The breeding distribution of Eurasian Spoonbills Platalea leucorodia along the East Atlantic seaboards of northwest Africa and Europe is rather disjunct (Cramp & Simmons 1977). There is a cluster of colonies in northwest Europe centred in The Netherlands, there are breeding colonies 2000 km further south in southern Spain, and, again separated by a 2000 km gap, a concentration of colonies on Banc d'Arguin, Mauritania, on the very verge of the Western Palearctic (Cramp & Simmons 1977). The Spanish and Dutch breeding birds have been assigned firmly to the subspecies leucorodia, while birds breeding on the offshore islands of Banc d'Arguin on the basis of their distinct morphology received recognition as a separate subspecies balsaci by de Naurois and Roux (1974). Mauritania-breeding birds are somewhat smaller, have less pronounced yellow breast feathers in breeding plumage (de Naurois & Roux 1974, Cramp & Simmons 1977), and adults, but not immatures, have a uniformly black upper bill (O. Overdijk, unpubl. data).

Perhaps surprisingly, the proposal for this subspecies distinction has yet to be revisited in the ornithological literature. Here we examine whether the variation in microsatellites that were initially developed for Black-faced Spoonbills P. minor (Yeung et al. 2009) and Roseate Spoonbills P. ajaja (Sawyer & Benjamin 2006) is greater between Spoonbills breeding in Mauritania and The Netherlands than among Spoonbills from a range of breeding islands within the Dutch Wadden Sea. We will reflect on dispersal patterns documented for individually colour-marked birds to interpret the molecular findings.

METHODS

The long-term colour-marking project of Dutch Spoonbills (a population close to 4000 breeding pairs) that underlies the present study, is introduced in Lok et al. (2009, 2011). Blood samples were collected from leucorodia breeding colonies at four different Dutch Wadden Sea islands: Schiermonnikoog (n = 26), Terschelling (n = 21), Vlieland (n = 40) and Wieringen (n = 18) in June 2011. Blood samples of the balsaci subspecies (n = 25), now consisting of under 1000 breeding pairs (O. Overdijk & E.M. El-Hacen, unpubl. data), were obtained at one of the four regular colonies on Banc d'Arguin, at Toufat in October 2007. In all cases we captured chicks before fledging in or near the breeding colonies, and this means that we cannot exclude the possibility that we occasionally sampled relatives. Blood was stored in 95% ethanol, and DNA was extracted from these blood samples using a salt extraction method (Richardson et al. 2001). As individual breeders may shift between colonies (O. Overdijk & E.M. El-Hacen, unpubl. data), the Toufat samples fairly represent the balsaci population.

Table 1.

Polymorphic microsatellites for Eurasian Spoonbills. Ta = annealing temperature, N_A = number of alleles, and H_o and H_E = Observed and Expected Heterozygosity. Cone, primer = primer concentration in 10 µl volume PCR reaction (n = 130).

To study genetic variation among the spoonbill samples, six primer sets from P. ajaja (Sawyer & Benjamin 2006) and 15 primer sets from P. minor (Yeung et al. 2009) were tested for amplification and polymorphism in eight samples (four leucorodia samples from Schiermonnikoog and four balsaci from Banc d'Arguin). Of these 21 microsatellites, 12 were found to amplify a PCR product and to be polymorphic (Table 1). Subsequently, all 130 DNA samples were analysed for these 12 microsatellite loci.

Microsatellites were amplified in multiplex PCRs of 3 separate panels (Table 1). PCR reactions were carried out in 10 µl volume containing 20–50 ng DNA, 0.2 mM of each dNTP, 10 mM Tris-HCl, 50 mM KCl, 3.0 mM MgCl₂, 0.25 U Taq DNA polymerase (Roche Diagnostics GmbH, Mannheim, Germany), and a variable amount of each primer. See Table 1 for the amount of each primer in the PCR reaction. PCR program was as follows: 94°C for 1 min, 35 cycles of 94°C for 30 s, 60°C for 60 s and 72°C for 60 s, followed by 72°C for 2 min. Fluorescently labelled PCR products were separated on an AB3730 DNA analyser, and allele-lengths were determined using Genemapper 4.0 software.

Population genetic parameters were estimated using Arlequin 3.1 (Excoffier et al. 2005) and FSTAT 2.9.3 (Goudet 2001).

RESULTS

Microsatellites from P. minor and P. ajaja showed moderate to high levels of variation in P. leucorodia. Number of alleles (N_A) per microsatellite locus ranged from 2–40 and the expected heterozygosity (H_E) ranged from 0.23–0.93 (Table 1). Eleven loci showed no significant deviations from HardyWeinberg-equilibrium, the twelfth locus PM2—16 had a significant excess of heterozygotes in birds from the Banc d'Arguin after Bonferroni correction. The mean expected heterozygosity (H_E) was 0.57 for Banc d'Arguin birds and ranged from 0.67–0.69 for the Dutch colonies (Figure 1), even though the difference in H_E among breeding colonies overall was not significant (Kruskal-Wallis ANOVA, P = 0.74). A single locus, PM3—16, was monomorphic in Mauritania: it was fixed for allele 201 that had a frequency of 0.83–0.88 in the Dutch populations. Eleven loci showed (rare) alleles unique to either Mauritania or The Netherlands, but these alleles generally had a low frequency (<0.10).

Figure 1.

Level of genetic variation (mean expected heterozygosity, H_E) within the five sampled populations. Vertical bars denote 0.95 confidence intervals.

A hierarchical Analysis of Molecular Variance (AMOVA) was used to partition the total genetic variation into two components: (1) variation between the source countries Mauritania and The Netherlands (i.e. between subspecies balsaci and leucorodia), and (2) variation within the breeding sites. The AMOVA revealed significant genetic differentiation between the breeding sites Mauritania (n = 25) and The Netherlands (n = 105, F_ST = 0.063, P < 0.00001). The remaining 93.7% of the variation was due to individual variation within breeding sites. In addition, tests of pair-wise genetic differentiation (F_ST) among all possible combinations of breeding colonies (Table 2), only showed significant genetic differentiation between the single Banc d'Arguin and each of the Dutch colonies. Genetic differentiation among the Dutch populations was small and not significant.

Table 2.

Pairwise genetic differentiation (F_ST) between Spoonbill colonies and populations. Significant F_ST values are bold, P ≤ 0.005 after Bonferroni correction.

DISCUSSION

The birds from Mauritania (balsaci) showed less genetic variation than the Dutch birds (leucorodia, Figure 1). Although this might suggest that the Banc d'Arguin population historically has gone through narrower population bottlenecks than the Dutch population, a more likely explanation is methodological: we probably sampled more families (and colonies) in The Netherlands than in Mauritania.

That there is little genetic structure within the leucorodia-spoonbills breeding on the Dutch Wadden Sea Islands is consistent with the considerable dispersal between the colonies on different islands. This can be illustrated by observations in the best studied colony on Schiermonnikoog in 1994–2009 of colour-ringed breeding birds born on islands with yearly colour-ringing of spoonbill chicks (Vlieland, Terschelling and, from 1997 onwards, Schiermonnikoog, see Lok et al. 2009; O. Overdijk, unpubl. data). During this period, 229 colourringed birds were observed breeding on Schiermonnikoog, of which 85 were born on Vlieland or Terschelling (dispersal), and 144 on Schiermonnikoog (philopatry). In only one case, a bird was known to have previously bred in another colony. Although the level of breeding dispersal is probably underestimated due to the lower observation effort in colonies other than Schiermonnikoog, it nevertheless suggests that natal dispersal is more common than breeding dispersal. Figure 2 shows that soon after establishment of the Schiermonnikoog colony in 1992 most breeding birds of 2–13 years old were born on Vlieland and Terschelling, whereas in recent years most recruiting birds were locally born individuals (but note that on Schiermonnikoog about twice as many chicks were colour-ringed each year compared to Vlieland and Terschelling). This suggests that gene flow is very high during the establishment of a colony, but strongly decreases during colony growth when locally born individuals start to recruit. Consequently, genetic differentiation among Dutch colonies may increase in the future.

Figure 2.

Total number of individually colour-marked Spoonbills observed breeding on Schiermonnikoog (first established 1992) per source colony and year of first breeding observation. Note that colour-ringing of chicks on Schiermonnikoog started in 1997.

We observed significant genetic differentiation between balsaci breeding in Mauritania and leucorodia breeding in The Netherlands. This suggests that gene flow between The Netherlands and Mauritania is much lower than the level of gene flow among the Dutch breeding colonies. Assuming that movements between the two populations are now in equilibrium, and an island model of migration (Wright 1951; F_ST = (1/4Nm+1), the level of dispersers between Wadden Sea and Banc d'Arguin (or vice versa) should be about four of five individuals per generation. In fact, there is some equivocal observational evidence for such dispersal, albeit in a single direction only (Table 3). There are several cases of European-born Spoonbills copulating with balsaci-looking local breeders. In 2000 a Netherlands-born leucorodia was observed feeding chicks in a balsaci-colony on the island of Nair. However, the best documented example of dispersal from The Netherlands to Mauritania (but not necessarily subspeciescrossbreeding) is of a colour-ringed male Spoonbill born on Terschelling in 1997 that was breeding on Nair (probably in 2002 and certainly in 2010). When, on 17 May 2010, it was observed on its nest and on 30 June 2010, it was observed feeding chicks, it was paired with a leucorodia-looking bird probably of European origin (Figure 3).

Table 3.

Evidence for natal dispersal by leucorodia Spoonbills into the balsaci breeding population on the Banc d'Arguin, Mauritania, based on notes by O. Overdijk. In addition there are five cases of birds in balsaci breeding colonies showing the symmetric yellow spots on the upper bill characteristic of leucorodia.

Breeding colonies of the subspecies leucorodia are widely distributed across the Eurasian continent (Cramp & Simmons 1977). Therefore, genetic differentiation between geographic populations of leucordia, at least in theory, could be larger than the estimate of divergence between the two subspecies based on Spoonbills breeding in Mauritania and The Netherlands. This question would need additional work, but the question of whether the genetic differentiation found is high enough to support a separation into two subspecies, balsaci and leucorodia, can be answered. As a gauge, we can compare the level of microsatellite differentiation between leucorodia and balsaci observed (F_ST= 0.06) with values found in other microsatellite studies on avian subspecies. A sample of studies on migrant bird species, although generally using slightly smaller sets of microsatellite loci, showed rather comparable levels of differentiation; in order of increasing F_ST: Dunlin Calidris alpina (three subspecies, 7 loci, F_ST ≤ 0.010; Marthinsen et al. 2007), Canada Goose Branta canadensis (two subspecies, 6 loci, F_ST = 0.021; Mylecraine et al. 2008), Bluethroat Luscinia svecica (seven subspecies, 7–11 loci, F_ST = 0.042; Johnsen et al. 2006), Reed Bunting Emberiza schoeniclus (two subspecies, 6 loci, F_ST = 0.043; Kvist et al. 2011), and Piping Plover Charadrius melodus (two subspecies, 8 loci, F_ST = 0.104; Miller et al. 2010). Thus, with F_ST = 0.063, the observed genetic differentiation between the two Spoonbill subspecies seems to fall in the upper range of values. On the basis of this comparison the subspecies status of balsaci seems entirely valid.

Figure 3.

A colour-ringed bird born on Terschelling in 1997 standing on its nest on Nair (Banc d'Arguin) on 17 May 2010. A month later it was observed feeding chicks. Its partner (sitting on the nest) has a yellow bill tip. Given that male spoonbills generally have larger bills than females, the standing bird is probably a male.