A large herring gull with unusual primary patterns was present at Chew Valley Lake in May-July 2008. The bird showed many characteristics of the American herring gull Larus smithsonianus. Many birders went to see the gull hoping they could tick this transatlantic vagrant. Knowing that American herring gulls had distinctive genetic characteristics, I offered to arrange sequencing of the DNA from a feather to confirm identification of the bird. Keith Vinicombe took up the challenge, and jumped over the rails at Herriot’s Pool to collect a preened down feather for me! A recent PhD student of mine, Dr Jon Flanders, obtained clean DNA sequences from the feather with assistance from Dr Steve Rossiter at Queen Mary College, University of London.

With a few modifications to our usual skin tissue extraction techniques, excellent DNA was obtained from the feather. We used published primers (Crochet et al. 2002) as start markers for amplifying tiny quantities of DNA via the polymerase chain reaction. We sequenced portions of two mitochondrial genes called the control region (about 650 base pairs sequenced) and cytochrome b (about 300 base pairs sequenced). We chose these genes because they had been sequenced in studies of ‘large white-headed gulls’ previously, and thus provided valuable reference material with which we could compare our results.

I use DNA sequencing regularly to study and identify bat species (and even more recently to identify insects in bat droppings!). Typically, each bat species has a characteristic gene sequence (termed a haplotype) for each mitochondrial gene. In the large white-headed gulls, the situation is more complex, with each species often containing individuals with different haplotypes. Although some species (including L. smithsonianus) seem to possess unique and distinctive haplotypes carried by all individuals studied to date, other species have multiple haplotypes. The haplotypes are named after the species that contains the majority of that haplotype. Hence the MIC haplotype is found mainly (but not exclusively) in L. michahellis, the yellow-legged gull. For example, a sample of 18 Larus argentatus argentatus, (the subspecies of herring gull found in northern Europe), comprised 6% fuscus, 33% argentatus, 22% marinus, 6% michahellis and 33% haplotypes of a type called ‘70’. The subspecies of herring gull that breeds in Britain, Larus argentatus argenteus showed less haplotype diversity, and comprised 78% argentatus and 22% marinus haplotypes (Crochet et al. 2002).

Why might a herring gull contain mitochondrial DNA more typically found in other species such as the yellow-legged gull or even the great black-backed gull? This mixing is the consequence of hybridization events occurring, often in the past, resulting in a phenomenon known as introgression. Hybridization occurs even today, with populations of Caspian gulls L. cachinnans and L. argentatus producing hybrids routinely where there occur together in Poland (Gay et al. 2007).

What did the gene sequences tell us? First, consider the control region. L. smithsonianus has a distinctive haplotype, termed SMI below. Distinctive parts of the sequence are highlighted in blue below. At two bases. SMI has a c (cytosine) instead of a t (thymine). The Chew bird (CVL) did not have the smithsonianus haplotype. Rather, it has a haplotype called MIC, found in L. michahellis and L. argentatus. So the possibility of L. smithsonianus can be rejected from the results of the control region sequences.

What about the cytochrome b sequence? Although this fragment is shorter, it shows more variability among taxa. The argentatus haplotype for example has 5 bases different from the michahellis haplotype, and smithsonianus is different from both of these. Once again the Chew bird had a haplotype identical with the MIC cytochrome b haplotype, and the possibility of L. smithsonianus could be rejected.

The results show that the bird was definitely not an American herring gull. Interestingly, the American herring gull has a haplotype more similar to other American large white-headed gulls (L. californicus, L. glaucoides, L. thayeri, and L. hyperboreus) than to European herring gulls, again suggesting that hybridization has been important in the evolution of this species. The gene sequences point to a Eurasian bird, but which one? This cannot be answered with absolute certainty.

Crochet et al. (2002) found the MIC cytochrome b haplotype in L. argentatus argentatus but not in L. a. argenteus (although their sample was limited to only 18 birds of each subspecies). This suggests that the bird may be more likely to be argentatus than argenteus, and this fits with its larger size and more bulky appearance. To my mind this represents the most parsimonious explanation.  However, argentatus usually has a darker mantle than argenteus, and this was not apparent. The far eastern Larus vegae also includes individuals with the MIC haplotype, but again this should have a dark mantle and the chance of one of these birds reaching the UK seems excessively remote. In addition, whether vegae represents a distinct species is open to question.

Although pinning the bird down in terms of what it is can be problematic, we can say it definitely is not smithsonianus (or at least its mother was not - mitochondrial DNA is maternally inherited - so it could possibly be a hybrid). So those of you who ticked the Chew bird as an American herring gull, untick it now!

Birding needs bold attempts that might result in the identification of diagnostic features for the identification of difficult taxa such as adult American herring gulls. This could have happened if the identification of the Chew bird as smithsonianus was confirmed. Ultimately, however we’ve been involved in a fascinating detective story that has seen the use of DNA technology to better understand identification in gulls, and conclude that perhaps the identification of adult American herring gulls in the UK still remains as a challenge that may prove impossible to achieve.

References

Crochet, P-A, J-D LeBreton, & F. Bonhomme (2002). Systematics of large white-headed gulls: Patterns of mitochondrial DNA variation in western European taxa. Auk 119:603-620.

Gay L., G. Neubauer, M. Zagalska-Neubauer, C. Debain, J-M. Pons, P. David and P-A. Crochet. (2007). Molecular and morphological patterns of introgression between two large white-headed gull species in a zone of recent secondary contact. Molecular Ecology 16: 3215–3227.