- Email: [email protected]
Floater males gain reproductive success through extrapair fertilizations in the stitchbird
ANIMAL BEHAVIOUR, 1999, 58, 321–328 Article No. anbe.1999.1150, available online at http://www.idealibrary.com on
Floater males gain reproductive success through extrapair fertilizations in the stitchbird JOHN G. EWEN, D. P. ARMSTRONG & DAVID M. LAMBERT
Ecology Group, Institute of Natural Resources, Massey University (Received 21 July 1998; initial acceptance 4 November 1998; final acceptance 8 April 1999; MS. number: 5941)
We used minisatellite DNA profiling to assign parentage to stitchbird, Notiomystis cincta, chicks from a breeding population on Tiritiri Matangi Island off the coast of the North Island of New Zealand. The small population size allowed samples to be collected from all potential parents and nearly (33/34 nestlings) complete assignment of paternity. Analysis revealed that 35% of nestlings (12/34) were the result of extrapair copulation and that extrapair young were present in 80% of nests (8/10). About half of the extrapair nestlings were the offspring of unpaired males. This is substantially higher than predicted from the literature, which suggests that extrapair paternity is typically gained by paired males.
a gel. It is relatively easy to sample both putative parents at a nest and test for genetic relationships with any nestlings (e.g. great tit: Verboven & Mateman 1997; reed bunting, Emberiza schoeniclus: Dixon et al. 1994). Owing to the nature of running gels, and the generally high mobility of the animals being studied, however, it quickly becomes impractical to attempt to assign maternity and paternity to those cases where extrapair parentage has been revealed. Some studies have attempted to address this question by comparing extrapair young with neighbouring adults (e.g. tree swallows, Tachycineta bicolor: Lifjeld et al. 1993; indigo bunting: Westneat 1990). Such studies have revealed that extrapair fertilization is often by other paired males, usually resident on neighbouring territories. As far as we know, only two studies have sampled all possible parents and been able to assign parentage to all nestlings (Gibbs et al. 1990; Hasselquist et al. 1995). Both these studies are consistent with the literature suggesting EPCs were by other paired males. In this study we also sampled all possible parents and aimed to assign paternity to all nestlings. We had two aims. Our first aim was to document the presence of extrapair parentage in a small, isolated and reintroduced population of stitchbirds, Notiomystis cincta. Previous research on stitchbird breeding has reported: (1) the presence of EPCs (Castro et al. 1996); (2) the observation of multimale chases of females during the breeding season (Angehr 1984; Lovegrove 1985); and (3) a reproductive anatomy indicative of a species with intense competition for reproductive success (Castro et al. 1996). Additionally, stitchbirds copulate in a face-to-face position possibly unique among birds (Castro et al. 1996; Ewen 1998). However, no previous study has investigated
Many species whose mating systems have been categorized according to social associations have recently been re-evaluated since genetic analysis of parentage (Burke & Bruford 1987; Wetton et al. 1987) has revealed either extrapair fertilizations (e.g. house sparrow, Passer domesticus: Wetton et al. 1987; great reed warbler, Acrocephalus arundinaceus: Hasselquist et al. 1995) or intraspecific brood parasitism (e.g. zebra finch, Taeniopygia guttata: Birkhead et al. 1990; purple martin, Progne subis: Morton et al. 1990). Results from such studies indicate that there is greater interspecific variability in the extent to which extrapair parentage occurs than previously appreciated (Petrie & Kempenaers 1998), even where extrapair copulations (EPCs) have been observed (e.g. indigo buntings, Passerina cyanea: Westneat 1987, 1990). Research to date indicates that EPCs are part of mixed reproductive behaviours, with paired males seeking copulations with other already paired females (Westneat et al. 1990). Females may also actively seek copulations. Use of DNA profiling allows direct quantification of reproductive success resulting from different behaviours of both males (e.g. red-winged blackbirds, Agelaius phoeniceus: Gibbs et al. 1990) and females (e.g. great reed warbler: Hasselquist et al. 1996; blue tit, Parus caeruleus: Kempenaers et al. 1992). One limitation in the use of minisatellite DNA profiling is the number of individuals that can be run together on Correspondence and present address: J. Ewen, School of Zoology, Faculty of Science and Technology, La Trobe University, Melbourne, Australia (email: [email protected]). D. P. Armstrong and D. M. Lambert are at the Ecology Group, Institute of Natural Resources, Massey University, Private Bag 11-222, Palmerston North, New Zealand. 0003–3472/99/080321+08 $30.00/0
1999 The Association for the Study of Animal Behaviour
321
1999 The Association for the Study of Animal Behaviour
322
ANIMAL BEHAVIOUR, 58, 2
the presence of extrapair parentage in this species. Our second aim was to measure realized reproductive success among males in this population and determine whether males gaining extrapair fertilizations were mated males on neighbouring territories, nonterritorial males or a mixture of the two. METHODS
Behavioural Observation The stitchbird, or hihi, is a medium-sized, sexually dimorphic, forest-dwelling passerine. It is one of three endemic honeyeaters (Meliphagidae) found in New Zealand (Craig et al. 1981; Craig 1984; Rasch & Craig 1988). Males weigh about 40 g and females 28 g (Craig et al. 1982). We studied a population on Tiritiri Matangi Island from the time of their release in August 1995 through their first two breeding seasons of 1995–1996 and 1996– 1997. Tiritiri Matangi Island is an island situated 3 km off the Whangaparaoa Peninsula, 25 km north of Auckland, New Zealand. Stitchbirds were assumed to have previously occupied this island prior to 1890, as it lies well within the historical range of the species. Ten nests were monitored during the study with a total population size of 15 and 18 adult birds (for the 1995–1996 and 1996–1997 seasons, respectively). Stitchbirds breed from October to the end of February (Oliver 1955; Craig 1985). During this time, we checked nestboxes and made behavioural observations to locate breeding females to determine putative parentage. Owing to the lack of mature forest on the island, artificial nestboxes were provided in all large bush patches. We checked nestboxes daily until all females on the island had commenced nest building and again when each female neared the completion of nestling rearing in case she attempted a second clutch. During systematic searches of nestboxes any stitchbirds seen or heard were followed for as long as possible. As the birds are highly mobile, this was never for longer than 30 min. All birds in this population were identifiable by unique combinations of coloured plastic bands and a numbered aluminium band. During nest building and egg laying in stitchbirds in other populations, there are regular chases between males, between females and between males and females (Lovegrove 1985; Rasch 1985; Castro 1995). These chases are accompanied by increased calling by both sexes. In the present study, we used instances of such behaviour as indicators of the initiation of nest building, and followed birds until we found the nest site. Once nest sites were discovered, we observed the pairs daily during nest building and egg laying, and then every third day until the end of the nestling stage (Ewen 1998). Observation periods were between 50 and 120 min and involved a single observer moving quietly around a 30-m radius of the nestbox. These movements were largely restricted to tracks and boardwalks designed for the large numbers of people moving over the island (a sanctuary open to the public) on a daily basis. Female stitchbirds typically have sole responsibility for nest building and incubation (Oliver 1955; Craig 1985)
and they also do most of the provisioning of nestlings (Ewen 1998). Putative mothers were therefore assumed to be the female involved with such behaviours. More than one female has been observed nest building and incubating at one nest in the Kapiti Island population (Castro 1995). However, only a single female was ever observed building and incubating in each nest on Tiritiri Matangi Island. We observed 1–11 male stitchbirds near each nest site. These males were involved in chases with females and other males (Ewen 1998). Despite this, one male spent more time than the other males with each female and attempted to defend a ‘territory’ around the nest site. These males also helped feed the nestlings, albeit at a lower level than the females (Ewen 1998). Such birds were termed ‘paired’ males and were assigned as putative fathers at each nest. All other males observed near the nest (within 30 m) were recorded as putative extrapair males and ranked in relation to the number of times they were seen and the number of EPCs they attempted. This was necessary because there are limits to the number of individuals that can be compared directly on the same gel (see below). In the two breeding seasons studied, the population of stitchbirds was heavily biased towards males, 12:4 in 1995–1996 and 12:6 in 1996–1997. Despite this bias, birds attempted to breed as socially monogamous pairs, resulting in eight and six males, respectively, remaining unpaired in each breeding season.
Minisatellite DNA Profiling Methods Techniques used for collection of blood were as detailed in Ardern et al. (1994). Blood samples (56–76 ìl, mean 66 ìl) were obtained from all adults and surviving nestlings by wing venipuncture of the brachial vein, using a 13-mm, 27-gauge needle. The total amount of blood taken did not exceed 10–20% of the bird’s total blood volume. Blood sampling took ca. 2 min. Once eggs had hatched, we routinely checked nests every 2–3 daylight hours. This was continued until all surviving chicks had been banded and blood samples collected. Female stitchbirds typically remove dead chicks from the nest and discard them soon after discovery. Dead chicks found in the nest were collected for genetic analysis and post mortem examination. If the female had removed chicks, we searched an area of about 5 m in radius in an effort to recover them. Typically, white eggshells gave the location of nest refuse away and subsequently the location of any dead chicks. All chicks recovered had their wings and legs removed and placed in a 20C freezer for between 0.5 and 2 months. All broods with surviving young were bled at 21–25 days after hatching. Nestlings typically fledge 28–30 days after hatching. Therefore, sampling birds at this stage allowed for maximum chick growth with little chance of chicks fledging before sampling. We sampled 10 families for parentage analysis: the single successful clutch of four young in the 1995–1996 breeding season, and nine clutches comprising 36 young in the 1996–1997 breeding season.
EWEN ET AL.: EXTRAPAIR PATERNITY IN STITCHBIRDS
DNA extraction, precipitation and resuspension followed techniques detailed in Ardern & Lambert (1997). DNA from tissue samples was extracted similarly with the following additions. A piece of tissue measuring ca. 663 mm was removed from each dead chick. This was sliced into small pieces with a sterile scalpel blade and added to SET buffer. We added 30 ìl of 10% SDS and 30 ìl of proteinase K and incubated the sample at 55C for 3 h. A further 10 ìl of proteinase K were then added and the sample returned to 55C overnight. An extra Trisbuffered Phenol wash was also performed to remove additional proteins present in tissue. To 20 ìl of genomic DNA, we added 4 ìl 10Buffer (react 2), 2 ìl 2 mg/ml BSA, 1 ìl 160 nM Spermidine, 1 ìl of the restriction enzyme HaeIII and 11 ìl milli-Q water. This mixture was incubated overnight at 37C. Another 1 ìl of HaeIII was added the following day and the sample was incubated at 37C for a further 1 h. Digested samples were then stored at 20C. About 5 ìg of digested DNA were loaded in each lane of the gel. DNA fragments were resolved on a 0.8% agarose gel (1927 cm) in 1TBE running buffer at 55 V for 72 h. We then denatured the DNA by washing each gel for 15 min in 0.25 M HCl and then for 45 min in 0.5 M NaOH, 1.5 M NaCl. Gels were then neutralized by two 15-min washes in 1.5 M NaCl, 0.5 M Tris-HCl pH 7.2, 1 mM EDTA. Southern blot techniques were used to transfer DNA from agarose gels to nylon membranes (Amersham, Little Chalfont, Bucks, U.K.) in 6SSC. Membranes were then dried for 10 min at 37C before being baked at 80C for 2 h. Baked membranes were soaked in prehybridization mix (75 ml 0.5 M disodium hydrogen orthophosphate pH 7.2, 75 ml milli-Q water, 300 ìl 0.5 M EDTA pH 8.0, 10.5 g SDS) for 2 h at 65C. First, Jeffreys 33.15 probe (Jeffreys et al. 1985) was labelled with á-32PdCTP by random priming with Amersham radprime kit. Unincorporated label was removed using a G50 sephadex column. Hybridization of Jeffreys 33.15 to membranes was at 65C for a minimum of 18 h. Membranes were then washed twice with 5SSC, 0.1% SDS at 65C. DNA fragments hybridized to the 33.15 probe were exposed on X-ray film at either 80C with one intensifying screen or at room temperature for 1–6 days. After adequate exposure, membranes were stripped and reprobed with CA probe, which was similarly labelled with á-32PdCTP. Hybridization to membranes was at 55C for a minimum of 18 h. Membranes were washed twice with 5SSC, 0.1% SDS at 55C, and then exposed as detailed for Jeffreys 33.15 probe.
Data Analysis Samples from two broods were loaded on to each gel. All chicks from a brood were loaded with their putative parents on either side. Remaining lanes, situated in the centre of the gel, were loaded with all putative extrapair fathers for those families. This sequence of loading samples minimized distance for between-lane comparisons, a potential source of error in reading gels (Burke et al. 1991). All gels contained molecular weight markers in the outermost lanes (1 and 20). This was important to
determine even running within each gel and allow estimation of band sizes. In addition, gels included one common individual that acted as a genomic control to provide a means of standardization for hybridization conditions and the distance run between gels (Miller et al. 1994). We transposed band positions on to the acetate overlays by dotting band centres with a permanent marker pen (Galbraith et al. 1991). Bands were considered identical in two individuals if there was no more than a two-fold difference in intensity estimated by eye (Bruford et al. 1991; Hasselquist et al. 1995) and they fitted into bin sizes of no more than 1.5 mm (Lambert et al. 1994; Finch & Lambert 1996). We assigned parentage by identifying novel bands (Westneat 1990, 1993) and from the band-sharing coefficients of putative parents (Wetton et al. 1987; Bruford et al. 1991). All summary statistics are given as meansSD.
Ethical Note We did this research under approval from the Department of Conservation Banding Office and the Massey University Animal Ethics Committee. Permission was granted from the Auckland Conservancy of the Department of Conservation to conduct scientific research on Department of Conservation Estate and to conduct the above procedures with stitchbirds. Stitchbirds are listed as a globally threatened species and vulnerable by Birds to Watch 2 (Collar et al. 1994). From this study, we gained information on whether inbreeding was occurring, which would jeopardize the species’ future, and on how many birds were breeding in relation to the number released on Tiritiri Matangi Island. The results were given to the New Zealand Department of Conservation and used in the continued development of conservation initiatives for stitchbirds, in particular the preferred sex ratio and size of populations to be translocated. Birds were caught, banded and bled as part of a standard stitchbird translocation procedure. No obvious change in behaviour was observed after each bird’s release and no bird died as a direct result of such procedures. A previous study (Ardern et al. 1994), which specifically investigated the effects on behaviour and survival of blood sampling an endangered passerine, also found no mortality resulting from the use of these techniques. The experiment caused only temporary disturbances during observation times and nest checks. Most observations were made from boardwalks/tracks used by large numbers of public moving around the island. We therefore assumed an observer would not represent a novel stimulus. Nests were checked only after the female had been observed leaving the nestbox, to avoid disturbing her while she was brooding young. No nests were deserted during the experiment. RESULTS
Banding Profiles Both probes produced variable banding profiles for most individuals (Fig. 1). Five muscle tissue samples
323
324
ANIMAL BEHAVIOUR, 58, 2
Figure 1. DNA fingerprint of a brood of stitchbird nestlings (C1–C3), with their putative mother (PF) and putative father (PM) positioned on either side. Remaining lanes of the fingerprint are putative extrapair fathers (PEPMs). Arrows for chick 2 show bands not present in either putative parent but present in an extrapair male, which indicates an extrapair fertilization.
probed with 33.15 and six muscle tissue samples probed with CA were severely degraded and did not produce scorable profiles. These individuals were ignored for all subsequent analyses. Bands were scored in the approximate size range of 5–23 kb for probe 33.15 and 5–12 kb for probe CA. Samples extracted from whole blood produced banding profiles with more resolvable bands than those extracted from muscle tissue. An average of 21.143.19 and 12.293.4 bands were scored from blood and muscle tissue, respectively, for probe 33.15 and 19.394.15 and 15.25.07 from blood and tissue, respectively, for probe CA. Each probe provided a different set of fragments, although there was some degree of overlap: mean overlap 0.3480.098, N=11 for 33.15 versus CA; 0.2910.087, N=11 for CA versus 33.15.
Parentage Analysis Sixteen of the 34 scorable chicks were completely compatible with their social parents. All remaining chicks had at least one novel band present in their DNA profiles. DNA samples extracted from blood showed either just
one, or six or more novel bands and those derived from tissue showed either one, three or five novel bands. Assuming that the occurrence of a single novel fragment was the outcome of a mutation event, the mutation rate can be calculated. Given that 1192 clearly resolved bands were scored for all chicks and a single novel band was recorded on nine occasions, the mutation rate can be calculated as 9/1192=7.5510 3. This observed mutation rate lies within the order of 10 3 per band per meiotic event calculated for numerous other species (Jeffreys et al. 1985; Burke & Bruford 1987; Burke et al. 1989; Westneat et al. 1990; Ardern et al. 1997; Verboven & Mateman 1997). This suggests that the single novel bands are the result of mutation and not misassigned parentage. Twenty-two of the nestlings either had zero or one novel band, and were therefore considered the genetic offspring of their social parents. The remaining 12 (10 samples derived from blood and two from tissue) must be the result of either extrapair fertilizations or intraspecific brood parasitism. The criterion for assigning parentage based on the presence or absence of novel fragments was supported by comparison of band-sharing coefficients (Fig. 2). Although not completely distinct, those nestlings with no or few novel bands also had high band-sharing values compared with their putative father. Similarly, all nestlings with large numbers of novel bands had lower band-sharing values. Compared with males assigned as genetic fathers through investigation of novel bands, nestlings from which we used blood samples had a mean band-sharing coefficient of 0.710.08 (N=27) for probe 33.15 and 0.700.09 (N=27) for probe CA. This value is similar to that calculated for the mean maternal band sharing: 0.700.06 (N=28) for probe 33.15 and 0.730.07 (N=28) for probe CA. Band-sharing coefficients calculated for nestlings and all nonfathers and nonmothers gave mean values of 0.470.1 (N=200) for probe 33.15 and 0.490.1 (N=200) for probe CA. For each nestling, the male assigned as the father provided a higher bandsharing coefficient than did other males. This value was always similar to that calculated for the genetic mother, offering support that these values are indicative of first-order relatives. We tried to assign paternity to those nestlings resulting from extrapair fertilization by reconstructing frequency distributions of both novel bands and band-sharing coefficients between these extrapair nestlings and all other males run on each gel. As each nestling was compared with at least nine males on a gel, an element of pseudoreplication was present. However, the importance of the results comes from the trends produced rather than actual frequencies obtained. If the actual genetic father is among the males run on the gel, few or no novel bands should be detected. All other comparisons, between nestlings and nonfathers, should result in more fragments than can be explained by mutation. Furthermore, comparison of band-sharing coefficients should reveal an individual extrapair male with values similar to those gained for first-order relatives as discussed above. A comparison between the extrapair male who has most potential to be
EWEN ET AL.: EXTRAPAIR PATERNITY IN STITCHBIRDS
1.0
0.9
(a)
0.9
0.8
0.8
0.7
0.7
(a)
0.6
0.6
0.5
0.5
0.4 0.4
Band-sharing coefficients
0.2 0.1
0
0.9
1
2
3
4
5
6
7
8
9
10
(b)
0.8
Band-sharing coefficients
0.3 0.3
0.2 0.1
0.9
2
4
6
8
10
(b)
0.8
0.7
0.7
0.6
0.6 0.5
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1 0.1
0
0
1
2 3 No. of novel bands
4
5
Figure 2. Relationship between the band-sharing coefficient and number of novel bands for each nestling compared with the putative father. (a) Minisatellite probe 33.15; (b) minisatellite probe CA. d: Samples extracted from blood; j: samples extracted from tissue.
the genetic father (based on both novel bands and bandsharing coefficients) and the putative father revealed distinct distributions (Fig. 3). The most likely extrapair males had band-sharing coefficients and numbers of novel bands indicative of first-order relatives. They are therefore more likely than the putative fathers to be the genetic fathers. Genetic parentage was assigned for nine of the 10 nestlings whose samples derived from blood. The unassigned individual contained 8–12 novel bands and also had band-sharing coefficients that fell among other nonparent chick dyads when compared with all males run on the same gel. This confirms the individual’s genetic father was not run on the gel. The two nestlings sampled from
0
1
2 3 No. of novel bands
4
5
Figure 3. Relationship between the band-sharing coefficient and number of novel fragments for each extrapair nestling compared with the putative father and the male most likely to be the genetic father. (a) Minisatellite probe 33.15; (b) minisatellite probe CA. j: The relationship between extrapair nestlings and the most likely extrapair male; d: the relationship between those extrapair nestlings and the putative father; m: the two extrapair nestlings sampled from tissue compared with the most likely father (downward pointing) and the putative father (upward pointing).
tissue had all their paternal bands present in an extrapair male run on the same gel. However, one nestling had its paternal bands present in two extrapair males, and therefore paternity was ambiguous. The small number of bands scored in this nestling were lighter than in all other samples, which indicates that the DNA was either underloaded and/or partially degraded. Unfortunately, the majority of these fragments were explained by the female, leaving few potential paternal fragments. Band-sharing distributions for nestling samples derived from tissue showed large overlap between dyads of parent and chicks
325
326
ANIMAL BEHAVIOUR, 58, 2
Figure 4. Realized reproductive success of male stitchbirds on Tiritiri Matangi Island. Circles show nest sites defended by single males. The fractions in each circle show the number of chicks fathered by the resident male over the number sampled from each clutch. Lines with arrows indicate instances of extrapair fertilizations: lines whose origins start in a circle represent instances of paired males fathering nestlings with females other than their mates, and lines starting from floating males represent instances of unpaired males fathering nestlings. Numbers adjacent to lines show the number of young fathered by each extrapair male and the fractions beside floating males indicate the number of successful floating males over the total number of floating males in each breeding season. *Chick whose paternity could not be determined.
and nonparent and chicks. Therefore, assigning parentage was difficult using this method alone. However, these values were useful in determining paternity compared with novel fragments. The single case where an extrapair nestling had all its fragments explained by two males showed clear distinctions in band-sharing coefficients between related versus unrelated dyads. One male provided the highest band-sharing coefficient of 0.774, whereas the second gave a value of 0.667 (among four nonparent chick dyads that lie above 0.6 for this individual).
Realized Reproductive Success Of the 34 scorable nestlings, 35% were the result of extrapair copulation and extrapair paternity was found in eight of 10 clutches. Both paired males (N=3) and unpaired male floaters (N=3) pursued extrapair copulations in this population (Fig. 4). One of eight unpaired male floaters fathered two of four nestlings in a clutch during the 1995–1996 breeding season, while during the 1996– 1997 breeding season two of six unpaired male floaters fathered two nestlings from a clutch of three and a single
EWEN ET AL.: EXTRAPAIR PATERNITY IN STITCHBIRDS
nestling in a clutch of four, respectively. Additionally, three paired males mated with other paired females over the 2 years. The percentage of extrapair young in nests ranged from 0 to 75% and within each nest these young were fathered by the same extrapair male. DISCUSSION Stitchbirds had a high rate of extrapair parentage, comparable to that of some other species (e.g. 35% in indigo buntings: Westneat 1990; 38% in tree swallows: Lifjeld et al. 1993). Extrapair nestlings can be the result of: (1) intraspecific brood parasitism (e.g zebra finches: Birkhead et al. 1990); (2) rapid mate switching and/or brood adoption (Birkhead et al. 1990); or (3) extrapair copulations (review in Birkhead & Møller 1995). We assumed that no instances of intraspecific brood parasitism occurred in the study population (deduced from behavioural observation), and this was confirmed by high maternal band-sharing coefficients. Mate switching and/or brood adoption did not occur. This is important because such behaviour potentially overestimates the reproductive success resulting from extrapair copulations (Birkhead et al. 1990). Where extrapair paternity is high, the observed reproductive output may differ greatly from the realized reproductive success (Gibbs et al. 1990). Assigning paternity to all but one chick allowed us to measure the reproductive success of individuals accurately. Stitchbird males achieved substantial reproductive success through extrapair copulations (Fig. 4). One particular male was paired with a nesting female yet fathered only one nestling from a clutch of three; however, he fathered a further four chicks through extrapair copulation with the neighbouring female. About half of the extrapair nestlings (5/11) were fathered by unpaired males. This differs from the results of both Gibbs et al. (1990) and Hasselquist et al. (1995) where all or the majority of extrapair fertilizations were by already paired males (20/26 and 17/17 extrapair fertilizations, respectively). Westneat et al. (1990) reviewed the literature indicating that among birds most extrapair young result from such mixed behaviours by paired males. In contrast, three of the male floaters gained reproductive success from EPCs in our stitchbird population. To our knowledge, this study is the first to provide evidence that extrapair fertilization by unpaired males can be a successful alternative reproductive strategy. Two of the unpaired floaters had more fertilizations than one paired male who had two broods. The high proportion of stitchbird young fathered by unpaired males may be due to the male bias within this population. Half the adult males were unable to pair and hence their only option was to associate with already paired birds (Ewen 1998). Our results could be an artefact of this population having just been reintroduced. For example, the differential survivorship of sexes after release, the small number of birds in a recently founded population, the lack of mature forest and/or the use of artificial nestboxes may all have altered reproductive behaviours of both males and females. However, the
observed reproductive behaviours are similar to those described by other researchers, both on the only naturally occurring population (Angehr 1984), and on other reintroduced island populations (Lovegrove 1985; Castro et al. 1996). Furthermore, the only naturally occurring population is also thought to be male biased (Rasch 1989). Male-biased populations have been reported in numerous bird species (e.g eastern bluebirds, Sialia sialis: Macdougall-Shackleton et al. 1996; violet-green swallows, Tachycineta thalassina: Beasley 1996) and have been implicated in increasing both male–male competition for females and mate-guarding behaviour (Beasley 1996). Studies of red-winged blackbirds indicate that even in species with unpaired male floaters (Searcy & Yasukawa 1995) extrapair paternity is still predominantly the result of EPCs by already paired, neighbouring males (Gibbs et al. 1990; Westneat 1993). Stitchbirds provide an alternative, indicating that unpaired male floaters may gain substantial reproductive success. This provides evidence for increased competition for females when populations become male biased. Further genetic studies directed at male-biased populations may reveal other instances of unpaired males gaining reproductive success, and thus provide evidence that such biases cause increased competition for breeding opportunities. Acknowledgments We thank Ed Minot, Tim Birkhead, Michael Clarke, John Craig, Bruce Robertson, Tamsin Ward-Smith and Brent Stephenson for their comments on the manuscript. Ray and Barbara Walter, Shaun Dunning, Brice Ebert and Lars Holst Hansen provided support and helped with data collection in the field. We also thank Tania King, Tania Waghorn, Karen Mason and Bruce Robertson for their help in the Molecular Ecology Laboratory at Massey, the Department of Conservation for allowing us to work on an endangered species, and the New Zealand Lottery Grants Board, Research Projects and Funding Committee of Massey University and Massey University Graduate Research Fund for funding this project. References Angehr, G. R. 1984. Ecology and Behaviour of the Stitchbird with Recommendations for Management and Future Research. Internal Report. Wellington: Department of Internal Affairs. Ardern, S. L. & Lambert, D. M. 1997. Is the black robin in genetic peril? Molecular Ecology, 6, 21–28. Ardern, S. L., McLean, I. G., Anderson, S., Maloney, R. & Lambert, D. M. 1994. The effects of blood sampling on the behaviour and survival of the endangered Chatham Island black robin (Petroica traversi). Conservation Biology, 8, 857–862. Ardern, S. L., Ma, W., Ewen, J. G., Armstrong, D. A. & Lambert, D. A. 1997. Social and sexual monogamy in translocated New Zealand robin populations detected using minisatellite DNA. Auk, 114, 120–126. Beasley, B. A. 1996. Males on guard: paternity defences in violet-green swallows and tree swallows. Animal Behaviour, 52, 1211–1224. Birkhead, T. R. & Møller, A. P. 1995. Extra-pair copulation and extra-pair paternity in birds. Animal Behaviour, 49, 843–848.
327
328
ANIMAL BEHAVIOUR, 58, 2
Birkhead, T. R., Burke, T., Zann, R., Hunter, F. M. & Krupa, A. P. 1990. Extra-pair paternity and intraspecific brood parasitism in wild zebra finches Taeniopygia guttata, revealed by DNA fingerprinting. Behavioral Ecology and Sociobiology, 27, 315– 324. Bruford, M. W., Hanotte, O., Brookfield, J. F. Y. & Burke, T. 1991. Single-locus and multilocus DNA fingerprinting. In: Molecular Genetic Analysis of Populations: A Practical Approach (Ed. by R. Hoelzl), pp. 225–269. Oxford: IRL Press. Burke, T. & Bruford, M. W. 1987. DNA fingerprinting in birds. Nature, 327, 149–152. Burke, T., Davies, N. B., Bruford, M. W. & Hatchwell, B. J. 1989. Parental care and mating behaviour of polyandrous dunnocks Prunella modularis related to paternity by DNA fingerprinting. Nature, 338, 249–251. Burke, T., Hanotte, O., Bruford, M. W. & Cairns, E. 1991. Multilocus and single locus minisatellite analysis in population biological studies. In: DNA Fingerprinting: Approaches and Applications (Ed. by T. Burke, G. Dolf, A. J. Jeffreys & R. Wolf), pp. 154–168. Basel: Birkhauser Verlag. Castro, I. 1995. Behavioural ecology and management of hihi (Notiomystis cincta), an endemic New Zealand honeyeater. Ph.D. thesis, Massey University. Castro, I., Minot, E. O., Fordham, R. & Birkhead, T. R. 1996. Polygynandry, face-to-face copulation and sperm competition in the hihi, Notiomystis cincta (Aves: Meliphagidae). Ibis, 138, 765–771. Collar, N. J., Crosby, M. J. & Stuttersfield, A. J. 1994. Birds to Watch 2. The Birdlife International World Checklist of Birds. Cambridge: Birdlife International. Craig, J. L. 1984. Wing noises, wing slots, and aggression in New Zealand honeyeaters (Aves: Meliphagidae). New Zealand Journal of Zoology, 11, 195–200. Craig, J. L. 1985. Stitchbird. In: Complete Book of New Zealand Birds (Ed. by C. J. R. Robertson), page 289. Sydney: Readers Digest. Craig, J. L., Stewart, A. M. & Douglas, M. E. 1981. The foraging of New Zealand honeyeaters. New Zealand Journal of Zoology, 8, 87–91. Craig, J. L., Douglas, M. E., Stewart, A. M. & Veitch, C. R. 1982. Specific and sexual differences in body measurements of New Zealand honeyeaters. Notornis, 28, 121–128. Dixon, A., Ross, D., O’Malley, S. L. C. & Burke, T. 1994. Paternal investment inversely related to degree of extra-pair paternity in the reed bunting. Nature, 371, 698–700. Ewen, J. G. 1998. A genetic and behavioural investigation of extra-pair copulation in stitchbirds (Notiomystis cincta) breeding on Tiritiri Matangi Island. M.Sc. thesis, Massey University. Finch, M. O. & Lambert, D. M. 1996. Kinship and genetic divergence among populations of tuatara Sphenodon punctatus as revealed by minisatellite DNA profiling. Molecular Ecology, 5, 651–658. Galbraith, D. A., Boag, P. T., Gibbs, H. L. & White, B. N. 1991. Sizing bands on autoradiograms: a study of precision for scoring DNA fingerprints. Electrophoresis, 12, 210–220. Gibbs, L. H., Weatherhead, P. J., Boag, P. T., White, B. N., Tabak, L. M. & Hoysak, D. J. 1990. Realized reproductive success of polygynous red-winged blackbirds revealed by DNA markers. Science, 250, 1394–1397. Hasselquist, D., Bensch, S. & von Schantz, T. 1995. Low frequency of extrapair paternity in the polgynous great reed warbler, Acrocephalus arundinaceus. Behavioral Ecology, 6, 27–38. Hasselquist, D., Bensch, S. & von Schantz, T. 1996. Correlation between male song repertoire, extra-pair paternity and offspring survival in the great reed warbler. Nature, 381, 229–232.
Jeffreys, A. J., Wilson, V. & Thein, S. L. 1985. Hypervariable ‘minisatellite’ regions in human DNA. Nature, 314, 67–73. Kempenaers, B., Verheyen, G. R., Van den Broeck, M., Burke, T., Van Broeckhoven, C. & Dhondt, A. A. 1992. Extra-pair paternity results from female preference for high-quality males in the blue tit. Nature, 357, 494–496. Lambert, D. M., Millar, C. D., Jack, K., Anderson, S. & Craig, J. L. 1994. Single- and multilocus DNA fingerprinting of communally breeding pukeko: do copulations or dominance ensure reproductive success? Proceedings of the National Academy of Sciences U.S.A., 91, 9641–9645. Lifjeld, J. T., Dunn, P. O., Robertson, R. J. & Boag, P. T. 1993. Extra-pair paternity in monogamous tree swallows. Animal Behaviour, 45, 213–229. Lovegrove, T. 1985. Stitchbirds on Kapiti Island. Internal Report. Wellington: Department of Lands and Survey. Macdougall-Shackleton, E. A., Robertson, R. J. & Boag, P. T. 1996. Temporary male removal increases extra-pair paternity in eastern bluebirds. Animal Behaviour, 52, 1177–1183. Miller, C. D., Anthony, I., Lambert, D. M., Stapleton, P. M., Bergmann, C. C., Bellamy, A. R. & Young, E. C. 1994. Patterns of reproductive success determined by DNA fingerprinting in a communally breeding oceanic bird. Biological Journal of the Linnean Society, 52, 31–48. Morton, E. S., Forman, L. & Braun, M. 1990. Extrapair fertilisations and the evolution of colonial breeding in purple martins. Auk, 107, 275–283. Oliver, W. R. B. 1955. New Zealand Birds. Wellington: A. H. & A. W. Reed. Petrie, M. & Kempenaers, B. 1998. Extra-pair paternity in birds: explaining variation between species and populations. Trends in Ecology and Evolution, 13, 52–58. Rasch, G. 1985. The ecology of cavity nesting in the stitchbird (Notiomystis cincta). New Zealand Journal of Zoology, 12, 637–642. Rasch, G. 1989. Cavity nesting in stitchbirds and the use of artificial nest sites. Notornis, 36, 27–36. Rasch, G. & Craig, J. L. 1988. Partitioning of nectar resources by New Zealand honeyeaters. New Zealand Journal of Zoology, 15, 185–190. Rasch, G., Boyd, S. & Clegg, S. 1996. Stitchbird (Hihi), Notiomystis cincta Recovery Plan. Threatened Species Recovery Plan. No. 20. Wellington: Department of Conservation. Searcy, W. A. & Yasukawa, K. 1995. Polygyny and Sexual Selection in Red-winged Blackbirds. Princeton, New Jersey: Princeton University Press. Verboven, N. & Mateman, A. C. 1997. Low frequency of extra-pair fertilisations in the great tit Parus major revealed by DNA fingerprinting. Journal of Avian Biology, 28, 231–239. Westneat, D. F. 1987. Extra-pair copulations in a predominantly monogamous bird: observations of behaviour. Animal Behaviour, 35, 865–876. Westneat, D. F. 1990. Genetic parentage in the indigo bunting: a study using DNA fingerprinting. Behavioral Ecology and Sociobiology, 27, 67–76. Westneat, D. F. 1993. Polygyny and extrapair fertilisations in eastern red-winged blackbirds (Agelaius phoeniceus). Behavioral Ecology, 4, 49–60. Westneat, D. F., Sherman, P. W. & Morton, M. L. 1990. The ecology and evolution of extra-pair copulations in birds. In: Current Ornithology (Ed. by D. M. Power), pp. 331–369. New York: Plenum. Wetton, J. H., Carter, R. E., Parken, D. T. & Walters, D. 1987. Demographic study of a wild house sparrow population by DNA fingerprinting. Nature, 327, 147–152.
Floater males gain reproductive success through extrapair fertilizations in the stitchbird JOHN G. EWEN, D. P. ARMSTRONG & DAVID M. LAMBERT
Ecology Group, Institute of Natural Resources, Massey University (Received 21 July 1998; initial acceptance 4 November 1998; final acceptance 8 April 1999; MS. number: 5941)
We used minisatellite DNA profiling to assign parentage to stitchbird, Notiomystis cincta, chicks from a breeding population on Tiritiri Matangi Island off the coast of the North Island of New Zealand. The small population size allowed samples to be collected from all potential parents and nearly (33/34 nestlings) complete assignment of paternity. Analysis revealed that 35% of nestlings (12/34) were the result of extrapair copulation and that extrapair young were present in 80% of nests (8/10). About half of the extrapair nestlings were the offspring of unpaired males. This is substantially higher than predicted from the literature, which suggests that extrapair paternity is typically gained by paired males.
a gel. It is relatively easy to sample both putative parents at a nest and test for genetic relationships with any nestlings (e.g. great tit: Verboven & Mateman 1997; reed bunting, Emberiza schoeniclus: Dixon et al. 1994). Owing to the nature of running gels, and the generally high mobility of the animals being studied, however, it quickly becomes impractical to attempt to assign maternity and paternity to those cases where extrapair parentage has been revealed. Some studies have attempted to address this question by comparing extrapair young with neighbouring adults (e.g. tree swallows, Tachycineta bicolor: Lifjeld et al. 1993; indigo bunting: Westneat 1990). Such studies have revealed that extrapair fertilization is often by other paired males, usually resident on neighbouring territories. As far as we know, only two studies have sampled all possible parents and been able to assign parentage to all nestlings (Gibbs et al. 1990; Hasselquist et al. 1995). Both these studies are consistent with the literature suggesting EPCs were by other paired males. In this study we also sampled all possible parents and aimed to assign paternity to all nestlings. We had two aims. Our first aim was to document the presence of extrapair parentage in a small, isolated and reintroduced population of stitchbirds, Notiomystis cincta. Previous research on stitchbird breeding has reported: (1) the presence of EPCs (Castro et al. 1996); (2) the observation of multimale chases of females during the breeding season (Angehr 1984; Lovegrove 1985); and (3) a reproductive anatomy indicative of a species with intense competition for reproductive success (Castro et al. 1996). Additionally, stitchbirds copulate in a face-to-face position possibly unique among birds (Castro et al. 1996; Ewen 1998). However, no previous study has investigated
Many species whose mating systems have been categorized according to social associations have recently been re-evaluated since genetic analysis of parentage (Burke & Bruford 1987; Wetton et al. 1987) has revealed either extrapair fertilizations (e.g. house sparrow, Passer domesticus: Wetton et al. 1987; great reed warbler, Acrocephalus arundinaceus: Hasselquist et al. 1995) or intraspecific brood parasitism (e.g. zebra finch, Taeniopygia guttata: Birkhead et al. 1990; purple martin, Progne subis: Morton et al. 1990). Results from such studies indicate that there is greater interspecific variability in the extent to which extrapair parentage occurs than previously appreciated (Petrie & Kempenaers 1998), even where extrapair copulations (EPCs) have been observed (e.g. indigo buntings, Passerina cyanea: Westneat 1987, 1990). Research to date indicates that EPCs are part of mixed reproductive behaviours, with paired males seeking copulations with other already paired females (Westneat et al. 1990). Females may also actively seek copulations. Use of DNA profiling allows direct quantification of reproductive success resulting from different behaviours of both males (e.g. red-winged blackbirds, Agelaius phoeniceus: Gibbs et al. 1990) and females (e.g. great reed warbler: Hasselquist et al. 1996; blue tit, Parus caeruleus: Kempenaers et al. 1992). One limitation in the use of minisatellite DNA profiling is the number of individuals that can be run together on Correspondence and present address: J. Ewen, School of Zoology, Faculty of Science and Technology, La Trobe University, Melbourne, Australia (email: [email protected]). D. P. Armstrong and D. M. Lambert are at the Ecology Group, Institute of Natural Resources, Massey University, Private Bag 11-222, Palmerston North, New Zealand. 0003–3472/99/080321+08 $30.00/0
1999 The Association for the Study of Animal Behaviour
321
1999 The Association for the Study of Animal Behaviour
322
ANIMAL BEHAVIOUR, 58, 2
the presence of extrapair parentage in this species. Our second aim was to measure realized reproductive success among males in this population and determine whether males gaining extrapair fertilizations were mated males on neighbouring territories, nonterritorial males or a mixture of the two. METHODS
Behavioural Observation The stitchbird, or hihi, is a medium-sized, sexually dimorphic, forest-dwelling passerine. It is one of three endemic honeyeaters (Meliphagidae) found in New Zealand (Craig et al. 1981; Craig 1984; Rasch & Craig 1988). Males weigh about 40 g and females 28 g (Craig et al. 1982). We studied a population on Tiritiri Matangi Island from the time of their release in August 1995 through their first two breeding seasons of 1995–1996 and 1996– 1997. Tiritiri Matangi Island is an island situated 3 km off the Whangaparaoa Peninsula, 25 km north of Auckland, New Zealand. Stitchbirds were assumed to have previously occupied this island prior to 1890, as it lies well within the historical range of the species. Ten nests were monitored during the study with a total population size of 15 and 18 adult birds (for the 1995–1996 and 1996–1997 seasons, respectively). Stitchbirds breed from October to the end of February (Oliver 1955; Craig 1985). During this time, we checked nestboxes and made behavioural observations to locate breeding females to determine putative parentage. Owing to the lack of mature forest on the island, artificial nestboxes were provided in all large bush patches. We checked nestboxes daily until all females on the island had commenced nest building and again when each female neared the completion of nestling rearing in case she attempted a second clutch. During systematic searches of nestboxes any stitchbirds seen or heard were followed for as long as possible. As the birds are highly mobile, this was never for longer than 30 min. All birds in this population were identifiable by unique combinations of coloured plastic bands and a numbered aluminium band. During nest building and egg laying in stitchbirds in other populations, there are regular chases between males, between females and between males and females (Lovegrove 1985; Rasch 1985; Castro 1995). These chases are accompanied by increased calling by both sexes. In the present study, we used instances of such behaviour as indicators of the initiation of nest building, and followed birds until we found the nest site. Once nest sites were discovered, we observed the pairs daily during nest building and egg laying, and then every third day until the end of the nestling stage (Ewen 1998). Observation periods were between 50 and 120 min and involved a single observer moving quietly around a 30-m radius of the nestbox. These movements were largely restricted to tracks and boardwalks designed for the large numbers of people moving over the island (a sanctuary open to the public) on a daily basis. Female stitchbirds typically have sole responsibility for nest building and incubation (Oliver 1955; Craig 1985)
and they also do most of the provisioning of nestlings (Ewen 1998). Putative mothers were therefore assumed to be the female involved with such behaviours. More than one female has been observed nest building and incubating at one nest in the Kapiti Island population (Castro 1995). However, only a single female was ever observed building and incubating in each nest on Tiritiri Matangi Island. We observed 1–11 male stitchbirds near each nest site. These males were involved in chases with females and other males (Ewen 1998). Despite this, one male spent more time than the other males with each female and attempted to defend a ‘territory’ around the nest site. These males also helped feed the nestlings, albeit at a lower level than the females (Ewen 1998). Such birds were termed ‘paired’ males and were assigned as putative fathers at each nest. All other males observed near the nest (within 30 m) were recorded as putative extrapair males and ranked in relation to the number of times they were seen and the number of EPCs they attempted. This was necessary because there are limits to the number of individuals that can be compared directly on the same gel (see below). In the two breeding seasons studied, the population of stitchbirds was heavily biased towards males, 12:4 in 1995–1996 and 12:6 in 1996–1997. Despite this bias, birds attempted to breed as socially monogamous pairs, resulting in eight and six males, respectively, remaining unpaired in each breeding season.
Minisatellite DNA Profiling Methods Techniques used for collection of blood were as detailed in Ardern et al. (1994). Blood samples (56–76 ìl, mean 66 ìl) were obtained from all adults and surviving nestlings by wing venipuncture of the brachial vein, using a 13-mm, 27-gauge needle. The total amount of blood taken did not exceed 10–20% of the bird’s total blood volume. Blood sampling took ca. 2 min. Once eggs had hatched, we routinely checked nests every 2–3 daylight hours. This was continued until all surviving chicks had been banded and blood samples collected. Female stitchbirds typically remove dead chicks from the nest and discard them soon after discovery. Dead chicks found in the nest were collected for genetic analysis and post mortem examination. If the female had removed chicks, we searched an area of about 5 m in radius in an effort to recover them. Typically, white eggshells gave the location of nest refuse away and subsequently the location of any dead chicks. All chicks recovered had their wings and legs removed and placed in a 20C freezer for between 0.5 and 2 months. All broods with surviving young were bled at 21–25 days after hatching. Nestlings typically fledge 28–30 days after hatching. Therefore, sampling birds at this stage allowed for maximum chick growth with little chance of chicks fledging before sampling. We sampled 10 families for parentage analysis: the single successful clutch of four young in the 1995–1996 breeding season, and nine clutches comprising 36 young in the 1996–1997 breeding season.
EWEN ET AL.: EXTRAPAIR PATERNITY IN STITCHBIRDS
DNA extraction, precipitation and resuspension followed techniques detailed in Ardern & Lambert (1997). DNA from tissue samples was extracted similarly with the following additions. A piece of tissue measuring ca. 663 mm was removed from each dead chick. This was sliced into small pieces with a sterile scalpel blade and added to SET buffer. We added 30 ìl of 10% SDS and 30 ìl of proteinase K and incubated the sample at 55C for 3 h. A further 10 ìl of proteinase K were then added and the sample returned to 55C overnight. An extra Trisbuffered Phenol wash was also performed to remove additional proteins present in tissue. To 20 ìl of genomic DNA, we added 4 ìl 10Buffer (react 2), 2 ìl 2 mg/ml BSA, 1 ìl 160 nM Spermidine, 1 ìl of the restriction enzyme HaeIII and 11 ìl milli-Q water. This mixture was incubated overnight at 37C. Another 1 ìl of HaeIII was added the following day and the sample was incubated at 37C for a further 1 h. Digested samples were then stored at 20C. About 5 ìg of digested DNA were loaded in each lane of the gel. DNA fragments were resolved on a 0.8% agarose gel (1927 cm) in 1TBE running buffer at 55 V for 72 h. We then denatured the DNA by washing each gel for 15 min in 0.25 M HCl and then for 45 min in 0.5 M NaOH, 1.5 M NaCl. Gels were then neutralized by two 15-min washes in 1.5 M NaCl, 0.5 M Tris-HCl pH 7.2, 1 mM EDTA. Southern blot techniques were used to transfer DNA from agarose gels to nylon membranes (Amersham, Little Chalfont, Bucks, U.K.) in 6SSC. Membranes were then dried for 10 min at 37C before being baked at 80C for 2 h. Baked membranes were soaked in prehybridization mix (75 ml 0.5 M disodium hydrogen orthophosphate pH 7.2, 75 ml milli-Q water, 300 ìl 0.5 M EDTA pH 8.0, 10.5 g SDS) for 2 h at 65C. First, Jeffreys 33.15 probe (Jeffreys et al. 1985) was labelled with á-32PdCTP by random priming with Amersham radprime kit. Unincorporated label was removed using a G50 sephadex column. Hybridization of Jeffreys 33.15 to membranes was at 65C for a minimum of 18 h. Membranes were then washed twice with 5SSC, 0.1% SDS at 65C. DNA fragments hybridized to the 33.15 probe were exposed on X-ray film at either 80C with one intensifying screen or at room temperature for 1–6 days. After adequate exposure, membranes were stripped and reprobed with CA probe, which was similarly labelled with á-32PdCTP. Hybridization to membranes was at 55C for a minimum of 18 h. Membranes were washed twice with 5SSC, 0.1% SDS at 55C, and then exposed as detailed for Jeffreys 33.15 probe.
Data Analysis Samples from two broods were loaded on to each gel. All chicks from a brood were loaded with their putative parents on either side. Remaining lanes, situated in the centre of the gel, were loaded with all putative extrapair fathers for those families. This sequence of loading samples minimized distance for between-lane comparisons, a potential source of error in reading gels (Burke et al. 1991). All gels contained molecular weight markers in the outermost lanes (1 and 20). This was important to
determine even running within each gel and allow estimation of band sizes. In addition, gels included one common individual that acted as a genomic control to provide a means of standardization for hybridization conditions and the distance run between gels (Miller et al. 1994). We transposed band positions on to the acetate overlays by dotting band centres with a permanent marker pen (Galbraith et al. 1991). Bands were considered identical in two individuals if there was no more than a two-fold difference in intensity estimated by eye (Bruford et al. 1991; Hasselquist et al. 1995) and they fitted into bin sizes of no more than 1.5 mm (Lambert et al. 1994; Finch & Lambert 1996). We assigned parentage by identifying novel bands (Westneat 1990, 1993) and from the band-sharing coefficients of putative parents (Wetton et al. 1987; Bruford et al. 1991). All summary statistics are given as meansSD.
Ethical Note We did this research under approval from the Department of Conservation Banding Office and the Massey University Animal Ethics Committee. Permission was granted from the Auckland Conservancy of the Department of Conservation to conduct scientific research on Department of Conservation Estate and to conduct the above procedures with stitchbirds. Stitchbirds are listed as a globally threatened species and vulnerable by Birds to Watch 2 (Collar et al. 1994). From this study, we gained information on whether inbreeding was occurring, which would jeopardize the species’ future, and on how many birds were breeding in relation to the number released on Tiritiri Matangi Island. The results were given to the New Zealand Department of Conservation and used in the continued development of conservation initiatives for stitchbirds, in particular the preferred sex ratio and size of populations to be translocated. Birds were caught, banded and bled as part of a standard stitchbird translocation procedure. No obvious change in behaviour was observed after each bird’s release and no bird died as a direct result of such procedures. A previous study (Ardern et al. 1994), which specifically investigated the effects on behaviour and survival of blood sampling an endangered passerine, also found no mortality resulting from the use of these techniques. The experiment caused only temporary disturbances during observation times and nest checks. Most observations were made from boardwalks/tracks used by large numbers of public moving around the island. We therefore assumed an observer would not represent a novel stimulus. Nests were checked only after the female had been observed leaving the nestbox, to avoid disturbing her while she was brooding young. No nests were deserted during the experiment. RESULTS
Banding Profiles Both probes produced variable banding profiles for most individuals (Fig. 1). Five muscle tissue samples
323
324
ANIMAL BEHAVIOUR, 58, 2
Figure 1. DNA fingerprint of a brood of stitchbird nestlings (C1–C3), with their putative mother (PF) and putative father (PM) positioned on either side. Remaining lanes of the fingerprint are putative extrapair fathers (PEPMs). Arrows for chick 2 show bands not present in either putative parent but present in an extrapair male, which indicates an extrapair fertilization.
probed with 33.15 and six muscle tissue samples probed with CA were severely degraded and did not produce scorable profiles. These individuals were ignored for all subsequent analyses. Bands were scored in the approximate size range of 5–23 kb for probe 33.15 and 5–12 kb for probe CA. Samples extracted from whole blood produced banding profiles with more resolvable bands than those extracted from muscle tissue. An average of 21.143.19 and 12.293.4 bands were scored from blood and muscle tissue, respectively, for probe 33.15 and 19.394.15 and 15.25.07 from blood and tissue, respectively, for probe CA. Each probe provided a different set of fragments, although there was some degree of overlap: mean overlap 0.3480.098, N=11 for 33.15 versus CA; 0.2910.087, N=11 for CA versus 33.15.
Parentage Analysis Sixteen of the 34 scorable chicks were completely compatible with their social parents. All remaining chicks had at least one novel band present in their DNA profiles. DNA samples extracted from blood showed either just
one, or six or more novel bands and those derived from tissue showed either one, three or five novel bands. Assuming that the occurrence of a single novel fragment was the outcome of a mutation event, the mutation rate can be calculated. Given that 1192 clearly resolved bands were scored for all chicks and a single novel band was recorded on nine occasions, the mutation rate can be calculated as 9/1192=7.5510 3. This observed mutation rate lies within the order of 10 3 per band per meiotic event calculated for numerous other species (Jeffreys et al. 1985; Burke & Bruford 1987; Burke et al. 1989; Westneat et al. 1990; Ardern et al. 1997; Verboven & Mateman 1997). This suggests that the single novel bands are the result of mutation and not misassigned parentage. Twenty-two of the nestlings either had zero or one novel band, and were therefore considered the genetic offspring of their social parents. The remaining 12 (10 samples derived from blood and two from tissue) must be the result of either extrapair fertilizations or intraspecific brood parasitism. The criterion for assigning parentage based on the presence or absence of novel fragments was supported by comparison of band-sharing coefficients (Fig. 2). Although not completely distinct, those nestlings with no or few novel bands also had high band-sharing values compared with their putative father. Similarly, all nestlings with large numbers of novel bands had lower band-sharing values. Compared with males assigned as genetic fathers through investigation of novel bands, nestlings from which we used blood samples had a mean band-sharing coefficient of 0.710.08 (N=27) for probe 33.15 and 0.700.09 (N=27) for probe CA. This value is similar to that calculated for the mean maternal band sharing: 0.700.06 (N=28) for probe 33.15 and 0.730.07 (N=28) for probe CA. Band-sharing coefficients calculated for nestlings and all nonfathers and nonmothers gave mean values of 0.470.1 (N=200) for probe 33.15 and 0.490.1 (N=200) for probe CA. For each nestling, the male assigned as the father provided a higher bandsharing coefficient than did other males. This value was always similar to that calculated for the genetic mother, offering support that these values are indicative of first-order relatives. We tried to assign paternity to those nestlings resulting from extrapair fertilization by reconstructing frequency distributions of both novel bands and band-sharing coefficients between these extrapair nestlings and all other males run on each gel. As each nestling was compared with at least nine males on a gel, an element of pseudoreplication was present. However, the importance of the results comes from the trends produced rather than actual frequencies obtained. If the actual genetic father is among the males run on the gel, few or no novel bands should be detected. All other comparisons, between nestlings and nonfathers, should result in more fragments than can be explained by mutation. Furthermore, comparison of band-sharing coefficients should reveal an individual extrapair male with values similar to those gained for first-order relatives as discussed above. A comparison between the extrapair male who has most potential to be
EWEN ET AL.: EXTRAPAIR PATERNITY IN STITCHBIRDS
1.0
0.9
(a)
0.9
0.8
0.8
0.7
0.7
(a)
0.6
0.6
0.5
0.5
0.4 0.4
Band-sharing coefficients
0.2 0.1
0
0.9
1
2
3
4
5
6
7
8
9
10
(b)
0.8
Band-sharing coefficients
0.3 0.3
0.2 0.1
0.9
2
4
6
8
10
(b)
0.8
0.7
0.7
0.6
0.6 0.5
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1 0.1
0
0
1
2 3 No. of novel bands
4
5
Figure 2. Relationship between the band-sharing coefficient and number of novel bands for each nestling compared with the putative father. (a) Minisatellite probe 33.15; (b) minisatellite probe CA. d: Samples extracted from blood; j: samples extracted from tissue.
the genetic father (based on both novel bands and bandsharing coefficients) and the putative father revealed distinct distributions (Fig. 3). The most likely extrapair males had band-sharing coefficients and numbers of novel bands indicative of first-order relatives. They are therefore more likely than the putative fathers to be the genetic fathers. Genetic parentage was assigned for nine of the 10 nestlings whose samples derived from blood. The unassigned individual contained 8–12 novel bands and also had band-sharing coefficients that fell among other nonparent chick dyads when compared with all males run on the same gel. This confirms the individual’s genetic father was not run on the gel. The two nestlings sampled from
0
1
2 3 No. of novel bands
4
5
Figure 3. Relationship between the band-sharing coefficient and number of novel fragments for each extrapair nestling compared with the putative father and the male most likely to be the genetic father. (a) Minisatellite probe 33.15; (b) minisatellite probe CA. j: The relationship between extrapair nestlings and the most likely extrapair male; d: the relationship between those extrapair nestlings and the putative father; m: the two extrapair nestlings sampled from tissue compared with the most likely father (downward pointing) and the putative father (upward pointing).
tissue had all their paternal bands present in an extrapair male run on the same gel. However, one nestling had its paternal bands present in two extrapair males, and therefore paternity was ambiguous. The small number of bands scored in this nestling were lighter than in all other samples, which indicates that the DNA was either underloaded and/or partially degraded. Unfortunately, the majority of these fragments were explained by the female, leaving few potential paternal fragments. Band-sharing distributions for nestling samples derived from tissue showed large overlap between dyads of parent and chicks
325
326
ANIMAL BEHAVIOUR, 58, 2
Figure 4. Realized reproductive success of male stitchbirds on Tiritiri Matangi Island. Circles show nest sites defended by single males. The fractions in each circle show the number of chicks fathered by the resident male over the number sampled from each clutch. Lines with arrows indicate instances of extrapair fertilizations: lines whose origins start in a circle represent instances of paired males fathering nestlings with females other than their mates, and lines starting from floating males represent instances of unpaired males fathering nestlings. Numbers adjacent to lines show the number of young fathered by each extrapair male and the fractions beside floating males indicate the number of successful floating males over the total number of floating males in each breeding season. *Chick whose paternity could not be determined.
and nonparent and chicks. Therefore, assigning parentage was difficult using this method alone. However, these values were useful in determining paternity compared with novel fragments. The single case where an extrapair nestling had all its fragments explained by two males showed clear distinctions in band-sharing coefficients between related versus unrelated dyads. One male provided the highest band-sharing coefficient of 0.774, whereas the second gave a value of 0.667 (among four nonparent chick dyads that lie above 0.6 for this individual).
Realized Reproductive Success Of the 34 scorable nestlings, 35% were the result of extrapair copulation and extrapair paternity was found in eight of 10 clutches. Both paired males (N=3) and unpaired male floaters (N=3) pursued extrapair copulations in this population (Fig. 4). One of eight unpaired male floaters fathered two of four nestlings in a clutch during the 1995–1996 breeding season, while during the 1996– 1997 breeding season two of six unpaired male floaters fathered two nestlings from a clutch of three and a single
EWEN ET AL.: EXTRAPAIR PATERNITY IN STITCHBIRDS
nestling in a clutch of four, respectively. Additionally, three paired males mated with other paired females over the 2 years. The percentage of extrapair young in nests ranged from 0 to 75% and within each nest these young were fathered by the same extrapair male. DISCUSSION Stitchbirds had a high rate of extrapair parentage, comparable to that of some other species (e.g. 35% in indigo buntings: Westneat 1990; 38% in tree swallows: Lifjeld et al. 1993). Extrapair nestlings can be the result of: (1) intraspecific brood parasitism (e.g zebra finches: Birkhead et al. 1990); (2) rapid mate switching and/or brood adoption (Birkhead et al. 1990); or (3) extrapair copulations (review in Birkhead & Møller 1995). We assumed that no instances of intraspecific brood parasitism occurred in the study population (deduced from behavioural observation), and this was confirmed by high maternal band-sharing coefficients. Mate switching and/or brood adoption did not occur. This is important because such behaviour potentially overestimates the reproductive success resulting from extrapair copulations (Birkhead et al. 1990). Where extrapair paternity is high, the observed reproductive output may differ greatly from the realized reproductive success (Gibbs et al. 1990). Assigning paternity to all but one chick allowed us to measure the reproductive success of individuals accurately. Stitchbird males achieved substantial reproductive success through extrapair copulations (Fig. 4). One particular male was paired with a nesting female yet fathered only one nestling from a clutch of three; however, he fathered a further four chicks through extrapair copulation with the neighbouring female. About half of the extrapair nestlings (5/11) were fathered by unpaired males. This differs from the results of both Gibbs et al. (1990) and Hasselquist et al. (1995) where all or the majority of extrapair fertilizations were by already paired males (20/26 and 17/17 extrapair fertilizations, respectively). Westneat et al. (1990) reviewed the literature indicating that among birds most extrapair young result from such mixed behaviours by paired males. In contrast, three of the male floaters gained reproductive success from EPCs in our stitchbird population. To our knowledge, this study is the first to provide evidence that extrapair fertilization by unpaired males can be a successful alternative reproductive strategy. Two of the unpaired floaters had more fertilizations than one paired male who had two broods. The high proportion of stitchbird young fathered by unpaired males may be due to the male bias within this population. Half the adult males were unable to pair and hence their only option was to associate with already paired birds (Ewen 1998). Our results could be an artefact of this population having just been reintroduced. For example, the differential survivorship of sexes after release, the small number of birds in a recently founded population, the lack of mature forest and/or the use of artificial nestboxes may all have altered reproductive behaviours of both males and females. However, the
observed reproductive behaviours are similar to those described by other researchers, both on the only naturally occurring population (Angehr 1984), and on other reintroduced island populations (Lovegrove 1985; Castro et al. 1996). Furthermore, the only naturally occurring population is also thought to be male biased (Rasch 1989). Male-biased populations have been reported in numerous bird species (e.g eastern bluebirds, Sialia sialis: Macdougall-Shackleton et al. 1996; violet-green swallows, Tachycineta thalassina: Beasley 1996) and have been implicated in increasing both male–male competition for females and mate-guarding behaviour (Beasley 1996). Studies of red-winged blackbirds indicate that even in species with unpaired male floaters (Searcy & Yasukawa 1995) extrapair paternity is still predominantly the result of EPCs by already paired, neighbouring males (Gibbs et al. 1990; Westneat 1993). Stitchbirds provide an alternative, indicating that unpaired male floaters may gain substantial reproductive success. This provides evidence for increased competition for females when populations become male biased. Further genetic studies directed at male-biased populations may reveal other instances of unpaired males gaining reproductive success, and thus provide evidence that such biases cause increased competition for breeding opportunities. Acknowledgments We thank Ed Minot, Tim Birkhead, Michael Clarke, John Craig, Bruce Robertson, Tamsin Ward-Smith and Brent Stephenson for their comments on the manuscript. Ray and Barbara Walter, Shaun Dunning, Brice Ebert and Lars Holst Hansen provided support and helped with data collection in the field. We also thank Tania King, Tania Waghorn, Karen Mason and Bruce Robertson for their help in the Molecular Ecology Laboratory at Massey, the Department of Conservation for allowing us to work on an endangered species, and the New Zealand Lottery Grants Board, Research Projects and Funding Committee of Massey University and Massey University Graduate Research Fund for funding this project. References Angehr, G. R. 1984. Ecology and Behaviour of the Stitchbird with Recommendations for Management and Future Research. Internal Report. Wellington: Department of Internal Affairs. Ardern, S. L. & Lambert, D. M. 1997. Is the black robin in genetic peril? Molecular Ecology, 6, 21–28. Ardern, S. L., McLean, I. G., Anderson, S., Maloney, R. & Lambert, D. M. 1994. The effects of blood sampling on the behaviour and survival of the endangered Chatham Island black robin (Petroica traversi). Conservation Biology, 8, 857–862. Ardern, S. L., Ma, W., Ewen, J. G., Armstrong, D. A. & Lambert, D. A. 1997. Social and sexual monogamy in translocated New Zealand robin populations detected using minisatellite DNA. Auk, 114, 120–126. Beasley, B. A. 1996. Males on guard: paternity defences in violet-green swallows and tree swallows. Animal Behaviour, 52, 1211–1224. Birkhead, T. R. & Møller, A. P. 1995. Extra-pair copulation and extra-pair paternity in birds. Animal Behaviour, 49, 843–848.
327
328
ANIMAL BEHAVIOUR, 58, 2
Birkhead, T. R., Burke, T., Zann, R., Hunter, F. M. & Krupa, A. P. 1990. Extra-pair paternity and intraspecific brood parasitism in wild zebra finches Taeniopygia guttata, revealed by DNA fingerprinting. Behavioral Ecology and Sociobiology, 27, 315– 324. Bruford, M. W., Hanotte, O., Brookfield, J. F. Y. & Burke, T. 1991. Single-locus and multilocus DNA fingerprinting. In: Molecular Genetic Analysis of Populations: A Practical Approach (Ed. by R. Hoelzl), pp. 225–269. Oxford: IRL Press. Burke, T. & Bruford, M. W. 1987. DNA fingerprinting in birds. Nature, 327, 149–152. Burke, T., Davies, N. B., Bruford, M. W. & Hatchwell, B. J. 1989. Parental care and mating behaviour of polyandrous dunnocks Prunella modularis related to paternity by DNA fingerprinting. Nature, 338, 249–251. Burke, T., Hanotte, O., Bruford, M. W. & Cairns, E. 1991. Multilocus and single locus minisatellite analysis in population biological studies. In: DNA Fingerprinting: Approaches and Applications (Ed. by T. Burke, G. Dolf, A. J. Jeffreys & R. Wolf), pp. 154–168. Basel: Birkhauser Verlag. Castro, I. 1995. Behavioural ecology and management of hihi (Notiomystis cincta), an endemic New Zealand honeyeater. Ph.D. thesis, Massey University. Castro, I., Minot, E. O., Fordham, R. & Birkhead, T. R. 1996. Polygynandry, face-to-face copulation and sperm competition in the hihi, Notiomystis cincta (Aves: Meliphagidae). Ibis, 138, 765–771. Collar, N. J., Crosby, M. J. & Stuttersfield, A. J. 1994. Birds to Watch 2. The Birdlife International World Checklist of Birds. Cambridge: Birdlife International. Craig, J. L. 1984. Wing noises, wing slots, and aggression in New Zealand honeyeaters (Aves: Meliphagidae). New Zealand Journal of Zoology, 11, 195–200. Craig, J. L. 1985. Stitchbird. In: Complete Book of New Zealand Birds (Ed. by C. J. R. Robertson), page 289. Sydney: Readers Digest. Craig, J. L., Stewart, A. M. & Douglas, M. E. 1981. The foraging of New Zealand honeyeaters. New Zealand Journal of Zoology, 8, 87–91. Craig, J. L., Douglas, M. E., Stewart, A. M. & Veitch, C. R. 1982. Specific and sexual differences in body measurements of New Zealand honeyeaters. Notornis, 28, 121–128. Dixon, A., Ross, D., O’Malley, S. L. C. & Burke, T. 1994. Paternal investment inversely related to degree of extra-pair paternity in the reed bunting. Nature, 371, 698–700. Ewen, J. G. 1998. A genetic and behavioural investigation of extra-pair copulation in stitchbirds (Notiomystis cincta) breeding on Tiritiri Matangi Island. M.Sc. thesis, Massey University. Finch, M. O. & Lambert, D. M. 1996. Kinship and genetic divergence among populations of tuatara Sphenodon punctatus as revealed by minisatellite DNA profiling. Molecular Ecology, 5, 651–658. Galbraith, D. A., Boag, P. T., Gibbs, H. L. & White, B. N. 1991. Sizing bands on autoradiograms: a study of precision for scoring DNA fingerprints. Electrophoresis, 12, 210–220. Gibbs, L. H., Weatherhead, P. J., Boag, P. T., White, B. N., Tabak, L. M. & Hoysak, D. J. 1990. Realized reproductive success of polygynous red-winged blackbirds revealed by DNA markers. Science, 250, 1394–1397. Hasselquist, D., Bensch, S. & von Schantz, T. 1995. Low frequency of extrapair paternity in the polgynous great reed warbler, Acrocephalus arundinaceus. Behavioral Ecology, 6, 27–38. Hasselquist, D., Bensch, S. & von Schantz, T. 1996. Correlation between male song repertoire, extra-pair paternity and offspring survival in the great reed warbler. Nature, 381, 229–232.
Jeffreys, A. J., Wilson, V. & Thein, S. L. 1985. Hypervariable ‘minisatellite’ regions in human DNA. Nature, 314, 67–73. Kempenaers, B., Verheyen, G. R., Van den Broeck, M., Burke, T., Van Broeckhoven, C. & Dhondt, A. A. 1992. Extra-pair paternity results from female preference for high-quality males in the blue tit. Nature, 357, 494–496. Lambert, D. M., Millar, C. D., Jack, K., Anderson, S. & Craig, J. L. 1994. Single- and multilocus DNA fingerprinting of communally breeding pukeko: do copulations or dominance ensure reproductive success? Proceedings of the National Academy of Sciences U.S.A., 91, 9641–9645. Lifjeld, J. T., Dunn, P. O., Robertson, R. J. & Boag, P. T. 1993. Extra-pair paternity in monogamous tree swallows. Animal Behaviour, 45, 213–229. Lovegrove, T. 1985. Stitchbirds on Kapiti Island. Internal Report. Wellington: Department of Lands and Survey. Macdougall-Shackleton, E. A., Robertson, R. J. & Boag, P. T. 1996. Temporary male removal increases extra-pair paternity in eastern bluebirds. Animal Behaviour, 52, 1177–1183. Miller, C. D., Anthony, I., Lambert, D. M., Stapleton, P. M., Bergmann, C. C., Bellamy, A. R. & Young, E. C. 1994. Patterns of reproductive success determined by DNA fingerprinting in a communally breeding oceanic bird. Biological Journal of the Linnean Society, 52, 31–48. Morton, E. S., Forman, L. & Braun, M. 1990. Extrapair fertilisations and the evolution of colonial breeding in purple martins. Auk, 107, 275–283. Oliver, W. R. B. 1955. New Zealand Birds. Wellington: A. H. & A. W. Reed. Petrie, M. & Kempenaers, B. 1998. Extra-pair paternity in birds: explaining variation between species and populations. Trends in Ecology and Evolution, 13, 52–58. Rasch, G. 1985. The ecology of cavity nesting in the stitchbird (Notiomystis cincta). New Zealand Journal of Zoology, 12, 637–642. Rasch, G. 1989. Cavity nesting in stitchbirds and the use of artificial nest sites. Notornis, 36, 27–36. Rasch, G. & Craig, J. L. 1988. Partitioning of nectar resources by New Zealand honeyeaters. New Zealand Journal of Zoology, 15, 185–190. Rasch, G., Boyd, S. & Clegg, S. 1996. Stitchbird (Hihi), Notiomystis cincta Recovery Plan. Threatened Species Recovery Plan. No. 20. Wellington: Department of Conservation. Searcy, W. A. & Yasukawa, K. 1995. Polygyny and Sexual Selection in Red-winged Blackbirds. Princeton, New Jersey: Princeton University Press. Verboven, N. & Mateman, A. C. 1997. Low frequency of extra-pair fertilisations in the great tit Parus major revealed by DNA fingerprinting. Journal of Avian Biology, 28, 231–239. Westneat, D. F. 1987. Extra-pair copulations in a predominantly monogamous bird: observations of behaviour. Animal Behaviour, 35, 865–876. Westneat, D. F. 1990. Genetic parentage in the indigo bunting: a study using DNA fingerprinting. Behavioral Ecology and Sociobiology, 27, 67–76. Westneat, D. F. 1993. Polygyny and extrapair fertilisations in eastern red-winged blackbirds (Agelaius phoeniceus). Behavioral Ecology, 4, 49–60. Westneat, D. F., Sherman, P. W. & Morton, M. L. 1990. The ecology and evolution of extra-pair copulations in birds. In: Current Ornithology (Ed. by D. M. Power), pp. 331–369. New York: Plenum. Wetton, J. H., Carter, R. E., Parken, D. T. & Walters, D. 1987. Demographic study of a wild house sparrow population by DNA fingerprinting. Nature, 327, 147–152.