Introduction
Although inter-specific hybridization is fairly common among some birds (Grant & Grant Reference Grant and Grant1992), it is rare in albatrosses, and the few hybrid pairs that have been recorded usually fail to produce chicks (Phillips et al. Reference Phillips, Cooper and Burg2018). Some mixed pairs of black-footed (Phoebastria nigripes) and Laysan albatrosses (P. immutabilis) occasionally produce hybrid young, of which at least some are viable (Fisher Reference Fisher1948, Reference Fisher1972, McCarthy Reference McCarthy2006, Rutt Reference Rutt2013). At Bird Island a black-browed (Thalassarche melanophris) and a white-capped albatross (T. [cauta] steadi) pair produced two hybrid chicks, but both died before fledging (Phillips et al. Reference Phillips, Cooper and Burg2018). Most other records of hybridization among albatrosses are confined to closely-related taxa formerly regarded as subspecies such as between southern royal (D. [e.] epomophora) and northern royal albatrosses (Diomedea [e.] sanfordi; Warham Reference Warham1996, McCarthy Reference McCarthy2006), and between black-browed and Campbell albatrosses (T. [m.] impavida; Moore et al. Reference Moore, Burg, Taylor and Millar2001). At least in the latter case there is evidence of high levels of genetic mixing among these taxa (Burg & Croxall Reference Burg and Croxall2001).
Genetic studies have found surprisingly high rates of extra-pair paternity among albatrosses, given their obligate social monogamy and strong mate fidelity between years (Huyvaert et al. Reference Huyvaert, Anderson, Jones, Duan and Parker2000, Abbott et al. Reference Abbott, Double and Cockburn2006, Burg & Croxall Reference Burg and Croxall2006, Huyvaert et al. Reference Huyvaert, Anderson and Parker2006, Jouventin et al. Reference Jouventin, Charmantier, Dubois, Jarne and Bried2007, Jones et al. Reference Jones, Techow and Ryan2012). These observations raise concerns about the validity of hybrid records based solely on social pairings, particularly in cases where the male parent is a vagrant in the breeding colony. We report the first record of successful hybridization between black-browed and grey-headed albatrosses Thalassarche chrysostoma. Genetic paternity tests were conducted to confirm the putative hybrid chick's provenance. The chick was ringed and it returned to its natal colony 9 and 11 years later, so we also provide a description of the adult phenotype of this hybrid offspring.
Methods
Bird observations have been conducted at Marion Island, the larger of the two Prince Edward Islands, sporadically since the 1950s, and more regularly since the 1970s (Cooper & Brown Reference Cooper and Brown1990). The grey-headed albatross is the only Thalassarche albatross that breeds on the island. Annual censuses of grey-headed albatross colonies have been made since 1997, when a study colony in which all adults and chicks are ringed was established on Grey-headed Albatross Ridge (46°52'S, 37°41'E, Converse et al. Reference Converse, Kendall, Doherty and Ryan2009). In December 2000, a black-browed albatross was found incubating an egg in another part of the grey-headed albatross colony on Grey-headed Albatross Ridge (away from the study area). It was caught and banded, and its partner subsequently confirmed to be a grey-headed albatross. Observers have looked for the presence of the black-browed albatross in most years since its initial discovery.
Blood was collected from both putative parents and their chick in 2007 and stored in lysis buffer. The sex of the parents was determined by amplification of the CHD gene, using primers 2550F and 2718R following the protocol of Fridolfsson & Ellegren (Reference Fridolfsson and Ellegren1999). Total genomic DNA was extracted using an extraction solution of 10% Chelex® 100 Resin (BioRad), 10 mM Tris, 0.2% SDS and 5 µl of Proteinase K (100 mg/mL). Approximately 20 µl of blood was added to 200 µl of the extraction solution and incubated at 65°C for at least 8 hours followed by boiling for 10 min. Eight microsatellite loci were used to genotype both parents and the chick: 6A3, 11F3, 12H8, 12C8, 10C5, 12E1, 11H1 and 7D8 (Dubois et al. Reference Dubois, Jarne and Jouventin2005). Microsatellite loci were amplified by polymerase chain reaction (PCR). PCRs were multiplexed in 2 subsets of loci (6A3, 11F3, 12H8, 12C8, 10C5, 11H1 and 7D8; 12E1) in a 10 µl final volume, including 0.2 µM of each primer and 1 µl genomic DNA, using the Qiagen multiplex PCR kit (Qiagen). PCRs were conducted using an ABI GeneAmp® PCR System 2700 under the following conditions: 15 min activation of the HotStartTaq DNA polymerase at 95°C, 30 cycles of 30 sec of initial denaturation at 94°C, 90 sec annealing (58°C for 6A3, 11F3, 12H8, 12C8, 10C5, 11H1 and 7D8; 54°C for 12E1) and 60 sec extension at 72°C, and then a final extension of 30 min at 60°C. PCR products were combined and electrophoresed on an ABI3730xl using POP7 and a 50 cm capillary using Rox350 (Applied Biosystems) as the standard at the Central DNA Sequencing Facility of the University of Stellenbosch (http://www.sun.ac.za/saf). Profiles were analysed using GeneMapper Software version 3 (Applied Biosystems). Genotypes were confirmed by repeating PCRs and electrophoresis for each individual.
CERVUS version 3.0.7 (Marshall et al. Reference Marshall, Slate, Kruuk and Pemberton1998, Kalinowski et al. Reference Kalinowski, Taper and Marshall2007) was used to assess maternity and paternity. The programme uses a likelihood approach that allows for genotyping error and mutation to estimate the probability of maternity and paternity. Adult grey-headed albatrosses (n = 200) sampled from the same colony on Marion Island and screened for the same microsatellite loci (FitzPatrick Institute unpublished data), were used to calculate allele frequencies and to compute simulations. An LOD (logarithm of odds) score as well as a Delta score were calculated for the mother, putative father, and both parents and chick (trio). LOD scores are the sum of the log-likelihood ratios at each locus calculated for each candidate. Delta is a derivative of the LOD score and may be used as an additional measurement. When considering a set of candidates with a positive LOD score, Delta is defined as the difference in LOD scores between the most likely candidate parent and the second most likely candidate parent. If no candidate has a positive LOD score, then Delta is undefined. Delta is especially useful when several candidates have positive LOD scores (Marshall et al. Reference Marshall, Slate, Kruuk and Pemberton1998). CERVUS compares the distribution of LOD or Delta scores of the simulation with the actual data. Consequently, confidence in assignment is defined as the proportion of candidate parents with LOD or Delta scores that exceed a given LOD or Delta score that are true parents (Marshall et al. Reference Marshall, Slate, Kruuk and Pemberton1998, Slate et al. Reference Slate, Marshall and Pemberton2000).
Results
The vagrant black-browed albatross ringed on Grey-headed Albatross Ridge in the 2000/01 breeding season has been resighted in 12 subsequent years from 2001/02 to 2018/19. However, breeding has only been recorded in four years: 2000/01, 2002/03, 2006/07 and 2008/09. It was only in 2006 that the partner was confirmed to be a grey-headed albatross, when it was also ringed. The egg apparently failed to hatch in three of these attempts; it was only in 2006/07 that a chick hatched (sometime between 26 December and 14 January). The chick was banded in March and fledged in May 2007. The black-browed albatross was seen in the 2007/08 season but did not breed (presumably because its partner was on sabbatical after raising a chick the preceding year). The same pair was found incubating in 2008, but the breeding attempt failed at the egg stage. Since then, the black-browed albatross has been observed loafing in the vicinity of its former breeding site in 2011/12 and each breeding season from 2013/14 to 2018/19. Its partner has not been re-sighted since 2008.
Genetic sexing indicated that the black-browed albatross was female. Unfortunately, despite running the sample from the grey-headed partner several times, we were unable to unequivocally confirm its sex, although one run showed faint amplification suggesting it was female. This raises the possibility of a female-female pairing (cf. Young et al. Reference Young, Zaun and Vanderwerf2008). CERVUS version 3.0.7 analysis of maternity indicated the black-browed albatross to be the mother of the chick (no mismatches between genotypes, LOD of 2.16 and a Delta of 1.86, P < 0.05). A CERVUS analysis of paternity showed that although the grey-headed partner had an 80% confidence level of being the father when considered alone (three mismatches between genotypes), when considered with the known mother the trio LOD score was -7.7 (Delta 0.0; four mismatches between genotypes) and the Delta score for the father-chick pair was < 80% confidence level.
Juveniles of both parental species exhibit considerable variation in head pattern and can be difficult to differentiate (Marchant & Higgins Reference Marchant and Higgins1990, Brooke Reference Brooke2004), but the fledgling had the grey head typical of a juvenile grey-headed albatross, and lacked the white face and grey collar characteristic of juvenile black-browed albatrosses, suggesting the possibility that it may not be of hybrid origin. However, in 2018, 11 years later the putative hybrid was photographed in its natal breeding colony and its phenotype was intermediate between the two species (Fig. 1). Its nape was white, shading to pale grey towards the face, and the yellow culminicorn and lower ramicorn were broader than on an adult grey-headed albatross, with a smaller, mottled black region (on the upper ramicorn and sulcus) that petered out towards the gonys. The peach-orange tip of the maxillary unguis was more extensive and paler than in a grey-headed albatross, resembling that of a black-browed albatross (Fig. 1). A second individual, resembling the hybrid, was seen in the colony in 2017. Although its history is unknown, its locality and hybrid phenotype suggest that the black-browed albatross may have fledged another chick.
Black-browed × grey-headed albatross hybrid photographed at its natal colony on Grey-headed Albatross Ridge, Marion Island, 11 years after fledging (photograph by C.W. Jones)
Discussion
Although some hybrids among Phoebastria albatrosses apparently result from forced copulations (Rohwer et al. Reference Rohwer, Harris and Walsh2014), our observations suggest that the black-browed albatross formed a mixed species pair with a grey-headed albatross. We can only confirm it was the same partner in two breeding attempts, but it is possible that the same birds were involved in all four breeding attempts. The fact that only one egg hatched might reflect some level of genetic incompatibility between these two individuals; similar apparent incompatibility has been reported among within-species pairs of other albatrosses (e.g. Jones et al. Reference Jones, Techow and Ryan2012). The intermediate adult phenotype of the sole chick raised by this pair confirms its hybrid origin, and together with the microsatellite results, suggests that the chick resulted from an extra-pair fertilization by another grey-headed albatross.
This is the third record of a mixed-species pairing between black-browed albatrosses and another Thalassarche albatross. Pairings between the genetically close (Chambers et al. Reference Chambers, Moeke, Steel and Trueman2009) Campbell and black-browed albatrosses at Campbell Island have resulted in viable F1 hybrids (Moore et al. Reference Moore, Burg, Taylor and Millar2001). However, at South Georgia both chicks hatched by a female black-browed albatross and sired by a more distantly related white-capped albatross (Chambers et al. Reference Chambers, Moeke, Steel and Trueman2009) died before fledging (Phillips et al. Reference Phillips, Cooper and Burg2018). The grey-headed albatross is more closely related to black-browed/Campbell albatrosses than is the white-capped albatross (Chambers et al. Reference Chambers, Moeke, Steel and Trueman2009), which might explain the production of a hybrid that was able to both fledge and survive to return to the breeding colony a decade later.
Black-browed and grey-headed albatrosses share breeding colonies at South Georgia (Prince & Ricketts Reference Prince and Ricketts1981), the Crozets (Weimerskirch et al. Reference Weimerskirch, Jouventin and Stahl1986), Campbell Island (Waugh et al. Reference Waugh, Weimerskirch, Cherel and Prince2000) and Macquarie Island (Jaimie Cleeland, personal communication 2018). Their breeding phenology and behaviours, including courtship and aggressive repertoires, are similar (Tickell Reference Tickell1984). Courtship displays have been observed between the two species at South Georgia (Warham Reference Warham1996), and experimental cross fostering of eggs between the species has shown that each can successfully rear chicks of the other species (Prince & Ricketts Reference Prince and Ricketts1981). Ours is the first confirmed record of a mixed species pair between the two species (Marchant & Higgins Reference Marchant and Higgins1990, Tickell Reference Tickell2000). However, at least one suspected black-browed × grey-headed albatross hybrid has been photographed off Australia (http://www.pbase.com/wildlifeimages/port_fairy_albatross). The appearance of the Australian bird(s), photographed in 2008, 2012 and 2017 (https://birdlifephotography.org.au/index.php?option=com_blphotos&view=show_image&id=25679) is similar to (but not the same as) the hybrids seen on Marion in 2016 and 2018. Hybridization among birds often occurs when a species is out of its normal range and has few if any opportunities to breed with conspecifics (Grant & Grant Reference Grant and Grant1992), as is the case in our study.
Strong mate fidelity is to be expected in albatrosses given their long pair bond formation and low reproductive rates (Moller Reference Moller2000). However, at Bird Island, South Georgia, rates of extra-pair paternity were 0–9% among black-browed albatrosses and 3–10% among grey-headed albatrosses (Burg & Croxall Reference Burg and Croxall2006). Extra-pair copulations insure against male infertility or partner incompatibility and may enable females to produce young while benefiting from the assistance of her social mate (Wink & Dyrcz Reference Wink and Dyrcz1999). Further, black-browed albatross typically breed annually whereas grey-headed albatrosses biennially (Prince et al. Reference Prince, Rothery, Croxall and Wood1994, Ryan et al. Reference Ryan, Phillips, Nel and Wood2007) which may hinder the development of a stable pair bond and render such a mixed species pair more prone to extra-pair behaviour. Genetic compatibility in a black-browed and grey-headed albatross pairing may be limited and might explain the failure of the pair's other breeding attempts. Extra-pair fertilisation was possibly the only option for the female black-browed albatross (Gowaty Reference Gowaty and Black1996) but she was still constrained to mate with a grey-headed albatross since they are the only Thalassarche albatross present at her breeding site on Marion Island.
Albatrosses generally are philopatric to their natal colonies, so the appearance of a black-browed albatross at Marion Island is remarkable. However, occasional long-range dispersal movements do occur (e.g. Inchausti & Weimerskirch Reference Inchausti and Weimerskirch2002), and are relatively common among black-browed albatrosses, as indicated by their recent colonisation at Campbell, Macquarie, Antipodes islands and The Snares (Moore et al. Reference Moore, Burg, Taylor and Millar2001) and the famous individual that was observed in gannet colonies in Scotland (Agreement on the Conservation of Albatrosses and Petrels 2010). It is unclear whether this reflects a greater dispersal tendency in this species, or is merely a result of their large population size and wide geographic range (Phillips et al. Reference Phillips, Gales, Baker, Double, Favero, Quintana, Tasker, Weimerskirch, Uhart and Wolfaardt2016).
Acknowledgements
We thank the many field workers on Marion Island who looked for and reported sightings of the black-browed albatross. Permits (including animal research protocols), logistical and financial support were provided by the South African Department of Environmental Affairs, through the South African National Antarctic Programme (SANAP), the National Research Foundation and the University of Cape Town. We also thank two reviewers for their helpful comments.
Author contributions
MGWJ, NMST and PGR wrote the manuscript and MMR and CWJ added to its preparation. NMST conducted the genetic analyses. MGWJ, MMR, CWJ, QAH and FT conducted the field work. PGR supervised the research.
