Biological Journal of the Linnean Society, 2010, 100, 725–736. With 3 figures Widespread hybridization between the Greater Spotted Eagle Aquila clanga and the Lesser Spotted Eagle Aquila pomarina (Aves: Accipitriformes) in Europe ÜLO VÄLI1,2*, VALERY DOMBROVSKI3, RIMGAUDAS TREINYS4, UGIS BERGMANIS5, SZILÁRD J. DARÓCZI6, MIROSLAV DRAVECKY7, VLADIMIR IVANOVSKI8, JAN LONTKOWSKI9, GRZEGORZ MACIOROWSKI10, BERND-ULRICH MEYBURG11, TADEUSZ MIZERA10, RÓBERT ZEITZ6 and HANS ELLEGREN1 1 Department of Evolutionary Biology, Evolutionary Biology Centre, Uppsala University, Norbyvägen 18D, SE-75236 Uppsala, Sweden 2 Institute of Agricultural and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51014 Tartu, Estonia 3 Laboratory of Ornithology, Institute of Zoology of National Academy of Sciences, Akademichnaya str. 27, 220072 Minsk, Belarus 4 Institute of Ecology of Nature Research Centre, Akademijos 2, LT–08412 Vilnius, Lithuania 5 Nature Reserve Teiči, Aiviekstes 3, LV-4862 Laudona, Madonas raj, Latvia 6 Milvus Group, Crinului 22, 540343 Tg.-Mures, Romania 7 Rovniková 8, SK-04012 Košice, Slovakia 8 APB-Birds Conservation Belarus, P.O. Box 306, 220050 Minsk, Belarus 9 Museum of Natural History, Wrocław University, ul. Sienkiewicza 21, PL 50-335 Wrocław, Poland 10 University of Life Sciences, Zoology Department, Wojska Polskiego 71c, PL 60-625 Poznań, Poland 11 World Working Group of Birds of Prey, Wangenheimstraße 32, D-14193 Berlin, Germany Received 6 November 2009; revised 15 February 2010; accepted for publication 15 February 2010 bij_1455 725..736 Hybridization is a significant threat for endangered species and could potentially even lead to their extinction. This concern applies to the globally vulnerable Greater Spotted Eagle Aquila clanga, a species that co-occurs, and potentially interbreeds, with the more common Lesser Spotted Eagle Aquila pomarina in a vast area of Eastern Europe. We applied single nucleotide polymorphism (SNP) and microsatellite markers in order to study hybridization and introgression in 14 European spotted eagle populations. We detected hybridization and/or introgression in all studied sympatric populations. In most regions, hybridization took place prevalently between A. pomarina males and A. clanga females, with introgression to the more common A. pomarina. However, such a pattern was not as obvious in regions where A. clanga is still numerous. In the course of 16 years of genetic monitoring of a mixed population in Estonia, we observed the abandonment of A. clanga breeding territories and the replacement of A. clanga pairs by A. pomarina, whereby on several occasions hybridization was an intermediate step before the disappearance of A. clanga. Although the total number of Estonian A. clanga ¥ A. pomarina pairs was twice as high as that of A. clanga pairs, the number of pairs recorded yearly were approximately equal, which suggests a higher turnover rate in interbreeding pairs. This study shows that interspecific introgressive hybridization occurs rather frequently in a hybrid zone at least 1700-km wide: it poses an additional threat for the vulnerable A. clanga, and may contribute to the extinction of its populations. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736. ADDITIONAL KEYWORDS: avian hybridization – extinction – hybrid identification – introgression – microsatellites – raptor – single nucleotide polymorphism – spotted eagles. *Corresponding author. Current address: Institute of Agricultural and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51014 Tartu, Estonia. E-mail: ulo.vali@emu.ee © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 725 726 Ü. VÄLI ET AL. INTRODUCTION Interspecific hybridization, i.e. the interbreeding of individuals from different species, occurs surprisingly frequently in nature (Arnold, 1997). Hybridization could play a role in speciation, and hence provide new material for evolution (Newton, 2003; Price, 2008). Hybridization is often directly associated with rarity: rare species hybridize more easily because the low numbers that lead to a lack of mates is often the main factor to trigger hybridization (Hubbs, 1955; Short, 1969; Randler, 2006). Hybridization is therefore of conservation importance, and presents a serious threat to many endangered species (Rhymer & Simberloff, 1996): it is often associated with the swamping of one species by another (Allendorf et al., 2001; Newton, 2003). Although introgression of foreign genes may be inhibited by the inviability or sterility of hybrids of the heterogametic sex (Haldane, 1922), hybridization is a waste of reproductive effort for the rarer species, calling its continuing sustainability into question. Hybridization is relatively common in birds. Almost one-tenth of bird species have been considered to interbreed in nature and produce hybrid offspring (Panov, 1989; Grant & Grant, 1992), although the proportion may be even twice as high as that (McCarthy, 2006; Aliabadian & Nijman, 2007). However, regular hybridization is rather uncommon in raptors (Panov, 1989; McCarthy, 2006), and there are only relatively few well-studied examples of hybridization occurring in this avian group, such as a hybridizing complex of large falcons (Eastham & Nicholls, 2005; Nittinger et al., 2007). The Greater Spotted Eagle Aquila clanga Pallas, 1811 and the Lesser Spotted Eagle Aquila pomarina Brehm, 1831 are two closely related Eurasian raptors, with overlapping ranges in Eastern Europe, whose numbers have decreased during the last century mainly because of a decline in habitats (Hagemeijer & Blair, 1997). The decline has been particularly dramatic in A. clanga, whose vast range across Eurasia is occupied by only a few thousand pairs, with less than a thousand pairs breeding in Europe (BirdLife International, 2004). In contrast, populations of A. pomarina are still dense, and in most regions significantly outnumber the sparsely represented A. clanga (Hagemeijer & Blair, 1997; BirdLife International, 2004). Both species are listed in Annex I of the EU Directive on the Conservation of Wild Birds (EEC/79/409), as well in the International Union for Conservation of Nature (IUCN) Red List: A. clanga as a globally vulnerable species and A. pomarina as a species of least conservation concern (IUCN, 2009). The two spotted eagles are relatively young species (Seibold et al., 1996; Lerner & Mindell, 2005; Helbig et al., 2005a). There is no complete reproduc- tive barrier between them and hybridization has been recorded repeatedly, although mainly in the form of anecdotal evidence of single cases of interbreeding or probable hybrids (Löwis, 1888, 1898; Feldt, 1909; Transehe, 1942, 1965; Bergmanis et al., 1997; Bergmanis & Strazds, 2001; Lõhmus & Väli, 2001; Dombrovski, 2002; Gutiérrez & Villa, 2002; Meyburg et al., 2005; Treinys, 2005). Studies to assess the extent of hybridization have been hindered by the difficulties posed by hybrid identification. Not all hybrids can be identified by morphology and, if they are fertile and the hybridization is introgressive, detection of backcrosses is even more challenging (e.g. Benedict, 1999; Eastham & Nicholls, 2005; Gaubert et al., 2005; McCarthy, 2006). Hybrid detection is particularly difficult when interbreeding species are similar, and spotted eagles undoubtedly belong to this category (Randler, 2004) because many of their morphological identification characters show overlapping variation (Bergmanis, 1996; Forsman, 1999; Väli & Lõhmus, 2004; Dombrovski, 2006). However, genetic methods have recently provided new tools for hybridization studies. In spotted eagles, after the first applications of maternally inherited mitochondrial DNA (Väli & Lõhmus, 2004; Helbig et al., 2005b), multilocus amplified fragment length polymorphism (AFLP) markers were used in order to study differentiation and gene flow among species (Helbig et al., 2005b), but these markers remained relatively ineffective for individual assignments. Recently, Väli et al. (2010) applied a set of single nucleotide polymorphism (SNP) markers that, in combination with microsatellites, provided enough resolution power to efficiently identify pure species, F1 hybrids (offspring of interbreeding A. clanga and A. pomarina), and backcrosses to parental species. Here, we apply this marker set in a spatiotemporal analysis of spotted eagle hybridization in Europe. On the spatial scale, we search for hybridization events across Europe and ask how widespread hybridization is and whether it poses a significant threat for the vulnerable A. clanga. On the temporal scale, we ask whether hybridization increases the risk of extinction in declining A. clanga populations, and verify this by results from the genetic monitoring of a mixed spotted eagle population in Estonia. MATERIAL AND METHODS SAMPLES In order to estimate the current distribution and frequency of hybridization, a total of 738 spotted eagles were sampled in the wild between 1994 and 2009 (after 2001 in most European populations). Samples studied represent 408 breeding pairs of © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 HYBRIDIZATION OF SPOTTED EAGLES IN EUROPE 727 Figure 1. Distribution and frequency of spotted eagle hybridization in Europe. We have focused on Aquila clanga and hybridizing pairs, and excluded Aquila pomarina pairs, in all populations except those where only A. pomarina was found (in black). Other pie charts show the relative proportion of A. clanga ¥ A. clanga (blue), A. clanga ¥ A. pomarina (red), F1 hybrid ¥ A. clanga (violet), F1 hybrid ¥ A. pomarina (green), and F1 ¥ F1 (yellow) in populations after excluding A. pomarina. The numbers above each pie chart indicate the sample size both including A. pomarina (before slash) and excluding A. pomarina (after slash). The distribution range of A. clanga is shaded in blue and that of A. pomarina is in green (mainly according to Cramp & Simmons, 1980 and Hagemeijer & Blair 1997), but new field observations and museum data from the eastern limit (Melnikov et al., 2001; Mischenko et al., 2010; V. Belik, V. Dombrovski & M. Dzmitranok, unpubl. data) have been taken into account as well. spotted eagles from 14 populations across Europe, covering both sympatric and allopatric regions of the distribution ranges (Fig. 1; Table 1), as well as two reference individuals from the Asian part of Russia, and thus remote from all A. pomarina populations. Instead of ‘individuals’, we prefer to use ‘breeding pair’ as a study unit because: (1) this is the common standard in the monitoring of raptor populations where non-territorial individuals are difficult to count; (2) samples were collected at nest sites; and (3) we made a consensus decision based on the assignment results of offspring and one or two adults (see below). In the spatial analysis, each pair was used only once in the final calculation of hybridization frequency. A temporal study was conducted in Estonia, at the north-western limit of both species’ distribution ranges, where spotted eagles have been continuously monitored (nests checked, birds scrutinized, and genetic samples collected) since the mid 1990s. According to estimates by Lõhmus (1998) from that period, A. clanga was present in some 5% of Estonian spotted eagle breeding pairs, i.e. in 20–30 out of 500–600 pairs. We included a total of 376 individuals from Estonia (a subsample of the total sample), representing 286 breeding attempts in 173 breeding territories (between 1994 and 2009, but mostly from 1999–2009; Appendix; Table 1). In the temporal study, assignment decisions were made on an annual basis, and several individuals of these long-lived species were studied repeatedly, which enabled us to verify the accuracy of genotyping. Blood or feather samples were obtained during the breeding season at nest sites, and in their vicinity, from both nestlings and adults. The morphology of the nestling (in spotted eagles usually only one offspring fledges) was scrutinized thoroughly during sampling, and parental birds were described when © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 728 Ü. VÄLI ET AL. Table 1. Classification of breeding pairs studied in 14 populations. In addition to the five groups included in the main analysis, intermediates and potential F1 ¥ F1 individuals are also shown No. Population N A. cla. A. cla./F1 ¥ A. cla. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Estonia North-western Russia Latvia Lithuania North-eastern Belarus North-eastern Germany Northern Poland North-eastern Poland Southern Belarus South-eastern Poland Slovakia Romania Greece Caucasian Russia Total 173 9 34 38 13 10 22 10 29 31 20 6 5 8 408 6 6 0 0 0 0 0 3 19 0 0 0 0 0 35 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 they were present (see Forsman, 1999; Väli & Lõhmus, 2004 for a comprehensive description of the morphological traits we used). If an A. pomarina nestling had characteristics typical of its species, and at least one of its parents was unequivocally assigned to the species by appearance, only the nestling sample was normally used in the genetic analysis; the adult’s feathers were used only occasionally to confirm the assignment of its nestling. However, territories inhabited by A. clanga or potential hybrids were usually studied more carefully, and whenever possible we included samples from adults. To exclude the possibility of sampling nest-visiting adults from other pairs (Meyburg, Meyburg & Franck-Neumann, 2007), we also checked the parentage in adult–nestling pairs by comparing individual genotypes (data not shown). Additional field observations on species identification and pair bonds were used to complement the genetic data, especially in any unsuccessful breeding years of the temporal study. Field records were particularly informative in the case of males, which usually do not shed feathers at nest sites (Lõhmus & Väli, 2004). On the other hand, males are easier to observe while hunting, and are more protective at nest sites, and therefore easier to trap and study in the hand. LABORATORY ANALYSIS DNA was extracted from blood cells and freshly plucked feathers using proteinase-K treatment followed by salting (Aljanabi & Martinez, 1997) or the phenol-chloroform purification method. From small moulted contour feathers, and some large flight feath- F1 ¥ A. cla A. cla ¥ A. pom F1 ¥ F1 F1 ¥ A. pom. 0 1 0 0 0 0 0 0 1 0 0 0 0 0 2 11 0 0 1 0 2 0 4 2 1 0 0 0 0 20 1 0 0 0 0 0 0 1 0 0 0 0 0 0 2 7 1 1 1 3 0 0 1 2 0 0 0 0 0 16 A. pom./F1 ¥ A. pom. A. pom. 4 0 1 1 0 0 0 1 0 0 0 0 0 0 6 144 1 32 35 10 8 22 0 4 30 20 6 5 8 326 ers, we extracted DNA from the basal tip using the DNEasy tissue kit (Qiagen). Usually, however, DNA was extracted from the blood clot in the superior umbilicus of flight feathers following the methodology of Horvath et al. (2005). This blood clot is secure from contamination and the quantity of DNA is relatively large, and thus the simple phenol-chloroform protocol could be used to obtain a high yield of good-quality DNA. In order to assign individuals to species, or to one of the hybrid classes, birds were genotyped by 30 nuclear markers. Twenty-two microsatellites, developed for other eagle species, were amplified and genotyped on a MegaBACE 1000 automated capillary sequencer (Amersham Biosciences), and eight nuclear SNP markers were analysed by restriction enzyme digestion and electrophoresis on 3% agarose gel (Table 2), as described by Väli et al. (2010). However, in order to ensure amplification success in poorer feather samples as well, we designed new primers for all SNP markers (only two 5′ primers were retained) for amplification of shorter DNA fragments (200– 300 bp) (Table 2). Although not used directly in the assignment process, information on the maternal lineage of the hybrid offspring studied was obtained from the mitochondrial DNA pseudo-control region, which differs by some 4% between the two species (Väli, 2002). Species-specific lineage was detected either by sequencing or by restriction enzyme analysis using the Cac8I enzyme, which has a specific recognition site in A. clanga. In order to sex adult birds, a portion of the CHD1 gene was amplified according to the methodology of Griffiths et al. (1998) © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 HYBRIDIZATION OF SPOTTED EAGLES IN EUROPE 729 Table 2. Genetic markers used in the assignment of individuals. For each marker, its type [single nucleotide polymorphism (SNP) or microsatellite (Ms)], primer sequence, and reference are presented. Primers marked with an asterisk are the same as in Väli et al. (2010). [References: this study; Martinez-Cruz et al. (2002); Hailer, Gautschi & Helander (2005); Busch et al. (2005).] Marker Type Forward primer (5′–3′) Reverse primer (5′–3′) Ref. 1.26928 4.12303 4.FIB 5.15691 7.04557 8.17388 13.12260 17.14657 Aa02 Aa12 Aa15 Aa26 Aa27 Aa35 Aa39 Aa43 Aa49 Aa53 Aa57 Hal1 Hal4 Hal7 Hal9 Hal13 Hal18 IEAAAG04 IEAAAG12 IEAAAG13 IEAAAG14 IEAAAG15 SNP SNP SNP SNP SNP SNP SNP SNP Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms Ms GACCTTCCAGAAGCTATTGC* TGTTCAAATGTCCTTGGTTG TGGGTCCTGAGGAAAGACAG AATGGTCCAACTTTGAAATCT ACTAACTTGCTTGCCATGTG CAGCGTAATGTACCAAATGC AAGCAGAAGCTGTCTTCCG* ACTTGGCCACAGTGAGTATG CTGCAGATTTCACCTGTTCTG TCATCAACCTGACCCTTTCC TCACTGACCTGCCCTCTACA GCAAAGGTAAACTGCATCTGG GAGATGTCTTCACAGCTTGGC GCAGCAGAAAGTGCATACGA TTCTGTTTTTCCACTTGCTTG CCACACTGAGAAACTCCTGTTG AGGAGGTGCCAGTTTTCTCC ATCGCTTCCATGAGCTGATT AACATTAAGGCAGATGTGGACA GAATACACCCAGAACAGCAACC TAAGGCTTTTCTTCGCGTGT TTCAGAAGGTGCATGCAGTAG TGAGCTTTGTAGTAGCAGTGGTG CCACTCAGTAAGGAGCTTTGC GACAGGGAGCGAGTTAGTGG GCATGTAACAAGTTTAATGTTGATGG GCTGCTGCTAAGAATCACTCATGTAC GAATACCACAATAAGAGGCAGAGTG GTCCAGATTCCCTGCTAGAAAGC GAGAATAATTTTTGAAAAGCAGTAGG ATTTAGCATGGGTGAGCCTA CTGCACGAGACATGCTTAAC TTCCCCAATCTAAACAATTCC TCAGCTTATTTGTTGCTCCA GCTGTCTGCCCTTCAGTAAT CCCAGTCTTTTCAGCTTAGG GCCCTTCACAAGTCAGGTAA TCACTAATGTTGCCTGGAGA CTTCCAGGTCTTGCAGTTTACC TGCACTGAAGTTTCTCGGC CCAACCCTCTAGTCGTCCAC ATGCACTATTGGTAAACAGGCA AAGTCTCAGAGACTGACGGACC GACCAAATGAAATGCGCC TATTGAGCTCACAAAAACAAAGG TTCCTGAGAGCTCTTCCTGC AGCGGGTCTGTGGCTCAT GAGTGCGGAGAGCTCTGC TACTGTGGACACGGACAGGA CCCAGCTGTGCTCATAACATAC TCAACAACCCCTCCGTAGAC GGGATGTGCAAAGAAATCTACC TGCAAAAATAGAGCCAATACCC CCTTGTGTTTGCTGCAGATG CCAGCCACAAAGGTACTAAGG GTTGGAAACAGGACATGTTAAGC GTGGGAAGGTGGGTTGTCAG GTCTAAAATGAAGTGAATCTGTAAGACAG GTTGGAGAGTCTAAGCACTGAATCAG GCTTAGTTGTTCAGAGGACGGTAAG 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 or Fridolfsson & Ellegren (1999), and Z and W chromosome-specific fragments were separated by electrophoresis on 2% agarose gel (the first of the two protocols also required restriction by the VspI enzyme, which has two recognition sites in W-specific fragments. but only one in the Z-specific fragments; Väli, 2004). DATA ANALYSIS Individuals were assigned to groups using Bayesian model-based Markov chain Monte Carlo (MCMC)simulation approaches in STRUCTURE 2.2 (Pritchard, Stephens & Donnelly, 2000) and NewHybrids 1.1 (Anderson & Thompson, 2002). In STRUCTURE 2.2, the number of assumed groups K was assumed to be two, as two interbreeding species were being studied. The analysis was run for 100 000 burn-in and 500 000 analytical iterations. For each individual, the probability q of belonging to one of the identified populations was calculated using correlated allele frequency models, assuming gene flow between species. Comparatively, all analyses, with no significant differences in results, were conducted using the independent allele frequency model. As the occurrence of two species was presumed, an admixture model without preliminary information on species ancestry was used in order to avoid any potential influence on results. In order to assign an individual to a group, we used the error rate 0.1 from the expected value, which resulted in the following ranges of q (indicated as a probability of recent A. clanga ancestry): A. pomarina, q = 0–0.1; backcross F1 ¥ A. pomarina, q = 0.15–0.35; F1, q = 0.4–0.6; backcross F1 ¥ A. clanga, q = 0.65–0.85; A. clanga, q = 0.9–1.0. Individuals with probabilities that were intermediate © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 730 Ü. VÄLI ET AL. between the two groups were treated as intermediate according to this analysis. Structure was also used for estimating allele frequencies in the two species in order to present the assignment power of and gene flow in various markers. NewHybrids 1.1 was developed specially for identifying hybrids between two species, and it aims to assign each individual into one of the categories: pure species, F1 or later generation hybrids, or backcrosses. Power analysis by synthetic genotypes shows that our marker sets may have problems with identifying F2 hybrids (Väli et al., 2010), and there is a tendency to assign false positives in this class (E.C. Anderson, pers. comm.). We therefore included only five potential classes in the analyses: A. clanga, A. pomarina, F1 hybrids, A. clanga ¥ F1, and A. pomarina ¥ F1. Individuals with all probabilities below 0.7 also showed, in most cases, a relatively high value in terms of belonging to another group, and were thus assigned as intermediate between these groups. However, we also performed simulations with a sixth, F2 class (data not shown), and in cases when subsequent probabilities for F2 were high (qn > 0.8), and in which the analysis of the five classes gave an unclear result (qn < 0.7 in all groups) in this analysis, we assigned individuals to the F2 class. Each run lasted for at least 25 000 sweeps during burn-in and 100 000 sweeps during the analysis stage. Simulations were performed numerous times using both uniform and Jeffreyslike priors. Except for a few intermediate individuals, no great differences were found between the different runs and approaches. Finally, each individual was assigned to a species or to a hybrid group, taking all assignment results into account. Assignments of a nestling and adults (if available) were combined, and the most plausible decision was made for each breeding pair. To depict individual assignments, genotypes of all birds were transformed into values in a multidimensional space by the Factorial Correspondence Analysis (using GENETIX 4.05) (Belkhir et al., 2004). We also estimated genetic differentiation between species by calculating Fst values between A. clanga and A. pomarina groups (hybrids and backcrosses excluded) using GENALEX 6.1 (Peakall & Smouse, 2006). ANALYSIS OF RESULTS THE EUROPEAN POPULATIONS Most of the 738 individuals genotyped by SNPs and microsatellites could be assigned with confidence to one of the five presumed classes – A. clanga, A. pomarina, F1 hybrids between the two species, A. clanga ¥ F1, or A. pomarina ¥ F1 – by both Bayesian methods (Table 1). Results from two approaches usually coincided completely, and in a few cases high probability from one method helped to clarify doubts raised by moderate probability from another method. Only seven samples were identified as intermediate by both approaches (Table 1). Even the less powerful factorial correspondence analysis (FCA) separated the two species clearly, whereas previously assigned hybrids and backcrosses appeared between the two species in an FCA plot (Fig. 2). Differentiation between the two species was evident in microsatellites (Fst = 0.12), SNPs (Fst = 0.44) and in both marker types combined (Fst = 0.20). Hybridization between A. clanga and A. pomarina was detected in six populations across the sympatric Figure 2. Distribution of the 738 spotted eagle individuals studied on the two-dimensional factorial correspondence analysis (FCA) plot. Prior to plotting, birds were classified according to the consensus of genotype-based Bayesian assignments. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 HYBRIDIZATION OF SPOTTED EAGLES IN EUROPE area (including Germany and south-eastern Poland, regions previously considered as being outside the A. clanga distribution range). A combination of field data, molecular sexing of adults, and mtDNA analysis indicated that A. clanga ¥ A. pomarina pairs were made up of male A. pomarina and female A. clanga in 17 cases, whereas the reverse situation occurred in four cases. In seven populations we found mixed pairs of F1 backcrossing to A. pomarina, the more common species in the western part of the sympatric area, whereas introgression (backcrossing) to A. clanga was detected only at the eastern limits of the distribution of A. pomarina in Russia and in southern Belarus, both of which hold a relatively large population of A. clanga. Altogether, 8.6% of the breeding pairs studied were composed of two pure A. clanga individuals, with hybridizing A. clanga ¥ A. pomarina pairs making up 4.9%, and second-generation backcrossing pairs (F1 ¥ pure species) making up 4.4%. Further introgression beyond the second generation apparently also occurs, as we discovered A. clanga mtDNA in four adults and four nestlings (from eight different pairs) that were assigned by nuclear markers to A. pomarina. Backcrossing was also suggested by similar probabilities for the assignment of seven individuals to two classes (pure species and backcrosses; Table 1). Some of these uncertainties may arise from the inadequate power of the analysis, but on at least one occasion we confirmed the nestling’s mother, and on a further occasion the father, as a backcross F1 ¥ A. pomarina; in another case one parent was an F1 ¥ A. clanga. Two nestlings were assigned to the F2 group because the probability of belonging to that class was found by NewHybrids 1.1 (with six classes) to be exceptionally high (0.89 and 731 0.90), whereas analysis with only five classes gave inadequate probabilities for all classes. Moreover, STRUCTURE 2.2 gave probabilities close to 0.50, as can be expected for F2 individuals, but not for backcrosses. MONITORING OF THE ESTONIAN POPULATION Most of the individuals studied in the Estonian population were assigned rather unequivocally to one of the classes (Appendix; Table 1). Ten out of 11 A. clanga ¥ A. pomarina pairs were made up of male A. pomarina and female A. clanga, whereas in one case the reverse was observed. We detected backcrossing to A. pomarina, but not to A. clanga. Additionally, one adult was identified as a backcross between F1 and A. pomarina in four pairs, and A. clanga mtDNA was found in two nestlings (from two pairs) that were assigned by nuclear markers to A. pomarina. Up to 2003, the intensive survey continuously revealed previously unknown A. clanga breeding territories (one or two territories per year), which obscured the possible changes in A. clanga population. After that, we noted a slight decrease in the numbers of A. clanga pairs and hybridizing pairs, although not in backcrossing pairs, which do not show a clear trend (Fig. 3). The small size of the sample, however, does not permit broader conclusions to be made. When we analysed separately the occupation history of each territory where an A. clanga, or an F1 hybrid, had been recorded, the prevalent abandonment of territories, or replacement of rare A. clanga with the common A. pomarina, was noticeable (Table 3). In contrast, the opposite development was rare – we found an adult A. pomarina replaced by an Figure 3. Number of breeding territories occupied by Aquila clanga and hybrids in Estonia in 2003–09. Genetic data are complemented with field observations. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 732 Ü. VÄLI ET AL. Table 3. Occupation history of spotted eagle breeding territories in Estonia, where at least one parent was an Aquila clanga or an F1 hybrid. Genetic data are complemented with field observations No. of territory Study years Assignment of breeding pairs to a group 1 2 3 4 5 2003–2009 2003–2009 2002–2009 1994–2009 1998–2009 A. A. A. A. A. 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 1997–2009 2004–2009 1994–2009 2001–2009 2004–2009 2001–2009 2000–2008 2002–2007 1999–2007 2006–2008 2002–2007 1999–2006 2004–2008 1998–2001 1999–2009 1999–2007 cla pair cla pair → single A. cla → A. pom pair cla pair → A. pom pair cla pair → single A. cla → abandoned since 2006 cla (pair?) → A. cla ¥ A. pom → F1 ¥ A. pom → A. pom pair → abandoned since 2008 A. cla ¥ A. pom A. cla ¥ A. pom A. cla ¥ A. pom → A. cla pair → A. cla ¥ A. pom → F1 ¥ A. pom A. cla ¥ A. pom → F1 ¥ A. pom → F1 ¥ F1 → F1 ¥ A. pom A. cla ¥ A. pom → A. pom pair A. cla ¥ A. pom → A. pom pair A. cla ¥ A. pom → A. pom pair A. cla ¥ A. pom → abandoned since 2004 A. cla ¥ A. pom → abandoned since 2003 F1 ¥ A. pom → A. pom pair F1 ¥ A. pom → A. pom pair F1 ¥ A. pom → A. pom pair F1 ¥ A. pom → A. pom pair F1 ¥ A. pom → A. pom pair A. pom pair → A. cla ¥ A. pom A. pom pair → F1 ¥ A. pom A. clanga or a hybrid on only two occasions, and once by a backcross F1 ¥ A. pomarina. In one other case a hybridizing pair was replaced by an A. clanga pair, but this changed back over a 7-year period, with ultimately the substitution of an F1 ¥ A. pomarina pair. By 2009, four A. clanga territories were abandoned, and now the species is only found in four breeding territories (interbreeding with A. pomarina in three of them). Three territories were occupied by an adult hybrid interbreeding with A. pomarina, and altogether ten previous A. clanga or hybrid territories were occupied by A. pomarina pairs. DISCUSSION EXTENT AND FREQUENCY OF HYBRIDIZATION The current study confirmed extensive introgressive hybridization between the two spotted eagle species in Europe, and highlights the significance of hybridization as an additional threat factor to the globally vulnerable A. clanga. Hybridization occurs over a very large area. Two interbreeding pairs of A. clanga and A. pomarina were discovered in Germany, whereas in central European Russia breeding of an A. pomarina pair and two backcrossing pairs were detected (Table 1). Given that only nine pairs were sampled in this Russian population, we believe that interbreeding also takes place here, and that the backcrosses detected do not just reflect gene flow from west to east. This suggests a hybrid zone of at least 1700 km in width. Even when only the extensively hybridizing population in the west (north-eastern Poland) is taken into account (excluding well-studied Germany, where only two cases are known), the zone width is at least 1100 km, which is much wider than previously recorded hybrid zones in birds (Price, 2008: 326–328). We undoubtedly underestimated the proportion of A. pomarina, as sampling in most populations was not random but was biased towards rarer species and hybrids. Comparisons made among these latter groups are more reliable, but are still likely to be somewhat biased, as most of the hybridizing populations studied originated from the western limit of sympatry, where A. clanga is rare. However, the fact that hybridizing A. clanga ¥ A. pomarina pairs and backcrossing pairs were together more numerous than A. clanga pairs (Table 1) nevertheless calls for vigilance. In the case of the more closely monitored Estonian population, the proportion of interbreeding pairs was almost equal to the proportion of A. clanga pairs when annual average estimates were considered, and was even twice as high when all breeding territories detected during the study period were con- © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 HYBRIDIZATION OF SPOTTED EAGLES IN EUROPE sidered together (Appendix; Fig. 3). This probably does not indicate a lack of selection for conspecifics in spotted eagles; rather it shows the shortage of A. clanga individuals and the fragility of the reproductive barrier separating the species. On the other hand, the difference between these estimates may indicate a higher exchange of individuals in hybridizing pairs. ASYMMETRY OF HYBRIDIZATION AND INTROGRESSION We established sexually asymmetrical hybridization, as in 81% of cases a male A. pomarina mated with a female A. clanga. Field data from Belarus suggest a more even sex ratio in hybridizing pairs (V. Dombrovski, unpubl. data), but the observed asymmetry is noteworthy, even if it is limited to the northwestern regions where we collected most of our samples. It can be accounted for in a number of ways. The sex bias could occur because females from a rare species tend to interbreed with males from common species, but not vice versa, because of the selectiveness of females (Wirtz, 1999). However, there is no clear evidence to support the validity of this theory in birds (Randler, 2002), and in some species males are more selective than females (Saether, Fiske & Kalas, 2001). It has also been suggested that the reason for asymmetry could be the reversed sexual dimorphism in raptors: males are smaller than females, leading to an advantage for the males of the smaller species, and the females of the larger species, thus maximising the size factor (Helbig et al., 2005b). Indeed, there is a fitness-related advantage of large body size in A. pomarina females (Lõhmus & Väli, 2004), and this may act as one of the triggers in hybridization because A. pomarina males prefer and compete for A. clanga females that are larger than A. pomarina females. Finally, the hybridization mechanism could simply be based on sexual differences in territory selection pattern. In many animals, including raptors, males select a territory whereas females select their male partner (Wirtz, 1999). Aquila clanga males can fail to establish a breeding territory because of a loss of suitable habitat, and therefore often remain vagrant, whereas A. clanga females are prepared to select A. pomarina males occupying a territory. The crucial question in the conservation of hybridizing species is the fertility of hybrids and introgression of genes from one species to another (Allendorf et al., 2001). Our study shows that hybrids of spotted eagles are fertile, as one could expect from their relatively recent divergence (Price & Bouvier, 2002). This applies to both males and females, and means that Haldane’s rule (Haldane, 1922) is not fully fol- 733 lowed. Earlier studies have reported that unidirectional introgression occurs, as A. clanga mtDNA has been found in A. pomarina or in intermediate birds, but not vice versa (Väli & Lõhmus, 2004; Helbig et al., 2005b). However, as hybridization is asymmetrical between female A. clanga and male A. pomarina, backcrossing to A. clanga would remain unnoticed using mtDNA analysis, as there is no difference between mtDNA molecules originating from hybrids or from pure species. Here, we also confirmed the same pattern using the more informative nuclear markers. There may be several explanations for an asymmetry towards A. pomarina. Backcrossing to the more common species is simply more probable, as hybrids have a lesser probability of meeting and mating with other hybrids, or with the rare species, than with individuals of the common species. Alternatively, Helbig et al. (2005b) suggested that asymmetrical introgression in sexually dimorphic spotted eagles could also result from size-oriented sexual selection, as hybrid females would mate with A. pomarina males, which are smaller than themselves, but not with A. clanga males that are larger or of equal size. FUTURE PROSPECTS The main factor for the separation of the two species in the breeding grounds is probably the difference in habitat selection. Aquila clanga prefers the vicinity of wetlands, whereas A. pomarina occupies less moist habitats (Cramp & Simmons, 1980). However, habitat use is rather flexible as at least some A. clanga are able to inhabit drained habitats less typical for the species (Dombrovski & Ivanovski, 2005), and these biotopes could also be occupied by A. pomarina (Lõhmus & Väli, 2005). Habitat nevertheless remains the key factor, as its loss leads to a decline in numbers. In essence, the smaller the numbers of A. clanga, the higher the probability of hybridization, as rarity is the most important factor triggering interbreeding (Hubbs, 1955; Short, 1969; Wirtz, 1999; Randler, 2006). As A. clanga numbers have been declining during the last century (Hagemeijer & Blair, 1997), a simultaneous increase of hybridization can be expected. Indeed, Dombrovski (unpubl. data) has studied the morphology of museum specimens, and reported the increase of hybrids during the second half of the 20th century. The accumulation of hybridization records in recent years, mentioned above, probably also highlights the increased frequency of interbreeding (but it must also be attributed in part to increased awareness). There are still A. clanga populations in Eastern Europe, such as those in Belarus and central European Russia, where the species is not outnumbered by hybridizing pairs © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 734 Ü. VÄLI ET AL. (see also Dombrovski, 2005 for field data). Hybridization pressure on these populations is probably growing as the numbers of A. clanga decline. The analysis of the Estonian population indicated the decline of A. clanga, but did not directly confirm a recent increase in hybridization. However, if the history of each breeding pair involving an A. clanga or a hybrid (Table 3) is studied, it clearly shows that the decline of A. clanga in Estonia is concomitant with extensive hybridization with A. pomarina, and hybridization is often an intermediate step on the way to the disappearance of A. clanga. Such a rapid decline and changing population structure of a longlived bird is alarming. The Estonian A. clanga population has few males, and there has been hardly any replacement among females (Ü. Väli, unpubl. data). Following the disappearance of an A. clanga individual, the territory is either abandoned or occupied by A. pomarina. Hybridization has been a significant threat for several species and subspecies (Rhymer & Simberloff, 1996). It could therefore be an important factor contributing to the decline, and potentially even the extinction, of A. clanga populations in Europe. This should be borne in mind when proposing measures for A. clanga conservation. ACKNOWLEDGEMENTS We thank Sven Aun, Igor Babkin, Viktor Belik, Marina Dzmitranok, Raivo Endrekson, Tarmo Evestus, Mikhail Ivanov, Roman Kiselev, Kristo Lauk, Asko Lõhmus, Anton Makarov, Riho Männik, Joachim Matthes, Vladimir Melnikov, Gennady Mindlin, Renno Nellis, Rein Nellis, Kostas Poirazidis, Svetlana Romanova, Vitaliy Ryabtsev, Pauli Saag, Lauri Saks, Wolfgang Scheller, Gunnar Sein, Urmas Sellis, Indrek Tammekänd, Jaak Tammekänd, Aarne Tuule, Eet Tuule, Joosep Tuvi, Dmitriy Zhuravlev, and Nikolay Yakovets for their help with sample collection. Frank Hailer kindly provided primers for the microsatellite analysis. Comments of four anonymous reviewers greatly improved the manuscript. Financial support was provided by APB-BirdLife Belarus, the Estonian State Nature Conservation Centre, the Estonian Science Foundation (grant nos ETF6050 and ETF7593), the Frankfurt Zoological Society, the Ornithological Society of the Middle East, the Royal Society for the Protection of Birds, and the Marie Curie Intra-European Fellowship for ÜV within the 6th European Community Framework programme. REFERENCES Aliabadian M, Nijman V. 2007. Avian hybrids: incidence and geographic distribution of hybridization in birds. Contributions to Zoology 76: 59–61. Aljanabi SM, Martinez I. 1997. Universal and rapid saltextraction of high quality genomic DNA for PCR-based techniques. Nucleic Acids Research 25: 4692–4693. Allendorf FW, Leary RF, Spruell P, Wenburg JK. 2001. The problems with hybrids: setting conservation guidelines. Trends in Ecology & Evolution 16: 613–622. Anderson EC, Thompson EA. 2002. A model-based method for identifying species hybrids using multilocus genetic data. Genetics 160: 1217–1229. Arnold ML. 1997. Natural hybridization and evolution. Oxford, New York: Oxford University Press. Belkhir K, Borsa P, Chikhi L, Raufaste N, Bonhomme F. 2004. GENETIX 4.05, logiciel sous Windows TM pour la génétique des populations. Laboratoire Génome, Populations, Interactions, CNRS UMR 5000. Montpellier, France: Université de Montpellier II. Benedict RA. 1999. Morphological and mitochondrial DNA variation in a hybrid zone between short-tailed shrews (Blarina) in Nebraska. Journal of Mammalogy 80: 112– 134. Bergmanis U. 1996. On the taxonomy of the Lesser Spotted Eagle Aquila pomarina and Greater Spotted Eagle A. clanga. In: Meyburg BU, Chancellor RD, eds. Eagle studies. Berlin, London & Paris: World Working Group of Birds of Prey, 199–207. Bergmanis U, Strazds M. 2001. Another possible hybridization case of Lesser and Greater Spotted eagle in Latvia. Putni Daba 11.2: 6–7. Bergmanis U, Petrinš A, Strazds M, Krams I. 1997. Possible case of hybridization of the Lesser Spotted eagle and the Greater Spotted eagle in Eastern Latvia. Putni Daba 3: 2–6. BirdLife International. 2004. Birds in Europe: Population estimates, trends and conservation status. Cambridge, UK: BirdLife International. Busch JD, Katzner TE, Bragin E, Keim P. 2005. Tetranucleotide microsatellites for Aquila and Haliaeetus eagles. Molecular Ecology Notes 5: 39–41. Cramp S, Simmons KEL. 1980. The birds of the Western Palearctic. Oxford: Oxford University Press. Dombrovski VC. 2002. Hybridization of Lesser and Greater Spotted Eagles (Aquila pomarina et A. clanga) in Belarus: rule or exception? Subbuteo 5: 23–31. Dombrovski VC. 2005. Hybridation entre Aigles criard Aquila clanga et pomarin A. pomarina en Biélorussie: conséquence taxonomique. Nos Oiseaux 52: 27–30. Dombrovski VC. 2006. Morphological characteristics and diagnostic features of the Greater Spotted (Aquila clanga), Lesser Spotted (A. pomarina) eagles, and their hybrids. Ornithologia 33: 29–41. Dombrovski VC, Ivanovski VV. 2005. Numbers, distribution and breeding ecology of the Greater spotted Eagle (Aquila clanga). Belarus. Ornithologia 32: 57–70. Eastham CP, Nicholls MK. 2005. Morphometric analysis of large Falco species and their hybrids with implications for conservation. Journal of Raptor Research 39: 386–393. Feldt. 1909. Ueber Raubvögel, Vogelschutz, Naturbeobachtung. Neue Baltische Waidmannsblätter 1909: 266–270. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 HYBRIDIZATION OF SPOTTED EAGLES IN EUROPE Forsman D. 1999. The raptors of Europe and the Middle East: a handbook of field identification. Poyser: London. Fridolfsson AK, Ellegren H. 1999. A simple and universal method for molecular sexing of non-ratite birds. Journal of Avian Biology 30: 116–121. Gaubert P, Taylor PJ, Fernandes CA, Bruford MW, Veron G. 2005. Patterns of cryptic hybridization revealed using an integrative approach: a case study on genets (Carnivora, Viverridae, Genetta spp.) from the southern African subregion. Biological Journal of the Linnean Society 86: 11–33. Grant PR, Grant R. 1992. Hybridization of bird species. Science 256: 193–197. Griffiths R, Double MC, Orr K, Dawson RJG. 1998. A DNA test to sex most birds. Molecular Ecology 7: 1071– 1075. Gutiérrez R, Villa S. 2002. A possible hybrid Spotted Eagle ¥ Lesser Spotted Eagle in Spain. Birding World 15: 104–105. Hagemeijer EJM, Blair MJ. 1997. The EBCC atlas of European breeding birds. London: Poyser. Hailer F, Gautschi B, Helander B. 2005. Development and multiplex PCR amplification of novel microsatellite markers in the White-tailed Sea Eagle, Haliaeetus albicilla (Aves: Falconiformes, Accipitridae). Molecular Ecology Notes 5: 938–940. Haldane JBS. 1922. Sex-ratio and unisexual sterility in hybrid animals. Journal of Genetics 12: 101–109. Helbig AJ, Kocum A, Seibold I, Braun MJ. 2005a. A multi-gene phylogeny of aquiline eagles (Aves : Accipitriformes) reveals extensive paraphyly at the genus level. Molecular Phylogenetics and Evolution 35: 147–164. Helbig AJ, Seibold I, Kocum A, Liebers D, Irwin J, Bergmanis U, Meyburg BU, Scheller W, Stubbe M, Bensch S. 2005b. Genetic differentiation and hybridization between Greater and Lesser Spotted Eagles (Accipitriformes: Aquila clanga, A. pomarina). Journal of Ornithology 146: 226–234. Horvath MB, Martinez-Cruz B, Negro JJ, Kalmar L, Godoy JA. 2005. An overlooked DNA source for noninvasive genetic analysis in birds. Journal of Avian Biology 36: 84–88. Hubbs CL. 1955. Hybridization between fish species in nature. Systematic Zoology 4: 1–20. IUCN. 2009. IUCN Red List of Threatened Species. Version 2009.2. Available at http://www.iucnredlist.org. Downloaded on 6 November 2009. Lerner HRL, Mindell DP. 2005. Phylogeny of eagles, Old World vultures, and other Accipitridae based on nuclear and mitochondrial DNA. Molecular Phylogenetics and Evolution 37: 327–346. Lõhmus A. 1998. Numbers of the Greater and the Lesser Spotted Eagle in Estonia. Hirundo 11: 24–34. Lõhmus A, Väli Ü. 2001. Interbreeding of the Greater Aquila clanga and Lesser Spotted Eagle A. pomarina. Acta Ornithoecologica 4: 377–384. Lõhmus A, Väli Ü. 2004. The effects of habitat quality and female size on the productivity of the Lesser Spotted Eagle 735 Aquila pomarina in the light of the alternative prey hypothesis. Journal of Avian Biology 35: 455–464. Lõhmus A, Väli Ü. 2005. Habitat use by the vulnerable Greater Spotted Eagle Aquila clanga, interbreeding with the Lesser Spotted Eagle Aquila pomarina in Estonia. Oryx 39: 170–177. Löwis O. 1888. Die baltischen Raubvögel. Baltische Monatsschrift 35: 527–556, 689–717. Löwis O. 1898. Diebe und Räuber in der Baltischen Vogelwelt. Riga: J. Deubner. Martinez-Cruz B, David VA, Godoy JA, Negro JJ, O’Brien SJ, Johnson WE. 2002. Eighteen polymorphic microsatellite markers for the highly endangered Spanish Imperial Eagle (Aquila adalberti) and related species. Molecular Ecology Notes 2: 323–326. McCarthy EM. 2006. Handbook of avian hybrids of the world. Oxford, New York: Oxford University Press. Melnikov VN, Solovkov DA, Kostin BA, Egorova NA, Bogomolov DV. 2001. Nesting of Lesser Spotted Eagle (Aquila pomarina) in Ivanovo region (Russia). Acta Ornithoecologica 4: 287–289. Meyburg B-U, Mizera T, Matthes J, Graszynski K, Schwanbeck JP, Maciorowski G. 2005. Hybridization of Greater Aquila clanga and Lesser Spotted Eagle A. pomarina in Poland and Germany. In: Mizera T, Meyburg BU, eds. International Meeting on Spotted Eagles (Aquila clanga, A. Pomarina and A. Hastata) – Research and Conservation. Osowiec: Biebrza National Park, 115–117. Meyburg BU, Meyburg C, Franck-Neumann F. 2007. Why do female Lesser Spotted Eagles (Aquila pomarina) visit strange nests remote from their own? Journal of Ornithology 148: 157–166. Mischenko A, Väli Ü, Treinys R, Bergmanis U, Domaschevski S. 2010. The Lesser Spotted Eagle Aquila pomarina C.L. Brehm, 1831. In: Galushin VM, Belik VP, Melnikov VN, Bragin EA, Numerov AD, eds. Birds of Russia and surrounding regions: Falconiformes. In press. Newton I. 2003. The speciation and biogeography of birds. London: Academic Press. Nittinger F, Gamauf A, Pinsker W, Wink M, Haring E. 2007. Phylogeography and population structure of the saker falcon (Falco cherrug) and the influence of hybridization: mitochondrial and microsatellite data. Molecular Ecology 16: 1497–1517. Panov EN. 1989. Hybridization and ecological isolation in birds. Nauka: Moscow. Peakall R, Smouse PE. 2006. GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6: 288–295. Price T. 2008. Speciation in birds. Greenwood Village, CO: Roberts and Company. Price TD, Bouvier MM. 2002. The evolution of F-1 postzygotic incompatibilities in birds. Evolution 56: 2083–2089. Pritchard JK, Stephens M, Donnelly P. 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945–959. Randler C. 2002. Avian hybridization, mixed pairing and female choice. Animal Behaviour 63: 103–119. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 736 Ü. VÄLI ET AL. Randler C. 2004. Frequency of bird hybrids: does detectability make all the difference? Journal of Ornithology 145: 123–128. Randler C. 2006. Behavioural and ecological correlates of natural hybridization in birds. Ibis 148: 459–467. Rhymer JM, Simberloff D. 1996. Extinction by hybridization and introgression. Annual Review of Ecology and Systematics 27: 83–109. Saether SA, Fiske P, Kalas JA. 2001. Male mate choice, sexual conflict and strategic allocation of copulations in a lekking bird. Proceedings of the Royal Society of London Series B-Biological Sciences 268: 2097–2102. Seibold I, Helbig AJ, Meyburg BU, Negro JJ, Wink M. 1996. Genetic differentiation and molecular phylogeny of European Aquila eagles according to cytochrome b nucleotide sequences. In: Meyburg B-U, Chancellor RD, eds. Eagle studies. Berlin, London, Paris: World Working Group of Birds of Prey, 1–15. Short LL. 1969. Taxonomic aspects in avian hybridization. Auk 86: 84–105. Transehe N. 1942. Das Vogelleben des Angerschen Sees. Korrespondenzblatt des Naturforscher-Vereins zu Riga 64: 101–124. Transehe Nv. 1965. Die vogelwelt lettlands. HannoverDöhren: Harro von Hirschheydt. Treinys R. 2005. The Greater Spotted Eagle Aquila clanga: previous, current status and hybridization in Lithuania. Acta Zoologica Lituanica 15: 31–38. Väli Ü. 2002. Mitochondrial pseudo-control region in old world eagles (genus Aquila). Molecular Ecology 11: 2189– 2194. Väli Ü. 2004. Sex ratio of Lesser Spotted Eagle Aquila pomarina nestlings in good and poor breeding years. Bird Study 51: 189–191. Väli Ü, Lõhmus A. 2004. Nestling characteristics and identification of the Lesser Spotted Eagle Aquila pomarina, Greater Spotted Eagle A. clanga, and their hybrids. Journal of Ornithology 145: 256–263. Väli Ü, Saag P, Dombrovski V, Meyburg B-U, Maciorowski G, Mizera T, Treinys R, Fagerberg S. 2010. Microsatellites and single nucleotide polymorphisms in avian hybrid identification: a comparative case study. Journal of Avian Biology 41: 34–49. Wirtz P. 1999. Mother species–father species: unidirectional hybridization in animals with female choice. Animal Behaviour 58: 1–12. APPENDIX Number of breeding pairs of spotted eagles studied with genetic markers in 1994–2009 in Estonia, assigned to different groups. The last row includes all territories registered, not the sum of annual results. As a result of the changes in occupation history (Table 3), some territories are included in several groups. A. cla pair N 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Total no. of studied pairs A. cla ¥ A. pom F1 ¥ F1 F1 ¥ A. pom F1 ¥ A. pom/ A. pom A. pom pair 3 1 1 5 6 31 27 30 53 12 17 34 38 28 20 28 1 1 1 2 0 2 2 2 3 4 1 2 3 1 1 1 1 0 0 0 0 3 2 3 4 2 4 1 2 1 2 3 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 1 0 1 0 1 2 2 1 2 2 0 0 0 0 0 1 0 1 1 0 0 1 0 0 0 0 1 0 0 3 6 24 22 24 44 6 11 28 31 24 15 22 173 6 11 1 7 4 144 © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 725–736 View publication stats