What is a ring species? What is a species?

A Greenish Warbler in the hand (c) Paul J Leader

Ring species show speciation in spite of gene flow, and also raise interesting questions about the nature of species. In a paper published in Science entitled 'Speciation by Distance in a Ring Species', Irwin et al. (2005) present the results of a detailed study on the Greenish Warbler Phylloscopus trochiloides, an Asian bird with a very intriguing biogeographic history. The warbler occurs in Asia, from the southern foothills of the Himlayas to the the extensive forests of Siberia to the north. However, the birds carefully (and for good reason) avoid the barren and inhospitable Tibetan plateau - no place for a minute tree-loving warbler (check out the map below). This peculiar scenario drew the attention of Irwin and collaborators, for two reasons. First, the ring-like shape of the species' distribution is an immediate sign that things could be interesting. Second, there are multiple subspecies of Greenish Warbler, two of which co-occur in parts of Siberia. You might think of this zone of overlap as the crucial connection point of the ring. But what is the nature of this link? Have the 'atoms' in the ring fused, resulting in a seamless transition from two disparate parts to one, united whole? Or is the link more tenuous, fragile, and, ultimately, more intriguing?

The global distribution of Greenish Warbler. Different colours represent subspecies; colour gradients show regions where subspecies intergrade. The blue/red hatched area shows the zone of overlap of P. t. viridanus (blue) and P. t. plumbeitarsus (red). The break in north=eastern China is likely to be due to recent habitat destruction. Taken from Irwin et al. (2005). 

Crucially, the two subspecies of Greenish Warbler that coexist in Siberia do not interbreed. In a previous paper on this species, Irwin et al. (2001) showed that there is a northward trend of increasing song complexity and length, but that this has occurred differently in P. t. plumbeitarsus and P. t. viridanus, with the former having many short phrases, and the latter, few long phrases. Further, they showed that females of one subspecies do not recognise the song of males of the other. Other traits have also been shown to vary with the same pattern,, suggesting that there is gene flow from one subspecies to the next. However, as all of these traits are likely to be under selection, it isn't really feasible to take this as conclusive evidence for gene flow. As a result, Irwin et al. (2005) turned to AFLP, which is a way of looking at variation across an individual's entire genome. They hoped to show that even though there is continuous gene flow around the ring, the two northern subspecies are still reproductively isolated, by sheer force of distance. And that is exactly what they did.

The implications of this are interesting in two ways. Firstly, in its own right, showing that populations can become reproductively isolated simply by being far enough away from each other, even if other populations of the same species occur in the intervening space, is of great significance. You don't always need a mountain range or a river or any of the other textbook examples, all you need is mileage.

But the other implication, which I find equally interesting, is that it challenges our concept of what a species even is. The discipline of systematics aims to reconstruct the ways in which the diversity of life has come about, but is dependent on its sister-discipline, taxonomy, to give names to that diversity. But what of the Greenish Warbler? In a way its rather vaguely descriptive common name is reflected in its nature: as a species, it is pretty vaguely defined. Being a birder, I am well aware of the dual challenge of not only identifying species (which occurs on a day-to-day basis), but also of just being able to tell when two things are in fact distinct species (which applies in a more general sense). Avian taxonomists are constantly 'splitting' and 'lumping', either adding more names to the tree of avian life, or culling them from it. But I don't think any taxonomist would really know what to do with the Greenish Warbler. Any naturalist tromping through the woods of Siberia, finding the two northern subspecies occurring together, would confidently call them different species. If that naturalist never travelled to the Himalayan foothills, and never studied the Greenish Warblers there, he or she would, in blissful ignorance, happily continue to do so. And if the 'intermediate' subspecies did not exist, so would we.

One could argue that this has occurred for many sister species in the past - the intermediates are now extinct, and we are seeing the end result. But what if this is the end result? Irwin et al. (2005) argue that in fact this system is likely to be stable, rather than progressing towards eventual speciation, and that only new processes such as habitat change have the potential to lead to what you might call 'full speciation'.

In my opinion, there is actually no solution to this problem. But that needn't worry us. It simply reflects the fact that the real world, the natural world, is immensely complex, and that sometimes our attempts at imposing order on it will be thwarted. We're just going to have to live with the Greenish Warbler being a sort-of species. I think that's great.