No no no no NO: the Herring gull is NOT a ring species!
So many animals that we totally take for granted are actually, when you think about them, really remarkable. In keeping with the theme of alien invaders (see previous posts on eagle owls and British big cats), I was going to talk about Collared doves Streptopelia decaocto, and the various egret species that are presently ‘invading’ the British Isles. But it’s another ordinary, yet remarkable, bird that I’m going to talk about now, and it’s the Herring gull Larus argentatus, a behaviourally flexible, adaptable and widespread bird that inhabits Eurasia and North America. It’s a large bird that can exceed 70 cm in length and have a wingspan of 1.3 m. It’s also a supreme generalist, capable of thriving on all kinds of food, and this explains its highly successful colonisation of urban environments.
While, as I said, we take urban gulls for granted, if you think about it, it’s pretty amazing that a seabird with a 1.3 m wingspan has successfully colonised towns and cities. They’ve learnt to drop shelled prey from heights in order to break it open (there’s the famous anecdote of a New Jersey highway bridge littered with clam shells dropped by enterprising gulls) and they also like to wash their food in rockpools. Their breeding behaviour and how they and their chicks respond to stimuli is tremendously well studied, with the studies of Goethe, Tinbergen and others being classic, pioneering works in ethology.
But perhaps what makes the Herring gull most ‘famous’ among biologists is it’s alleged status as the examplar par excellence of a ‘ring species’. As we’ll see below, this concept has now been all but debunked, and it’s because of this that I’m surprised whenever I see continuing references to it. Who might be the most naughty of recent offenders, I hear you ask? Richard Dawkins. Yes, he of Selfish Gene fame. I noted a while back that I was reading The Ancestors Tale (Dawkins 2004). It’s a good book, sure, but I’ve found it a tedious read (with awful artwork), and to be honest I’ve repeatedly given up on it and moved on to Mayor’s The First Fossil Hunters, Pianka & Vitt’s Lizards: Windows to the Evolution of Diversity, Patterson’s The Lions of Tsavo and Fisher & Lockley’s Seabirds [image at left, depicting the ring species concept, is from the Inside Science site].
The ring species concept is a big deal as it proposes a mechanism for one of the most important questions of evolutionary biology: the origin of species. And it was study of Herring gulls that led the late great Ernst Mayr (1942) to argue that speciation in Herring gulls had occurred by way of ‘isolation by distance’: while adjacent populations would be able to breed with one another, the genetic distance resulting from the expansion of a species far from its centre of origin would eventually produce an ‘end’ population so far removed from its ancestor that it would be incapable of interbreeding with it. It would now be a separate species. Mayr proposed that exactly this has happened, and that, after originating in the Aralo-Caspian region, Herring gulls had moved north to the Arctic Ocean. Here they expanded west, giving rise to dark-mantled forms of the Lesser black-backed gull, and also east, giving rise to the pale-mantled forms of Herring gull of Siberia and North America. Finally, North American Herring gulls crossed the Atlantic to invade Europe, and here they encountered the Lesser black-backed gulls that marked the other end of the ring. Both end points had now reached reproductive isolation, and today coexist as distinct species.
Don’t get me wrong: there’s good evidence that speciation does occur in this way in some instances (e.g. in Californian Ensatina salamanders, and southern Asian leaf warblers), but it seems to be very rare. And, as it happens, new study indicates that it did not happen in the case of Herring and Lesser black-backed gulls. Firstly, the taxa regarded by Mayr as subspecies of these two are now regarded as distinct enough to be regarded as separate species. The Aralo-Caspian gull regarded by Mayr (1942) as the ancestral Herring gull population is the Caspian or Steppe gull L. cachinnans, and the Mediterranean and eastern Atlantic gull thought by Mayr (1942) to be a westward excursion of the Caspian gull is the Yellow-legged gull L. michahellis. Furthermore, if a recently proposed subdivision of Lesser black-backed gulls is accepted (Sangster et al. 1998), then the proper name for this species is L. graellsii, and two taxa previously ranked as subspecies – the Tundra gull L. heuglini and Baltic gull L. fuscus – should be separated as species. Of course you could bring in here the debate over the whole subspecies concept: there are over 20 of these in the Herring gull-Lesser black-backed gull complex, so.. gack.. how many species should we be recognising? And once we do start to recognise at least some of these taxa as species, doesn’t this negate the whole raison d’être of Mayr’s proposed ring?
Furthermore, in a study of mtDNA in white-headed gulls, Liebers et al. (2004) found that white-headed gull phylogeny and biogeography was far more complex than Mayr and others had thought. Yellow-legged gulls [image at left] were not closest to Caspian gulls, but instead seem to have descended from a North Atlantic ancestral population. A separate ancestral population moved north from the Aralo-Caspian region toward the British Isles, giving rise to the Lesser black-backed gull, and east toward Siberia and North America, where Tundra gulls, Slaty-backed gull L. schistisagus and Glaucous-winged gulls L. glaucescens arose. Intriguingly, the Great black-backed gull L. marinus was not an outgroup to the Herring gull-Lesser black-backed complex as usually thought, but was actually nested within the complex and probably evolved (in allopatry with L. argentatus) in northeastern N. America. Glaucous gulls L. hyperboreus and Kelp gulls L. dominicanus were also nested within L. argentatus, and the discovery about the Kelp gull is interesting: this species is unique to the Southern Hemisphere, and Liebers et al. (2004) concluded that it must have evolved via long-distance colonisation ‘from the same ancestral population as the Lesser black-backed gull, suggesting that its ancestors were highly migratory, as nominate Lesser black-backed gulls still are today’ (p. 895). The central Asian L. mongolicus didn’t originate from Caspian gulls, but from Pacific gulls close to L. schistisagus.
If all of this seems horribly confusing, I think that’s because it is. The number and variety of white-headed gull taxa is baffling and sorting out any kind of historical pattern is highly, highly difficult. The picture is made more complex by the fact that populations which appear to belong to different lineages (e.g. L. michahellis from the Atlantic Iberian coast and western European L. argentatus) look similar, apparently due to convergence (Pons et al. 2004). It’s also difficult to tell whether strong genetic similarities reported between some taxa – such as Baltic gulls and Tundra gulls for example – result from recent separation or from ongoing gene flow (Liebers & Helbig 2002). There’s also the interesting discovery that supposed hybrids (of L. hyperboreus and L. argentatus) turned out to be light-winged L. argentatus founders that were expanding their range (Snell 1991).
But, most importantly, support for the simple ring model is lacking as there is no evidence that North American Herring gulls recolonised Europe to encounter the Lesser black-backed gulls that marked the other end of the ring. However, the great irony of all this is that L. graellsii is presently spreading westwards, and may eventually colonise North America. Should it do this (right now it breeds as far west as Greenland), it will encounter the North American Herring gull L. argentatus smithsonianus, and if these two forms prove incapable of interbreeding, then the ring species model would have been fulfilled… albeit it by birds moving from east to west, rather than west to east as Mayr proposed.
Well, all of that was pretty complicated. Feel free not to remember it, but take home at least the title of Liebers et al.’s paper: ‘The herring gull complex is not a ring species’. In the accompanying photo, James Coyne throws wotsits at a Herring gull on Anglesea. Coyne, where are you now?
PS - for the latest news on Tetrapod Zoology do go here.
Refs - -
Dawkins, R. 2004. The Ancestor’s Tale. Weidenfeld & Nicolson (London), pp. 528.
Liebers, D., de Knijff, P. & Helbig, A. J. 2004. The herring gull complex is not a ring species. Proceedings of the Royal Society of London B 271, 893-901.
- . & Helbig, A. J. 2002. Phylogeography and colonization history of Lesser black-backed gulls (Larus fuscus) as revealed by mtDNA sequences. Journal of Evolutionary Biology 15, 1021-1033.
Mayr, E. 1942. Systematics and the Origin of Species. Columbia University Press (New York), pp. 334.
Pons, J.-M., Crochet, P.-A., Thery, M. & Bermejo, A. 2004. Geographical variation in the yellow-legged gull: introgression or convergence from the herring gull? Journal of Zoological Systematics & Evolutionary Research 42, 245-256.
Sangster, G., Hazevoet, C. J., Berg, A. & van den Roselaar, C. S. 1998. Dutch avifaunal list: species concepts, taxonomic stability, and taxonomic changes in 1998. Dutch Birding 20, 22-32.
Snell, R. R. 1991. Variably plumaged Icelandic Herring gulls reflect founders not hybrids. The Auk 108, 329-341.