell Scientific Publications, Oxford.
  • Emlen, S.T. (1994) Benefits, constraints and the evolution of the family. Trends in Ecology and Evolution, 9, 282285.
  • Emlen, S.T. & Wrege, P.H. (1989) A test of alternate hypotheses for helping behavior in white-fronted bee-eaters of Kenya. Behavioral Ecology and Sociobiology, 25, 303319.
  • Griesser, M. (2003) Nepotistic vigilance behavior in Siberian jay parents. Behavioral Ecology, 14, 246250.
  • Griffiths, R., Double, M.C., Orr, K. & Dawson, R.J.G. (1998) A DNA test to sex most birds. Molecular Ecology, 7, 10711075.
  • Hatchwell, B.J. & Komdeur. J. (2000) Ecological constraints, life history traits and the evolution of cooperative breeding. Animal Behaviour, 59, 10791086.
  • Holm, S. (1979) A simple sequentially rejective Bonferroni test procedure. Scandinavian Journal of Statistics, 6, 6570.
  • Kenward, R.E. (1987) Wildlife Radio Tagging: Equipment, Field Technique and Data Analysis. Academic Press, San Diego.
  • Koenig, W.D., Pitelka, F.A., Carmen, W.J. & Mumme, R.L. & Stanback, M.T. (1992) The evolution of delayed dispersal in cooperative breeders. Quarterly Review of Biology, 67, 111150.
  • Kokko, H. & Ekman, J. (2002) Delayed dispersal as a route to breeding: territorial inheritance, safe havens, and ecological constraints. American Naturalist, 160, 468484.
  • Kokko, H., Johnstone, R.A. & Clutton-Brock, T.H. (2001) The evolution of cooperative breeding through group augmentation. Proceedings of the Royal Society of London, Series B, 268, 187196.
  • Kokko, H. & Lundberg, P. (2001) Dispersal, migration, and offspring retention in saturated habitats. American Naturalist, 157, 188202.
  • Komdeur, J. (1992) Importance of habitat saturation and territory quality for evolution of cooperative breeding in Seychelles warbler. Nature, 358, 493495.
  • Kraaijeveld, K. & Dickinson, J.L. (2001) Family-based winter territoriality in western bluebirds (Sialia mexicana): the structure and dynamics of winter groups. Animal Behaviour, 61, 109117.
  • Loman, J. (1985) Social organisation in a population of the hooded crow. Ardea, 73, 6175.
  • Macedo, R.H. & Bianchi, C.A. (1997) Communal breeding in tropical Guira cuckoos Guira guira: sociality in the absence of a saturated habitat. Journal of Avian Biology, 28, 207215.
  • Manly, B.F.J. (1991) Randomization and Monte Carlo Methods in Biology. Chapman & Hall, London.
  • Marcos, J.M. & Baglione, V. (2003) Adoptions in a co-operative population of carrion crows Corvus corone corone. Bird Study, 50, 306308.
  • Perrin, L. & Lehmann, L. (2001) Is sociality driven by the costs of dispersal or the benefits of philopatry? A role for kin discrimination mechanism. American Naturalist, 158, 471483.
  • Picozzi, N. (1975) A study of the carrion/hooded crow in north-east Scotland. British Birds, 68, 409419.
  • Pravosudova, E.V., Grubb, T.C. & Parker, P.G. (2001) The influence of kinship on nutritional condition and aggression levels in winter social groups of tufted titmice. Condor, 103, 821828.
  • Pruett-Jones, S.G. & Lewis, M.J. (1990) Sex ratio and habitat limitation promote delayed dispersal in superb fairy-wrens. Nature, 348, 541542.
  • Richner, H. (1989) Habitat-specific growth and fitness in carrion crows (Corvus corone corone). Journal of Animal Ecology, 58, 427440.
  • Richner, H. (1990) Helpers at the nest in carrion crows Corvus corone corone. Ibis, 132, 105108.
  • Sokal, R.R. & Braumann, C.A. (1980) Significance test for coefficients of variation and variability profiles. Systematic Zoology, 29, 5066.
  • Spray, C.J. (1978) Territorial behaviour of the carrion crow, Corvus corone L., in relation to food supplementation: an experimental study. PhD Thesis, University of Aberdeen, Aberdeen.
  • Stacey, P.B. & Ligon, J.D. (1991) The benefit-of-philopatry hypothesis for the evolution of cooperative breeding: variation in territory quality and group size effects. American Naturalist, 137, 831846.
  • Walters, J.R., Copeyon, C.K. & Carter, J.H. (1992) Test of the ecological basis of cooperative breeding in red-cockaded woodpeckers. Auk, 109, 9097.
  • Williams, D.A., Lawton, M.F. & Lawton, R.O. (1994) Population growth, range expansion, and competition in the cooperatively breeding brown jay, Cyanocorax morio. Animal Behaviour, 48, 309322.
  • Wittenberg, J. (1988) Langfristige Entwicklung einer Population der Rabenkrähe (Corvus c. corone) bei Braunschweig, ihre Zusammensetzung und ihr Einfluss auf andere Arten. Beiheft Veröffentlichungen Naturschutz und Landschaftspflege Baden-Württemberg, 53, 21112223.
  • Woolfenden, G.E. & Fitzpatrick, J.W. (1984) The Florida Scrub Jay: Demography of a Cooperative-Breeding Bird. Monographs in Population Biology no. 20. Princeton University Press, Princeton, NJ.