SEARCH

SEARCH BY CITATION

Keywords:

  • cooperative breeding;
  • dispersal distance;
  • sex-biased dispersal;
  • tropical birds;
  • vertebrate sociality

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Dispersal from the natal area, or between breeding sites, marks a critical transition for individual animals, and a fundamental demographic and genetic process. Limitations on dispersal in natural populations are important in predicting population dynamics and structure. Tropical habitats that can be naturally patchy are also being fragmented rapidly, and these habitats harbour many social species.
  • 2
    We studied dispersal in cooperatively breeding stripe-backed wrens (Campylorhynchus nuchalis) in the Venezuelan savanna, using 20 years of data from marked populations over 40 km2. We test the hypotheses that dispersal is limited by competition for social position and that such limitation is sex-specific.
  • 3
    Dispersal is strongly female-biased, and results mainly in attainment of breeding status in neighbouring groups. Reproductive success is determined by the structure of the breeding group, and is concentrated in a small minority of individuals.
  • 4
    Natal dispersal is delayed, and represents a dramatic shift from indirect contribution to kin production (helping) to the direct approach of breeding. Monogamy limits breeding opportunities, compared to other social species, and both natal and breeding dispersal are constrained to short distances, particularly when compared to non-cooperative breeders.
  • 5
    Males are more philopatric compared to females, for which more frequent and distant dispersal leads to greater reproductive success. Competition for breeding status in an established social order impedes dispersal, in part because experience and previous familiarity are important, particularly in males that rarely disperse to productive large groups.
  • 6
    Diverse patterns of sex bias in dispersal among cooperatively breeding birds can be explained by the comparative diversity of reproductive alternatives available to males and females. A ‘breeding diversity’ hypothesis is illustrated by comparison with male-biased dispersal in brown jays.
  • 7
    Social resistance to movement establishes a brake on gene flow and demographic interaction among populations that has important implications for population viability.

Introduction

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

Movement of individuals is important in determining the dynamics of populations and their genetic constitution, particularly when dispersers breed (Hanski & Gilpin 1991; Brawn & Robinson 1996; Hanski 1997; Beier & Noss 1998). Understanding interactions among intrinsic constraints, environmental barriers and selective pressures favouring dispersal is essential to predicting the effects of habitat fragmentation on the viability of populations (e.g. Paradis et al. 1998; Breininger 1999; Loison et al. 1999; Travis & Dytham 1999). Few empirical studies have assessed the importance of dispersal for population structure and dynamics in fragmented habitats (but see e.g. Smith et al. 1996; Stith et al. 1996; Baker, Mather & Hughes 2001; Girman et al. 2001), and very few have considered social animals in tropical habitats. Recent studies of factors leading to dispersal in social species have focused on the balance of selective forces favouring early natal dispersal, compared to philopatry or delayed dispersal (Stacey & Ligon 1991; Walters, Doerr & Carter 1992; Emlen 1994, 1995; Koenig et al. 1992; Daniels & Walters 2000b; Koenig et al. 2000; Pen & Weissing 2000). This balance is often different for males and females, resulting in a sex bias in frequency, distance or age of dispersal from the natal area, or in movement by breeders. In most birds, dispersal is female-biased with some exceptions in the families Laridae, Anatidae and Corvidae, and these patterns can be explained by asymmetries in resource-holding potential and reproductive strategies (Brown 1974; Greenwood 1980; Brown & Brown 1990; Craig & Jamieson 1990; Curry & Grant 1990; Koenig & Stacey 1990; Woolfenden & Fitzpatrick 1990; Langen 1996a,b; Clarke, Saether & Roskaft 1997; Wolff & Plissner 1998; Williams 2000). Ultimately, dispersal can reduce inbreeding (Waser & Jones 1983; Shields 1987; Clutton-Brock 1989; Keane 1990; Pusey & Wolf 1996; Dieckmann, O’Hara & Weisser 1999) and competitive effects (Greenwood 1980; Waser 1985).

Sociality can limit dispersal options by concentrating breeding opportunities in a few positions in social groups. Sociality can also increase alternatives for potential dispersers by making permanent or temporary philopatry a viable fitness strategy, and by facilitating exploration from the safety of the natal group. These forces can produce dispersal that is curtailed in both frequency and distance, that is strongly sex-biased, and that favours older, more experienced individuals (Jones 1986; Zack & Rabenold 1989). In most social birds a delay in natal dispersal is typical, and the probability of dispersing at all is often higher in females, while males are more likely to achieve breeding status in their natal groups (Walters 1990; Koenig et al. 1992; Russell & Rowley 1993). Helping by these philopatric individuals to rear the offspring of others (often siblings) is then the defining characteristic of cooperative breeding. Such helping behaviour is often compensated by resulting increased productivity of close kin (indirect fitness gains), improved survival and/or increased eventual fecundity of the donor (Rabenold 1984; Stacey & Koenig 1990; Emlen & Wrege 1991; Cockburn 1998).

Beyond physical limitations on movement, it is often further argued that very short movements, resulting in highly viscous dispersal patterns, are favoured by the risks of exposure to predators while searching for openings, and by the difficulty of assessing accurately the potential rewards of particular opportunities based on habitat quality or group characteristics (Brown 1987; Daniels & Walters 2000b). Such difficulties would be greater in social species in which breeding positions created by breeder deaths are scattered, habitat characteristics are complex and group composition is important in determining productivity.

Stripe-backed wrens (Campylorhynchus nuchalis) are cooperatively breeding birds of South American savannas and dry forests, in which non-breeding helpers greatly improve reproductive success. Most individuals never attain breeding status, the great majority of breeding is by monogamous pairs with helpers and close inbreeding is avoided (Rabenold 1990; Rabenold et al. 1991b; Piper & Slater 1993; see also Haydock et al. 1996). Previous field experiments have demonstrated that individuals (particularly the more dispersive females) discriminate good breeding opportunities and favourable helping/breeding trade-offs, and that contests for breeding positions promote short-distance dispersal and philopatry by favouring older resident or local individuals (Zack & Rabenold 1989; Rabenold 1990; Haydock and Rabenold, unpublished data).

We ask whether the effects of behavioural constraints associated with sociality can explain the pattern of dispersal observed in this species, and specifically whether distance and frequency of dispersal are consistent with the differing advantages of dispersal for males and females. The ‘experienced disperser’ hypothesis explains both delayed dispersal and very viscous dispersal by proposing that they are determined by experience produced by a combination of age and proximity of the home group to a breeding opportunity. This hypothesis could also apply more strongly to one sex than the other, or apply on different spatial scales (within-population vs. between-population dispersal). The viability of such populations would be reduced if they were both handicapped by low effective population size (Ne; due to strong social dominance in breeding) and by truncated dispersal potential. We use 21 years of field data on naturally occurring dispersal in large marked populations to ask: (1) is dispersal limited to neighbouring groups, by more experienced individuals, as expected from the experienced-disperser hypothesis above; (2) is dispersal strongly female-biased, as would be expected if females were better able to invade and usurp breeding status in non-natal groups than males that are more constrained by within-group dominance hierarchies; (3) is inheritance of natal-group breeding position by males strictly age-related, and is acquisition of breeding position by dispersers also determined by competition from other males; (4) is the correlation between age and competitiveness stronger in males than in females, leading to a longer delay in breeding among males; (5) is there a sex bias in the reproductive success of dispersers; and (6) is dispersal by either sex sensitive to the relative same-sex competitive environments, or relative reproductive potential, in origin and destination groups? These questions lead to the larger issue of whether sociality reduces the viability of small populations in increasingly fragmented habitats.

Methods

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

study site and populations

The study area is on Hato Masaguaral in the low, seasonal savanna (llanos) of Venezuela, Estado Guárico (8°30′ N, 67°36′ W). Vegetation is patchy, and social groups of stripe-backed wrens are concentrated in scattered groves of leguminous trees on slightly drier soils, dominated by the genera Albizia, Pithecellobium, Enterolobium, Guazuma and Cassia. Masaguaral is lightly grazed, and the surrounding landscape is mainly agricultural (rice and cattle), so that very few groups of stripe-backed wrens exist within 2 km outside the study area (see Rabenold 1990 for a fuller description of the study area). The sexes of stripe-backed wrens are physically monomorphic, but differ in patterns of incubation and duetting that occurs only between opposite-sex partners. Juveniles have diagnostic plumage and iris colour. Sex and parentage have been confirmed for hundreds of individuals with DNA ‘fingerprinting’ using hypervariable minisatellite loci hybridizing with Jeffreys’ probes (33·6 and 33·15) of digests of genomic DNA (Rabenold et al. 1990, 1991a,b). For individuals whose sex was still uncertain, we used PCR techniques with sex-specific primers (1237 L and 1272H) and the protocol developed by Kahn et al. (1998).

Social groups of two to 12 adults occupy well defined, all-purpose territories year-round, and all members participate in boundary enforcement, nest maintenance and care of young. Groups normally contain a single breeding pair, although multiple paternity occurs with immigrant female breeders, and adult helpers are most often at least half siblings to juveniles. The presence of helpers strongly influences reproductive success, and overt competition among adults is strong for breeding positions in large groups (Rabenold 1984, 1985, 1990; Zack & Rabenold 1989). Changes in group membership are recorded easily, consisting mainly of recruitment of juveniles and replacement of deceased breeders by immigrants.

The study area occupied approximately 40 km2 of varied savanna with natural habitat breaks between slightly raised ridges, where nesting trees and wren groups are clustered. Movement was inhibited between these clusters, and we identified five distinct populations, each with different floristics. Marking with uniquely coloured leg bands and monitoring of demography was begun in 1977 in two central populations, and data were gathered in three other populations starting in 1985. Results described below have been consistent throughout the study area. Complete censuses were performed twice each year in the dry and wet seasons, through observations at roosting nests (particularly at dawn and dusk) and on the traditional territories, often employing acoustic playback of vocalizations and netting. Most often, studies of breeding resulted in weekly visits to groups and detailed information concerning intergroup movements. Censuses were judged to be complete when two successive counts revealed no additional individuals. Very few individuals (< 1%) have ever been found that were not recorded in the previous census. Reproductive success was recorded as number of independent juveniles (4–8 months old) surviving to the semi-annual dry-season census. Other measures of reproductive success, including number of fledglings and 1-month juveniles, are not uniformly available for all groups in all years. Eighty-five per cent of individuals bore colour bands on average, and unmarked immigrants were identifiable because all other group members were marked normally.

Some dispersal will inevitably be missed in a finite study area with finite census effort (Koenig, Van Vuren & Hooge 1996; Koenig et al. 2000); however, several lines of evidence suggest that error is low. First, the study area is very large compared to the dispersal distances detected, and the maximum detectable distance of 8 km is more than 20 times the average detected distance within populations (75% of movements) and five times the average distance traversed between populations. Secondly, very few wren groups have existed outside the study area within 2 km, because appropriate habitat is replaced by treeless savanna, thick scrub or agriculture. None the less, these areas were surveyed annually, and fewer than 1% of movements (0·25% of banded individuals) were detected in this buffer zone. Thirdly, substantial rates of undetected dispersal would require higher survival for non-breeders of dispersal age than for established breeders and a radically bimodal distribution of dispersal distances, but neither seems plausible.

defining and measuring dispersal

Our definitions of dispersal generally follow those of Greenwood (1980), as in many other studies (e.g. Waser & Jones 1983; Clarke et al. 1997; Wolf & Plissner 1998), modified for application to social animals. We define dispersal as the movement of individuals established in an origin group to a destination group resulting in a lasting change in membership (always > 1 month, often years). This movement can be from a natal or non-natal group to a breeding (principal) position or to a non-breeding (auxiliary) position. We define ‘natal dispersal’ as movement from natal group (as a juvenile or non-breeding auxiliary adult) to another group where principal status is attained, establishing breeding potential. Natality is known with varying degrees of precision, from genetic determination of parentage to banding fledglings or juveniles recognized by plumage or iris colour. ‘Breeding dispersal’ is movement from principal (breeding) status in one group to principal status in another. ‘Auxiliary dispersal’ is movement from non-breeding status in one group to non-breeding status in another, as well as the relatively rare movement from principal status to auxiliary status in another group. ‘Effective dispersal’ is movement resulting in offspring production by the disperser. Individuals can move more than once, contributing repeatedly to our data set, but we consider each move an independent event (normally separated by years). In this report, we consider only dispersal within populations, not movement between populations, because such relatively long-distance dispersal between major habitat patches appears to be governed by more stringent conditions than the much more common local, within-population dispersal.

To map dispersal and dispersion of groups, high-resolution aerial photos from the Ministerio del Ambiente de Venezuela were digitized and analysed using the Geographical Information System program ArcView 3·0. Nest trees for each group in each year were located, and we calculated the area under study for each population each year using the minimum convex polygon enclosing the territories. We assumed that dispersers moved directly from the origin group's nest tree to that of the destination group (Fig. 1). These observed dispersal distances were compared to expected values derived from all distances between nest trees. We defined core areas of territories as 50 m radii around active nest trees to measure ‘social distance’ traversed. We used systat 7·0 software for all statistical analyses, relying mainly on non-parametric tests because of the categorical nature of many variables.

image

Figure 1. An example of within-population dispersal in stripe-backed wrens, from the Samán population in 1977. Circles represent 50 m radii from nest trees, and lines represent dispersal events (white for females; black for males). Solid lines represent natal dispersal, dashed lines breeding dispersal, and dots auxiliary dispersal.

Download figure to PowerPoint

Results

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

Over 21 years (1977–98), 222 social groups of stripe-backed wrens were studied at Hato Masaguaral, for a total of 1800 group-years. Few groups remained stable over that time, but groups (lineages) were normally faithful to the same territory over many years, and rarely moved to new areas. Of 1599 individuals marked during this study, 307 changed group membership (dispersed) at least once within populations. Another 89 individuals dispersed between populations, but we do not consider them here. The overall frequency of movement from the group of initial banding is nearly 25%. Most dispersal is by single individuals, but in some cases (32 events) two to four individuals moved together.

Previous studies have shown a strong relationship between group size and reproductive success, associated with the increased feeding rates and defence of nestlings by auxiliaries in large groups (Rabenold 1984, 1985, 1990; Zack & Rabenold 1989). This relationship is apparent even as group size changes on the same territory. Using data for all groups in all years, we analysed the effect of different populations, years and group size on the number of juveniles produced per breeding female per year (Fig. 2). The strongly positive relationship between group size and reproductive success did not differ significantly among populations, but the strength of the relationship can vary among years. We ask below whether dispersal is sensitive to this relationship. We have found no relationship between group size and survival of adults (all regressions, for all age/status classes, P > 0·10).

image

Figure 2. Effects of group size on annual reproductive success (number of juveniles at the end of the breeding season) in stripe-backed wrens (error bars show the mean standard error, and sample size in group-years).

Download figure to PowerPoint

Most dispersal is by auxiliaries leaving their natal group and acquiring principal status in another group (natal dispersal − 220 of 375 events, 59%). Breeding dispersal (one principalship to another) is less common (99 events − 26%), and auxiliary dispersal is least common (56 events − 15%). We consider these classes of dispersal in that order, separating and comparing males and females.

natal dispersal

Frequency

Nearly twice as many dispersing stripe-backed wrens are female (139 of 214 events in which sex was known − 65%) as male (75 events, 35%; Table 1). The route to principal status varies strongly with sex. Of 393 females banded in their natal group as juveniles or auxiliaries, 139 (35%) obtained principalship by dispersal, while only 10 (3%) obtained principal status by remaining in the natal group (χ2 = 111·69; P < 0·001). In contrast, of 480 males banded as juveniles or auxiliaries, only 74 (15%) obtained principalship by dispersal, while 88 (18%) obtained principal status by remaining in the natal group (male tendency to philopatry, χ2 = 1·21; P = 0·27; male–female difference χ2 = 81·52, P < 0·001). During the ages of 2–4, when most dispersal occurs, annual male survival was 75% compared to 50% for females (2–3 years: χ2 = 5·72, P = 0·02; 3–4 years: χ2 = 3·548, P = 0·06). At earlier and later ages, males and females showed similar survival rates.

Table 1.  Number of dispersal events and dispersal distances in stripe-backed wrens by sex and dispersal category. Numbers under dispersal categories are total sample size, individuals of unknown sex not included (six natal, five breeding). Observed dispersal distances, measured in metres and core areas traversed (social distance) are compared to expected distances if random within population
 Number of dispersal eventsDispersal distance (n)Number of cases observ. > expect.* distance (social)Number of cases expect. > observ.* distance (social)
  1. Female-biased dispersal was observed in all categories of dispersal (all P < 0·001): natal: 1·85:1; breeding: 3·27:1; auxiliary: 3:1. *Wilcoxon signed ranks test, all P < 0·01 (see text). **Mann–Whitney U-test, female vs. male natal dispersal, P = 0·04.

Natal (n = 220)
 Female139388 m**19 (29)112 (100)
 Male 75331 m**10 (15) 61 (55)
Breeding (n = 99)
 Female 72349 m 8 (13) 55 (50)
 Male 22376 m 3 (4) 16 (15)
Auxiliary (n = 56)
 Female 42272 m 3 (8) 36 (31)
 Male 14319 m 0 (1) 14 (13)
Effectiveness

Half of female natal dispersal resulted in some direct fitness gain: 63 of 136 females (46%) produced at least one juvenile after dispersing, while only 27 of 74 males (36%) did so (χ2 = 1·89; P = 0·17). Using average juvenile production per year as the measure of reproductive success, the direct fitness benefit of natal dispersal was nearly twice as great for females:  = 0·71 ± 0·09 for 136 females, compared to  = 0·37 ± 0·07 for 74 males (Mann–Whitney U= 5818, P = 0·04). Males tend to disperse disproportionately to small groups where reproductive success is likely to be low (Fig. 3; χ2 = 25·85; P < 0·001, n= 62), while the sizes of groups to which females disperse did not differ from the prevailing distribution of group sizes at the time of dispersal (χ2 = 7·212; P = 0·41, n = 111). Reproductive success differs between philopatric and dispersing males. Of 182 males that acquired principal status, 50 of 88 (57%) inheritors (breeding in their natal group) reproduced successfully, and 27 of 74 (36%) dispersers produced at least one juvenile (χ2 = 5·551, P = 0·02). Annual reproductive success was higher for inheritors ( = 0·71 ± 0·09) than for dispersers ( = 0·37 ± 0·07) (Mann–Whitney U= 4043, P = 0·004).

image

Figure 3. Frequency of natal dispersal by males to groups of various sizes. The expected frequency is based on the distribution of group sizes present in the year that dispersal occurred.

Download figure to PowerPoint

Distance

Among natal dispersers, females disperse further than males:  = 388 ± 22 m for 131 females;  = 331 ± 29 m for 71 males (Mann–Whitney U= 5468·5, P = 0·03; Table 1). Mean territory diameter is approximately 150 m, so that such distances are sufficient to carry the average disperser to a destination two territories removed from its origin. Females moved to adjacent or next-adjacent groups in 55 of 131 cases (42%), while males moved within this radius in 24 of 71 cases (34%). We compared observed dispersal distances to three null models: (1) random destinations among groups extant in the year of dispersal; (2) random destinations among the closest four neighbouring (usually contiguous) groups; and (3) random destinations among groups with actual same-sex breeding vacancies created by principal deaths in the year of dispersal or preceeding. Natal dispersers of both sexes move shorter distances than those expected to random destinations within the population in the same year (females: Wilcoxon signed-ranks, Z= 7·87, P < 0·001, Fig. 4a; males: Wilcoxon signed ranks, Z= 6·48, P < 0·001, Fig. 4b, Table 1). In contrast, observed dispersal by both males and females covered longer distances than those expected if dispersal were limited to the immediate neighbourhood (females: Wilcoxon signed ranks, Z = 6·62, P < 0·001, males: Wilcoxon signed ranks, Z = 3·10, P = 0·002). Both males and females dispersed shorter distances than those expected if dispersal were random with respect to breeding vacancies (females: Wilcoxon signed ranks, Z = 6·72, P < 0·001; males: Wilcoxon signed ranks, Z = 5·05, P < 0·001; Fig. 5a.b; expected female  = 585 ± 19 m, expected male  = 524 ± 23 m; both 200 m greater (+ 50%) than observed means).

image

Figure 4. Female (a) and male (b) natal dispersal. Actual distance (m) traversed is plotted vs. expected (if random destination groups within population). The data under the line represent cases where the expected distance is larger than the observed dispersal distance. The data above the line represent cases where the observed dispersal distance was larger than expected.

Download figure to PowerPoint

image

Figure 5. Female (a) and male (b) natal dispersal. Actual distance traversed (m) is plotted vs. expected if random from breeding opportunities present (see text).

Download figure to PowerPoint

Males and females crossed few core areas (50 m radii around nest trees) ( = 0·46 ± 0·07 for 130 females;  = 0·32 ± 0·06 for 71 males; Mann–Whitney U= 4834·5, P = 0·50). Both females and males moved shorter ‘social distances’, crossing fewer core areas than expected if destinations were chosen randomly from all groups in that year (females: Wilcoxon signed ranks Z = 6·31, P < 0·001; males: Wilcoxon signed ranks Z = 5·87, P < 0·001, Table 1). Social distances of female natal dispersers were shorter than expected even when compared to the immediate neighbourhood (Wilcoxon signed ranks Z = 3·35, P = 0·001), but male distances did not depart from expectation.

Age and composition of origin group

For those individuals captured as juveniles and for which exact age of natal dispersal could be determined, the mean age of dispersal for females ( = 1·9 years, n= 59) was significantly less than the mean age of dispersal for males ( = 2·9 years, n= 37; Mann–Whitney U= 532·5, P < 0·001). Females gain breeding status almost exclusively by dispersing (above), and the main impediment in obtaining a principal position appears to be competition with females from other social groups. In contrast, most males gain the principal position through philopatry, and males that gain principalship in the natal group are almost always the oldest male auxiliaries or are in the oldest age class. Of 77 groups in which all individuals were sexed and aged, the ascending male was either the only male (n = 26, 34%), the oldest (n = 36, 47%), or in the oldest age class (n = 14, 18%). In only one case (1% of total) was the male intermediate between older and younger males. Dispersing males, however, were less likely to be older: of 60 natal groups, the disperser was the only male (n = 12, 20%), oldest (n = 19, 32%), oldest class (n = 20, 33%), intermediate (n = 1, 1%) or the youngest (n = 8, 13%).

Both male and female natal dispersers tended to leave small and mid-size groups disproportionately (females χ2 = 21·59; P = 0·01, n= 111, Fig. 6; males χ2 = 16·25; P = 0·02, n= 61). In addition, both male and female dispersers moved from groups with more same-sex auxiliaries to groups with fewer (females: 58 of 66 cases, Wilcoxon signed-ranks Z = 5·892, P < 0·001; males: 39 of 40 cases, Wilcoxon signed ranks: Z= 5·464, P < 0·001). However, because potential dispersers are concentrated in large groups, while breeding positions are limited to one per group regardless of group size, such a pattern would be expected by chance. If dispersers moved at random, we would expect a difference of 1·7 between the size of the origin and destination groups, compared to the difference of 2·2 observed for females and 2·5 observed for males.

image

Figure 6. Frequency of dispersal from groups of various sizes by females. The expected frequency is based on the distribution of group sizes present in the year that dispersal occurred.

Download figure to PowerPoint

breeding dispersal

Dispersal from one breeding position to another generally occurs when a mate dies and is not quickly replaced, and when a mate (almost always a female) abandons a position in favour of another. Breeding dispersal is strongly female-biased (72 female, 22 male events; χ2 = 26·60, P < 0·001, Table 1), and more strongly female-biased than natal dispersal (χ2 = 4·10, P = 0·04). Since adult survival is indistinguishable between the sexes (62% annually, Rabenold 1990), the extra female dispersal appears to represent mate desertion by females. We detected no difference between the sexes in effectiveness of breeding dispersal, measured as probability of reproduction or as average annual juvenile production, nor in average age of breeding dispersal.

Breeding dispersal covered similar distances for males and females (male  = 375 m ± 87, female  = 348 m ± 31, Table 1) and these distances did not differ significantly from natal dispersal. Breeding dispersal was viscous for both sexes compared to all possible destinations (females Wilcoxon signed ranks Z = 5·48, P < 0·001; males Z = 3·09, P = 0·002), but further than the immediate neighbourhood only for females (Wilcoxon signed ranks Z = 3·40, P < 0·001; males Z = 1·53, P = 0·13). Males and females traverse similar social distances during breeding dispersal (male  = 0·58 ± 0·19 core areas, female  = 0·51 ± 0·10; Mann–Whitney U= 558·5, P = 0·60), and fewer than expected at random (females Wilcoxon signed ranks Z = 4·65, P < 0·001; males Z = 3·08, P < 0·01, Table 1). Social distances of female and male breeding dispersal were not shorter than expected when compared to the immediate neighbourhood (Wilcoxon signed ranks Z = 0·52, P = 0·60 for females; Z = 0·38, P = 0·70 for males).

Considering origin and destination group sizes of both males and females, we found no departure from the expected, prevailing distribution of group sizes. However, females dispersed to unaided pairs much less frequently than expected (χ2 = 14·21, P < 0·001), and tended to leave unaided pairs more frequently than expected (χ2 = 2·73, P = 0·10).

auxiliary dispersal

Dispersal from one auxiliary position to another accounted for 15% of all dispersal (n = 56). Most of these movements occurred in concert with other dispersers (n = 32), normally relatives (n = 28 siblings, parent/offspring, cousins; known first-order relatives in 15 cases). Auxiliary dispersal was strongly female-biased (42 female, 14 male events; χ2 = 14·00, P < 0·001, Table 1). In nearly half of these cases, auxiliary dispersers later rose to breeding status after a delay of at least a year, with or without another move (eight females after another move, seven in place; two males after another move, two in place). Auxiliary dispersal occurred before the second year on average, and the age of dispersal was not different between males and females (females:  = 1·62 years, n= 17; males:  = 1·75 years, n= 8; Mann–Whitney U= 63, P = 0·76).

Auxiliary dispersal covered similar distances for males and females (male  = 318 m ± 47, female  = 272 m ± 27, Table 1) and these distances were shorter for females than natal dispersal ( = 331 m ± 29, n= 131, Mann–Whitney U = 1855·5, P = 0·01). Auxiliary dispersal was viscous for both sexes compared to all possible destinations (females Wilcoxon signed ranks Z = 5·12, P < 0·001; males Z = 3·29, P = 0·001, Table 1), but consistent with expected distances drawn from the immediate neighbourhood (females Wilcoxon signed ranks Z = 0·78, P = 0·43; males Z = 1·66, P = 0·10). Males and females traverse similar social distances during auxiliary dispersal (male  = 0·07 ± 0·07 group core areas, female  = 0·31 ± 0·08; Mann–Whitney U= 331, P = 0·11), and fewer than expected at random (females Wilcoxon signed ranks Z = 4·26, P = 0·001; males Z = 3·05, P < 0·001, Table 1). Social distances of female and male auxiliary dispersal did not differ from expected when compared to the immediate neighbourhood (Wilcoxon signed ranks Z = 0·17, P = 0·87 for females; Z = 1·58, P = 0·11 for males).

We found no tendency, for either male or female auxiliaries, to disperse to or from groups of particular sizes, but females leave groups with a large number of individuals of the same sex relative to the destination group (Wilcoxon signed ranks: Z = 2·825, P = 0·005), while males tend to do the same (Wilcoxon signed ranks: Z = 1·801, P = 0·07).

Discussion

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

Dispersal by stripe-backed wrens is strongly delayed, female-biased, and generally constrained to short distances, particularly when compared to noncooperative breeders. Indirect contributions to fitness by effective philopatric, nonbreeding helpers are ubiquitous (75% of individuals never disperse and 65% never breed), but dispersal results mainly in immediate attainment of breeding status in already existing groups. These patterns establish a direct link between dispersal and gene flow, but also illustrate impediments to movement that curtail gene flow. Individuals of all age/sex/status classes disperse at least occasionally. The group in which breeding status is attained strongly affects ultimate fitness, and direct contributions to future gene pools are concentrated in a small minority of individuals. It is also important that these basic patterns have been consistent from year to year over 21 years of study (see Rabenold 1984, 1985, 1990 for previous analyses), as pointed out by Clarke et al. (1997). Competition for breeding positions in the context of an established social order, in which experience and previous familiarity appear to be important (Zack & Rabenold 1989), impedes dispersal, particularly in males that appear to be unable to gain breeding positions in large groups with established competitors.

Natal dispersal in this system represents a dramatic shift from an indirect mode of contributing to kin production (effective helping) to the direct approach of breeding. In large groups, per capita probability of dispersal declines, as dispersers from large groups are under-represented presumably because of sibling competition for breeding vacancies. Auxiliary dispersal, however, in which the destination group is either a new venue for helping a codispersing relative, or a stepping-stone to a future breeding position, is relatively rare (15% of movements) (see also Koenig & Mumme 1987; Ligon & Ligon 1990; Mulder 1995; Koenig, Haydock & Stanback 1998). Auxiliary dispersers tend to be very young females, tend to move only to the immediate neighbourhood, and tend to disperse again before breeding. Breeding dispersal in this system is generally a reaction to misfortune − either the death of a mate or breeding failure (see also Russell & Rowley 1993; Mulder 1995; Daniels & Walters 2000a). None the less, this form of dispersal is also strongly female-biased and short-distance. Departure of breeders from large groups, mainly by females following a mate's death, can often be interpreted as inbreeding avoidance because it is known in many cases that most remaining males are sons (Woolfenden & Fitzpatrick 1984, 1986; Walters, Doerr & Carter 1988, 1992; Rabenold 1990; Daniels & Walters 2000a; Hatchwell et al. 2000).

divergence of male and female dispersal strategies

Female birds generally are more likely to disperse from natal areas than males, and often disperse at earlier ages or farther than males, particularly in social species. A variety of explanations have been put forward, especially in contrast to generally male-biased dispersal in mammals (Greenwood 1980; Clarke et al. 1997; Wolff & Plissner 1998). Undoubtedly, current patterns are produced in part by history and phylogeny (Perrin & Mazalov 1999), but ecological explanations should still be sought for the maintenance of dispersal tendencies (Lewis 1982; Waser & Jones 1983; Shields 1987; Stacey & Ligon 1987, 1991; Marzluff & Balda 1989; Ligon, Ligon & Ford 1991; Koenig et al. 1992, 1998; Mumme 1992; Walters et al. 1992; Komdeur et al. 1995; Stith et al. 1996; Clarke et al. 1997; Wolff & Plissner 1998; Daniels & Walters 2000b; Hatchwell & Komdeur 2000). Natal dispersal produces a scattering of kin that can have the effects of preventing competition with siblings or parents, preventing inbreeding, and promoting the lineage throughout the population (Hamilton & May 1977; Emlen 1982, 1995; Waser 1985, 1988; Woolfenden & Fitzpatrick 1986; Shields 1987; Keane 1990; Pusey & Wolf 1996; Dieckmann et al. 1999). Avoidance of inbreeding would be further favoured by sex-biased dispersal, but this could be accomplished by a greater propensity for either sex to disperse. Philopatry can be favoured by benefits of defending spatially defined resources. It has been argued that the general tendency for male birds to aggressively defend resources circumscribed by territorial boundaries, coupled with their ability to effectively contribute to rearing young, establishes males as the philopatric sex and favours female dispersal, except in ‘female-defence’ systems that tend to be polygynous (Greenwood 1980; Shields 1987; Clarke et al. 1997; Wolff & Plissner 1998).

In social systems such as that of stripe-backed wrens, dispersal is targeted to existing social groups, so that dispersers must penetrate an established social order and subsequently dominate breeding efforts among individuals of the same sex. Social resistance to movement can contribute to philopatry beyond the effects of defense of resources or mates. Males are not simply the sex choosing the mating site (Wolff & Plissner 1998), but are limited in their dispersal options by the strength of the advantage of position in an age-related social queue for breeding status. Female breeding status is established by competition mainly among neighbourhood females for breeding positions in social groups. Age and local residency appear to be important, so that females are not simply searching for and defending mates.

The strong tendency for male natal dispersers to be limited to destinations with few or no resident males (especially pairs or trios), and their resulting poor reproductive success following dispersal, suggests that philopatry is enforced by strong ties to defended resources and/or hierarchies among males (Wiley & Rabenold 1984). The fact that males infrequently attempt breeding dispersal (22% of events) or auxiliary dispersal (25%), compared to females, also suggests that males are relatively unlikely to abandon a defended area and the established social position that goes with it. Patient philopatry by males generally leads to greater reproductive success than dispersal, notwithstanding a long delay, as males can inherit the breeding position in a large natal group once they reach the oldest age class. Philopatry produces the build-up of local male lineages, which reinforce further compensating benefits of helping during a reproductive delay. Results of this study support the conclusion of previous analyses of this system that, even during this delay, the average effect of helpers on the reproductive success of breeders that are often parents is very similar to the reproductive success that can be expected by a male disperser. We show here that dispersing males can expect 0·37 offspring per year compared to our previous estimate of 0·44 offspring equivalents for philopatric male helpers, reducing incentives to leave (Rabenold et al. 1990).

Female stripe-backed wrens, in contrast, do not often inherit breeding positions in their natal group. They appear to be better able to establish breeding dominance in a new large group, perhaps because the focus of breeding activity is so concentrated in a single nest, and they have inherently better access than males to information about the parentage of young in the nest. Female dispersers produce more young during their tenure as breeders than males, and this ability improves with age, since they appear to avoid unaided matings, and are often ‘trading up’ to better breeding opportunities (see also Wiklund 1996; Daniels & Walters 2000a). Female natal dispersers also move further and at a younger age than males, without being evicted as in some species (splendid fairy wren: Mulder 1995), further reinforcing the view that females are less constrained than males. Joint dispersal by coalitions of individuals occurs mainly among females, compared to the male bias in some social carnivores (Pusey & Packer 1987). Field experiments with stripe-backed wrens (Zack & Rabenold 1989) and with bicolored wrens (Camphylorhynchus griseus) (Haydock and Rabenold, unpublished data), in which breeding females were removed from groups of different sizes, show that older females have a clear advantage in competition for breeding positions, and that the collective response of nonbreeding females is much greater for positions in larger groups.

distance of dispersal

Natal dispersers of either sex seldom move further than two or three territory diameters, and the same is true of breeding dispersers. This is much shorter than the norm reported for non-cooperative resident species of approximately 10 territory diameters (Greenwood & Harvey 1982; Payne 1991; Verhulst, Perrins & Riddington 1997; Paradis et al. 1998; Wheelwright & Mauck 1998; Wolff & Plissner 1998), but comparable to other social species (Koenig et al. 1992; Russell & Rowley 1993; Stith et al. 1996; Daniels & Walters 2000b). This viscosity of movement is greater than expected if dispersal destinations were randomly drawn from the whole population, or from the pool of groups in which breeder deaths had recently created opportunities. This conclusion is unaffected by whether dispersal distance is measured as physical displacement or as social spheres crossed. Members of the dispersive sex, at the average ages of dispersal, have lower survival, suggesting some substantial cost of dispersal, but we cannot measure dispersal risk precisely because the exact timing and cause of mortality are rarely known. These data could also confound a sex-specific survival benefit of philopatry, but our observations of the equality of participation in most group activities (Rabenold 1985) do not support this. For 30-g birds whose normal locomotion is largely hopping and fluttering from branch to branch, in a patchy savanna habitat teeming with aerial predators, simply crossing between groves of trees appears to be inherently dangerous. It is also possible that the detection distance for breeder vacancies is short, so that individuals using the natal group for a base of operations can mainly detect openings near the home group.

Despite potential risks of dispersal (see also Russell 2000), removal experiments for both stripe-backed and bicolored wrens showed that some individuals respond to breeding vacancies from substantial distances. This suggests scouting well beyond the limits of immediately neighbouring groups, although such distant respondents to openings rarely succeed in competition. In 20 such experiments with stripe-backed wrens, an average of five females responded to large-group openings, and birds from adjacent groups won 70% of contests while comprising less than 40% of contestants (Zack & Rabenold 1989; Haydock and Rabenold, unpublished data). In the present data set, successful dispersal is not strictly limited to the nearest contiguous territories, although the modal destination in all dispersal categories is an adjacent group, 60% of all dispersers travel less than 300 m (two average territory diameters), and 45% remain within a map radius of two territories. The neighbourhood of potential dispersal is thereby effectively limited to a two- to three-territory radius encompassing fewer than 10 territories in a dense population. On average, three breeding females are likely to die annually within such a neighbourhood, with 15 non-breeding females to replace them. Because dispersal is limited almost exclusively to existing groups and breeder survival is relatively high, there is a permanent surplus of potential dispersers, and breeding opportunities are strongly contested (Hannon et al. 1985; Woolfenden & Fitzpatrick 1986; Russell & Rowley 1993; Koenig et al. 1998). Compounding this, potential dispersers respond less to small-group vacancies and breeders sometimes abandon small groups. This selectivity decreases the perceived availability of breeding opportunities and reinforces the importance of competitive ability. Familiarity with either the target territory, or with the prospective mate, as well as experience with competitive interactions, appears to confer an advantage and thereby favours both delayed and neighbourhood dispersal (Brown 1987; Zack & Rabenold 1989; Rabenold 1990; Zack 1990; Walters et al. 1992), as predicted by the ‘experienced disperser’ hypothesis presented earlier.

sex-biased dispersal and diversity of reproductive opportunities

Stripe-backed wrens represent the most common, relatively simple form of cooperative breeding, in which (a) groups generally smaller than 10 individuals occupy all-purpose territories on which a single mated pair breeds (with occasional polyandry), (b) both close inbreeding and cross-group breeding are avoided and (c) dispersal of all types is clearly female-biased (Rabenold 1990; Rabenold et al. 1991a; Piper & Slater 1993). Among cooperatively breeding birds, however, dispersal systems range from strong female bias to strong male bias (Brown 1974, 1987; Emlen 1982, 1995; Woolfenden & Fitzpatrick 1986, 1990; Stacey & Ligon 1987, 1991; Marzluff & Balda 1989; Brown & Brown 1990; Curry & Grant 1990; Koenig & Stacey 1990; Ligon & Ligon 1990; Koenig et al. 1992, 1998; Russell & Rowley 1993; Mulder 1995; Langen 1996a,b; Daniels & Walters 2000b; Koenig et al. 2000; Williams 2000). Dispersal that is strongly delayed, female-biased and limited to nearby groups is associated in this system with more fitness alternatives than in nonsocial species, but relatively few routes to breeding status in either sex compared to more complex societies.

In systems with larger groups and more complex patterns of parentage, dispersal systems depart from strong female bias. A greater diversity of opportunities for males, including prebreeding or cobreeding positions in non-natal groups, is associated with increased male dispersal (Brown & Brown 1990; Craig & Jamieson 1990; Langen 1996a,b; Williams 2000). Greater opportunities for females to breed in their natal groups, often in the form of plural or joint nesting, also occur in these systems, associated with the ‘reversal’ of the sex-bias in dispersal. This pattern appears to be especially well expressed in the Corvidae, although female bias persists in the simplest corvid social systems, like that of the Florida scrub-jay (Woolfenden & Fitzpatrick 1986, 1990). The clearest contrast with stripe-backed wrens is provided by Costa Rican brown jays (Williams 2000). In these jays, females appear constrained to breed in a stable age-related hierarchy within a lineage occupying a traditional territory, with some potential for plural breeding in the natal group, but little breeding potential for dispersers. Dispersive males can enter large groups either as breeders or non-breeders, sometimes sharing access to breeding females, and they can breed across territory boundaries, in sharp contrast to the limited options for male stripe-backed wrens. The brown jay system tends more to polyandry within groups, rather than the polygyny associated with male-biased dispersal in groups like ducks emphasized previously as the main exceptions to female bias in avian dispersal.

Sex-biased dispersal in social species appears to reflect the comparative diversity of opportunities for breeding in natal and non-natal groups for each sex. This perspective extends Greenwood's (1980) pioneering emphasis on mating systems by recognizing a gradient of effects among social species from philopatric males defending both territories and status while females pursue a spatially broader competitive strategy, to a ‘reversed’ situation in which females form more philopatric lineages, within which plural breeding is possible, while males pursue a broader search for breeding opportunities that can be shared in new groups. The philopatric sex need not be defending only physical resources directly, but also position in a stable hierarchy of known and related individuals. The dispersive sex could be employing a mixed strategy of simultaneously pursuing the benefits of investment in status within a group and opportunistic mating. This approach better covers the range of mating systems and sex-biased dispersal, given the more extensive empirical base that now exists for social species. We propose a more inclusive ‘breeding diversity’ hypothesis, predicting that the more philopatric sex will have reproductive options most constrained by the effects of local resource defense or social status, and the fewest options for breeding in non-natal groups, while a more prospective strategy by the other sex will be associated with diverse breeding opportunities beyond the natal area. If these two classes of variables were to be fairly balanced for the two sexes, sex-biased dispersal in either direction could occur, assuming that differential dispersal is driven by a strong underlying benefit of avoidance of competition and/or inbreeding (Perrin & Mazalov 1999).

implications for population dynamics

Understanding the balance of impetus and resistance for individual movement will improve our understanding of the resilience and viability of populations, which are likely to be strongly determined by exchange of individuals. Stripe-backed wrens can serve as a useful model for the functioning of dispersal systems, particularly in the tropics where sociality is common. Sociality can curtail gene flow and demographic interchange within and between populations. This is especially important in naturally patchy, and increasingly anthropogenically fragmented habitats like the threatened savanna/dry-forest ecosystem. Stable social groups, with generally only two breeding positions per group and high breeder survival, limit opportunities. In addition, strong competition limits access for younger individuals, those from large groups, and those based more than a few hundred meters from an opening. Resistance to immigrants by either within-group males or neighbourhood females establishes a local density-dependence for dispersal, which has been shown to be important for both single- and meta-population dynamics (Saether, Engen & Lande 1999).

The sex bias in dispersal, along with the delay and viscosity, could strongly limit effective response of immigrants to breeder mortality, gaps in populations, or low-density populations, and thereby depress the resilience of populations. When groups of stripe-backed wrens decline and disappear because of chance negative balance between recruitment and loss, the gaps that they leave, even in proven prime habitat, are rarely colonized by new groups. Neighbouring groups often shift their ranges, and sometimes split, resulting in coverage of the gap, but we observed newly formed associations of immigrants successfully colonizing vacancies only six times in 1800 group-years of study. Constraints suggested by patterns of interpopulation movement are similar to those we show here within populations, but in exaggerated form, and we analyse them elsewhere (Yaber and Rabenold, unpublished data). We emphasize the importance of social resources in addition to purely ecological ones, particularly the importance of inheritance of breeding status in stable same-sex hierarchies. Social resistance to movement by one or both sexes establishes a brake on gene flow and potential demographic ‘rescue’ (Pulliam & Danielson 1991) that has important implications for population viability, particularly in tropical habitats.

Acknowledgements

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

We are indebted to Don Tomás Blohm, owner of Hato Masaguaral, whose kindness, generosity and dedication made possible this and many other studies in the Venezuelan llanos. The foundation of field data was built up with the collaboration Joseph Haydock, Steve Zack, Walter Piper, Gary Slater, Carla Christensen, Tom Langsheid, Angela Garcia, Evelyne Laurent, Yvette Halpin, Anna Ciecka, Jen Sadowsky, Jeremy Hale, Jenny Drnevich, Chris Burke, Todd Schneeberger, Becky Smith, Nicole Strong, Jennifer Nead, Jeff Bigham, Marc Johnson, Jon Winters, Josiah Clark, Carolina Bertsch and Patricia Parker. Special thanks to Peter Waser, Jeff Lucas, John Dunning, Dean Williams, Steve Zack, Ken Norris, Jeff Albarella and an anonymous reviewer for reading previous versions of this report and offering important suggestions. This study was made possible by funding from Miguel Yaber, the National Science Foundation, Purdue Research Foundation, American Museum of Natural History (Frank M. Chapman fund), Sigma Delta Epsilon (Graduate Women in Science), Sigma Xi, Alton A Lindsey Graduate Fellowship in Ecology, Hampden-Sydney College and Purdue University

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Baker, A.M., Mather, P.B. & Hughes, J.M. (2001) Evidence for long-distance dispersal in a sedentary passerine, Gymnorhina tibicen (Artamidae). Biological Journal of the Linnean Society, 72, 333342.
  • Beier, P. & Noss, R.F. (1998) Do habitat corridors provide connectivity? Conservation Biology, 12, 12411252.
  • Brawn, J.D. & Robinson, S.K. (1996) Source–sink population dynamics may complicate the interpretation of long-term census data. Ecology, 77, 312.
  • Breininger, D.R. (1999) Florida scrub-jay demography and dispersal in a fragmented landscape. Auk, 116, 520527.
  • Brown, J.L. (1974) Alternate routes to sociality in jays − with a theory for the evolution of altruism and communal breeding. American Zoologist, 14, 6380.
  • Brown, J.L. (1987) Helping and communal breeding in birds. Princeton University Press, Princeton.
  • Brown, J.L. & Brown, E.R. (1990) Mexican jays: uncooperative breeding. Cooperative breeding in birds. Long term studies of ecology and behavior (eds P.B.Stacey & W.D.Koenig), pp. 267288. Cambridge University Press, New York.
  • Clarke, A.L., Saether, B.-E. & Roskaft, E. (1997) Sex biases in avian dispersal: a reappraisal. Oikos, 79, 429438.
  • Clutton-Brock, T.H. (1989) Female transfer and inbreeding avoidance in social animals. Nature, 337, 7072.
  • Cockburn, A. (1998) Evolution of helping behavior in cooperatively breeding birds. Annual Review of Ecology and Systematics, 29, 141177.
  • Craig, J.L. & Jamieson, I.G. (1990) Pukeko: different approaches and some different answers. Cooperative breeding in birds. Long term studies of ecology and behavior (eds P.B.Stacey & W.D.Koenig), pp. 385412. Cambridge University Press, New York.
  • Curry, R.L. & Grant, P.R. (1990) Galápagos mockingbirds: territorial cooperative breeding in a climatically variable environment. Cooperative breeding in birds. Long term studies of ecology and behavior (eds P.B.Stacey & W.D.Koenig), pp. 289332. Cambridge University Press, New York.
  • Daniels, S.J. & Walters, J.R. (2000a) Dispersal of breeding female red-cockaded woodpeckers. Ecology, 81, 24732484.
  • Daniels, S.J. & Walters, J.R. (2000b) Inbreeding depression and natal dispersal in red-cockaded woodpeckers. Condor, 102, 482491.
  • Dieckmann, U., O'Hara, B. & Weisser, W. (1999) The evolutionary ecology of dispersal. Trends in Ecology and Evolution, 14, 8890.
  • Emlen, S.T. (1982) The evolution of helping. I. An ecological constraints model. American Naturalist, 119, 2939.
  • Emlen, S.T. (1994) Benefits, constraints and the evolution of the family. Trends in Ecology and Evolution, 9, 282285.
  • Emlen, S.T. (1995) An evolutionary theory of the family. Proceedings of the National Academy of Science, 92, 80928099.
  • Emlen, S.T. & Wrege, P.H. (1991) Breeding biology of white-fronted bee-eaters at Nakuru (Kenya): the influence of helpers on breeder fitness. Journal of Animal Ecology, 60, 309326.
  • Girman, D.J., Vila, C., Geffen, E., Creel, S., Mills, M.G.L., McNutt, J.W., Ginsberg, J., Kat, P.W., Mimiya, K.H. & Wayne, R.K. (2001) Patterns of population subdivision, gene flow, and genetic variability in the African wild dog (Lycaon pictus). Molecular Ecology, 10, 17031723.
  • Greenwood, P.J. (1980) Mating systems, philopatry and dispersal in birds and mammals. Animal Behaviour, 28, 11401162.
  • Greenwood, P.J. & Harvey, P.H. (1982) The natal and breeding dispersal of birds. Annual Review of Ecology and Systematics, 13, 121.
  • Hamilton, W.D. & May, R.M. (1977) Dispersal in stable habitats. Nature, 269, 578581.
  • Hannon, S.J., Mumme, R.L., Koenig, W.D. & Pitelka, F.A. (1985) Replacement of breeders and within-group conflict in the cooperatively breeding acorn woodpecker. Behavioral Ecology and Sociobiology, 17, 303312.
  • Hanski, I. (1997) Metapopulation dynamics: from concepts and observations to predictive models. Metapopulation biology: ecology, genetics, and evolution (eds I.Hanski & M.E.Gilpin), pp. 6992. Academic Press, London.
  • Hanski, I. & Gilpin, M. (1991) Metapopulation dynamics: brief history and conceptual domain. Biological Journal of the Linnean Society, 42, 316.
  • Hatchwell, B.J. & Komdeur, J. (2000) Ecological constraints, life history traits and the evolution of cooperative breeding. Animal Behaviour, 59, 10791086.
  • Hatchwell, B.J., Russell, A.F., Ross, D.J. & Fowlie, M.K. (2000) Divorce in cooperatively breeding long-tailed tits: a consequence of inbreeding avoidance? Proceedings of the Royal Society of London, 267, 813819.
  • Haydock, J., Parker, P.G. & Rabenold, K.N. (1996) Extra-pair paternity uncommon in the cooperatively breeding bicolored wren. Behavioral Ecology and Sociobiology, 38, 116.
  • Jones, W.T. (1986) Survivorship in philopatric and dispersing kangaroo rats (Dipodomys spectabilis). Ecology, 67, 202207.
  • Kahn, N.W., St John, J. & Quinn, T.W. (1998) Chromosome-specific intron size differences in the avian CHD gene provide an efficient method for sex identification in birds. Auk, 115, 10741078.
  • Keane, B. (1990) Dispersal and inbreeding avoidance in the white-footed mouse, Peromyscus leucopus. Animal Behaviour, 40, 143152.
  • Koenig, W.D., Haydock, J. & Stanback, M.T. (1998) Reproductive roles in the cooperatively breeding acorn woodpecker: incest avoidance versus reproductive competition. American Naturalist, 151, 243255.
  • Koenig, W.D., Hooge, P.N., Stanback, M.T. & Haydock, J. (2000) Natal dispersal in the cooperatively breeding acorn woodpecker. Condor, 102, 492502.
  • Koenig, W.D. & Mumme, R.L. (1987) Population ecology of the cooperatively breeding acorn woodpecker. Princeton University Press, Princeton.
  • 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.
  • Koenig, W.D. & Stacey, P.B. (1990) Acorn woodpeckers: group-living and food storage under contrasting ecological conditions. Cooperative breeding in birds. Long term studies of ecology and behavior (eds P.B.Stacey & W.D.Koenig), pp. 413454. Cambridge University Press, New York.
  • Koenig, W.D., Van Vuren, D. & Hooge, P.N. (1996) Detectability, philopatry, and the distribution of dispersal distances in vertebrates. Trends in Ecology and Evolution, 11, 514517.
  • Komdeur, J., Huffstadt, A., Prast, W., Castle, G., Mileto, R. & Wattel, J. (1995) Transfer experiments of Seychelles warblers to new islands: changes in dispersal and helping behaviour. Animal Behaviour, 49, 695708.
  • Langen, T.A. (1996a) The mating system of the white-throated magpie-jay Calocitta formosa and Greenwood's hypothesis for sex-biased dispersal. Ibis, 138, 506513.
  • Langen, T.A. (1996b) Skill acquisition and the timing of natal dispersal in the white-throated magpie-jay, Calocitta formosa. Animal Behaviour, 51, 575588.
  • Lewis, R.A. (1982) Dispersal in a population of white-browned sparrow weavers. Condor, 84, 306312.
  • Ligon, J.D. & Ligon, S.H. (1990) Green woodhoopoes: life history traits and sociality. Cooperative breeding in birds. Long term studies of ecology and behavior (eds P.B.Stacey & W.D.Koenig), pp. 3166. Cambridge University Press, New York.
  • Ligon, J.D., Ligon, S.H. & Ford, H.A. (1991) An experimental study of the basis of male philopatry in the cooperatively breeding superb fairy-wren Malurus cyaneus. Ethology, 87, 134148.
  • Loison, A., Jullien, J.M. & Menaut, P. (1999) Subpopulation structure and dispersal in two populations of chamois. Journal of Mammalogy, 80, 620632.
  • Marzluff, J.M. & Balda, R. (1989) Causes and consequences of female-biased dispersal in a flock-living bird. Ecology, 70, 316328.
  • Mulder, R.A. (1995) Natal and breeding dispersal in a cooperative, extra-group-mating bird. Journal of Avian Biology, 26, 234240.
  • Mumme, R.L. (1992) Delayed dispersal and cooperative breeding in the Seychelles warbler. Trends in Ecology and Evolution, 7, 330331.
  • Paradis, E., Baillie, S.R., Sutherland, W.J. & Gregory, R.D. (1998) Patterns of natal and breeding dispersal in birds. Journal of Animal Ecology, 67, 518536.
  • Payne, R.B. (1991) Natal dispersal and population structure in a migratory songbird, the indigo bunting. Evolution, 45, 4962.
  • Pen, I. & Weissing, F.J. (2000) Towards a unified theory of cooperative breeding: the role of ecology and life history re-examined. Proceedings of the Royal Society of London, 267, 24112418.
  • Perrin, N. & Mazalov, V. (1999) Dispersal and inbreeding avoidance. American Naturalist, 154, 282292.
  • Piper, W.H. & Slater, G. (1993) Polyandry and incest avoidance in the cooperative stripe-backed wren of Venezuela. Behaviour, 124, 227247.
  • Pulliam, H.R. & Danielson, B.J. (1991) Sources, sinks, and habitat selection: a landscape perspective on population dynamics. American Naturalists, 137, s50s66.
  • Pusey, A.E. & Packer (1987) Dispersal and philopatry. Primate societies (eds B.B.Smuts, D.L.Cheney, R.M.Seyfarth, R.W.Wrangham & T.T.Struhsaker), pp. 250266. University of Chicago Press, Chicago.
  • Pusey, A.E. & Wolf, M. (1996) Inbreeding avoidance in animals. Trends in Ecology and Evolution, 11, 201206.
  • Rabenold, K.N. (1984) Cooperative enhancement of reproductive success in tropical wren societies. Ecology, 65, 871885.
  • Rabenold, K.N. (1985) Cooperation in breeding by nonreproductive wrens: kinship, reciprocity, and demography. Behavioral Ecology and Sociobiology, 17, 117.
  • Rabenold, K.N. (1990) Camplorhynchus wrens: the ecology of delayed dispersal and cooperation in the Venezuelan savanna. Cooperative breeding in birds. Long term studies of ecology and behavior (eds P.B.Stacey & W.D.Koenig), pp. 157196. Cambridge University Press, New York.
  • Rabenold, P.P., Rabenold, K.N., Piper, W.H., Decker, M.D. & Minchella, D.J. (1991a) Density-dependent dispersal in social wrens: genetic analysis using novel matriline markers. Animal Behaviour, 42, 144146.
  • Rabenold, P.P., Rabenold, K.N., Piper, W.H., Decker M.D. & Haydock, J. (1991b) Using DNA fingerprinting to assess kinship and genetic structure in avian populations. Proceedings of the Fourth International Congress of Systematic and Evolutionary Biology (ed. E.C.Dudley), pp. 611620. Dioscorides Press, Portland.
  • Rabenold, P.R., Rabenold, K.R., Piper, W.H., Haydock, J. & Zack, S.W. (1990) Shared paternity revealed by genetic analysis in cooperatively breeding tropical wrens. Nature, 348, 538540.
  • Russell, A.F. (2000) Dispersal costs set the scene for helping in an atypical avian cooperative breeder. Proceedings of the Royal Society of London, 268, 9599.
  • Russell, E.M. & Rowley, I.C.R. (1993) Philopatry or dispersal: competition for territory vacancies in the splendid fairy-wren, Malurus splendens. Animal Behaviour, 45, 519539.
  • Saether, B.E., Engen, S. & Lande, R. (1999) Finite metapopulation models with density-dependent migration and stochastic local dynamics. Procceedings of the Royal Society of London, 266, 113118.
  • Shields, W.M. (1987) Dispersal and mating systems: investigating their casual connections. Mammalian dispersal patterns (eds B.D.Chpeko-Sade & Z.T.Halpin), pp. 324. University of Chicago Press, Chicago.
  • Smith, J.N.M., Taitt, M.J., Rogers, C.M., Arcese, P., Keller, L.F., Cassidy, A.L.E.V. & Hochachka, W.M. (1996) A metapopulation approach to the population biology of the song sparrow Melospiza melodia. Ibis, 138, 120128.
  • Stacey, P.B. & Koenig, W.D., eds (1990) Cooperative breeding in birds: long-term studies of ecology and behavior. Cambridge University Press, New York.
  • Stacey, P.B. & Ligon, J.D. (1987) Territory quality and dispersal options in the acorn woodpecker, and a challenge to the habitat saturation model of cooperative breeding. American Naturalist, 130, 654676.
  • Stacey, P.B. & Ligon, J.D. (1991) The benefits-of-philopatry hypothesis for the evolution of cooperative breeding: variation in territory quality and group size effects. American Naturalist, 137, 831846.
  • Stith, B.M., Fitzpatrick, J.W., Woolfenden G.E. & Pranty, B. (1996) Classification and conservation of metapopulations: a case study of the florida scrub jays. Metapopulation and Wildlife Conservation (ed. D.R.McCullough), pp. 187215. Washington, D.C.
  • Travis, J.M.J. & Dytham, C. (1999) Habitat persistence, habitat availability and the evolution of dispersal. Proceedings of the Royal Society of London Series B − Biological Sciences, 266, 723728.
  • Verhulst, S., Perrins, C.M. & Riddington, R. (1997) Natal dispersal of great tits in a patchy environment. Ecology, 78, 864872.
  • Walters, J.R. (1990) Red-cockaded woodpeckers: a ‘primitive’ cooperative breeder. Cooperative breeding in birds. Long term studies of ecology and behavior (eds P.B.Stacey & W.D.Koenig), pp. 67102. Cambridge University Press, New York.
  • Walters, J.R.P.D., Doerr & Carter, III (1988) The cooperative breeding system of the red-cockaded woodpecker. Ethology, 78, 275305.
  • Walters, J.R., Doerr P.D. & Carter, J.H. III (1992) Delayed dispersal and reproduction as a life-history tactic in cooperative breeders: fitness calculations from red-cockaded woodpeckers. American Naturalist, 139, 623643.
  • Waser, P.M. (1985) Does competition drive dispersal? Ecology, 66, 11701175.
  • Waser, P.M. (1988) Resources, philopatry, and social interactions among mammals. The ecology of social behavior, pp. 109130. Academic Press, New York.
  • Waser, P.M. & Jones, W.T. (1983) Natal philopatry among solitary mammals. Quarterly Review of Biology, 58, 355390.
  • Wheelwright, N.T. & Mauck, R.A. (1998) Philopatry, natal dispersal, and inbreeding avoidance in an island population of savanna sparrows. Ecology, 79, 755767.
  • Wiklund, C.G. (1996) Determinants of dispersal in breeding merlins (Falco columbarius). Ecology, 77, 19201927.
  • Wiley, R.H. & Rabenold, K.N. (1984) The evolution of cooperative breeding by delayed reciprocity and queuing for favorable social positions. Evolution, 38, 609621.
  • Williams, D. (2000) Cooperative breeding in the Costa Rican brown jay. PhD Dissertation, Purdue University.
  • Wolff, J.O. & Plissner, J.H. (1998) Sex biases in avian natal dispersal – an extension of the mammalian model. Oikos, 83, 327330.
  • Woolfenden, G.E. & Fitzpatrick, J.W.&., eds. (1984) The Florida scrub jay. Demography of a cooperative-breeding bird. Princeton University Press, Princeton.
  • Woolfenden, G.E. & Fitzpatrick, eds. (1986) Sexual asymmetries in the life history of the Florida scrub jay. Ecological aspects of social evolution (eds D.I.Rubenstein & R.W.Wrangham), pp. 87107. Princeton University Press, Princeton.
  • Woolfenden, G.E. & Fitzpatrick, J.W., eds (1990) Florida scrub jays: a synopsis after 18 years of study. Cooperative breeding in birds. Long term studies of ecology and behavior (eds P.B.Stacey & W.D.Koenig), pp. 239266. Cambridge University Press, New York.
  • Zack, S. (1990) Coupling delayed breeding with short-distance dispersal: patience and assessment in cooperatively breeding birds. Ethology, 86, 265286.
  • Zack, S. & Rabenold, K.N. (1989) Assessment, age and proximity in dispersal contests among cooperative wrens: field experiments. Animal Behaviour, 38, 235247.