I’m reading Richard Dawkins’ A Devil’s Chaplain and in an essay on speciesism he mentions ‘ring species’, and the example he gives is the Herring Gull/Lesser Blackbacked Gull ring.
Wikipedia list two more ring species, Ensatina Salamanders and Greenish Warblers.
Are there any others?
I have to say I was quite intrigued by the whole concept… it was not something I’d heard of before, which I’m quite ashamed to admit to as a biological sciences graduate (microbiology).
Considering how clear an example they are of speciation in progress, they don’t seem to be talked about all that much.
Talkorigins mentions a few more examples here, but not in any detail, but perhaps that page might make an adequate springboard for further searches.
Greenish warbler, Phylloscopus trochiloides
Thanks, that helps a lot. 
Are there any examples (outside of fiction) of “ring” species which actually form a complete ring, and not just a chain? So not only can A breed with B, and B with C, … and Y with Z, but Z can also breed with A?
If that were the case, they wouldn’t be ring species, but just an ordinary set of subspecies. There are plenty of cases of circumpolar subspecies that are presumed to be able to breed with each adjacent one all the way around. (I am not sure how well actual interbreeding is documented in most cases, though.)
Ring species are very rare, because you have to have just the right set of geographical conditions for them to develop. More frequently the “ring” will be broken by the formation of full species within it; or else the terminal populations will still be able to interbreed and thus just be subspecies.
So “horseshoe species” might be a better name? 
I find the whole concept suspect. If you can interbreed very different animals, like asses and horse, lions and tigers, bison and cattle, then the thought that these barely visible differences would stop a couple of horny salamanders just defies credulity.
Instead, what I see is that biologists, always hoping to identify and name new species, for the sheer glory of it, have declared species separate that really aren’t.
I don’t know of any examples, but I believe it is at least logically possible for such a thing to occur; the trick is that A and Z aren’t joined ends, they’d have to just be divergent relatives the same as any other set of neighbours.
Hypothetical scenario in which this could occur? - volcanic activity or uplift creates a small island in the middle of an ocean; the shore of this island is colonised by some organism that is not particularly rapidly mobile and can only live in shallow waters.
Over time, the island continues to rise and grow larger; and the organisms continue to inhabit the growing coastline; because of the slightly limited gene flow caused by their sedentary nature, diversity starts to become more conspicuous and they diverge in all directions; eventually this expresses itself as ability of any given individual only to breed with local subspecies, but those subspecies form an unbroken ring around the cost of the enlarging island. So they would be ring species in both the logical and topographic sense.
You’ve actually got it slightly backwards - the argument is that the Ensatina salamanders are a single species, but one where the terminal populations that have come into sympatry from two different directions have diverged sufficiently that they won’t hybridize.
Horniness may have nothing to do with it - it might be a case of a postzygotic inviability ( i.e. hybrid sterility, as with a mule ), though that remains unproven far as I know. This causes a problem ( yet another 
) with the traditional Biological Species Concept ( and to be fair with other concepts as well ), in which you have an organism that breeds successfully on a geographic continuum in one direction, but not the other.
I don’t think you understand the Biological Species Concept. There many cases of extremely similar looking species (often referred to as sibling species) that are essentially indistinguishable to us, but which have been proved to be good species because they never or very rarely interbreed in the wild even though they occur in exactly the same places. Some such species may even be completely interfertile if they do happen to mate. Interbreeding is prevented by differing courtship behaviors, such as different songs, or other behaviors, such as being active at different times of years. That these entities are in fact true species can be demonstrated genetically.
Here is are Willow and Alder Flycatchers side by side. They cannot be reliably distinguished even in the hand; they can only surely be separated by call and by genetic analysis. It is very well established that they are good biological species.
Unless I am misunderstanding you and Chronos this is incorrect. This is not a ring species by definition. To be a so-called ring species in the sense of the Biological Species Concept, two of the populations must overlap and not interbreed in the area where they overlap. If each subspecies can interbreed with each neighboring subspecies, then they all constitute part of the same species.
According to the Biological Species Concept, two forms may be considered different species if they are genetically isolated from one another. It is OK if they occasionally hybridize, as long as this does not result in the free flow of genes between the two populations.
By this definition, ring species do not consitute true species. If each one of the adjacent subspecies can interbreed freely, then the terminal populations can exchange genes through the intermediate populations and must be considered to be part of the same species even if they cannot interbreed with each other. This may be the situation in the Greenish Warbler.
On the other hand, if hybridization between some of the populations is only occasional, then gene flow is restricted and these populations should be recognized as full species. Instead you will have a ring of closely related species, some of which hybridize on rare occasions.
Note that the classic example of ring species, the Herring Gull and related species, is no longer considered to be such a case. The various populations involved are probably best recognized as full species according to current taxonomic standards:
No no no The Herring Gull is NOT a ring species.
Likewise the Ensatina salamanders represent a complex situation and may best be recognized as several different species.
The bottom line is, however, that there is no clear cut dividing line between subspecies and species. All kinds of intermediate situations occur, and it is always going to be a judgement call where exactly to draw the line.
Not at all, since the gull species have overlapping range but do not interbreed. Whether they could is another question and I don’t know the answer to. I believe the species of snail mentioned by Dawkins also has overlapping range.
Mathematical in joke: I call them Riemannian species (since they live on a Riemannian manifold).
My understanding was that populations within ring species could breed with close neighbours, but not with distant ones; that the two ends of the gull ring species can’t breed is a result of their distance from each other (they’re not adjacent; they’re separated by the circumference of the entire ring.
In fact that isn’t really a ring at all - it’s a bent line. I think my hypothetical example, if it existed, would qualify if populations could breed with close neighbours, but not with distant ones - I don’t see why it matters that this would be happening in a topology that consisted of a point stretched into a circle instead of a point stretched into a line.
I’m not sure what point you’re trying to make with that link, Sage Rat. Could you elaborate?
My point is mainly that “ring species” has a specific definition, and that definition requires the terminal populations( that cannot interbreed) overlap. Since your populations don’t overlap, it’s not technically a ring species.
Also, while the situation you describe is theoretically possible, it would be difficult or impossible to test in the real world. The BSC depends on what happens in nature; since the populations in question don’t overlap one could not assess their species status directly. It could be possible that the populations would be intrisically intersterile due to distance, but this is unlikely if they are connected by other interfertile subspecies. Note that in several of the other cases of ring species, the lack of interbreeding is mainly due to differences in mating behavior, not to intrinsic intersterility.
Of course there are many cases in which the extreme populations of a wide-ranging species might not be able to interbreed. A true ring species is just an example of the rare case in which this is testable under natural conditions.
I didn’t realise the overlap was a part of the definition, thanks for explaining it.
Mankind made a full circle of the globe. We can all breed just fine.
If we’d had more time before Africans and white people/Native Americans had tried breeding, true we might have developed into a ring species, but I suspect that now we’ll be stuck as simply being regional variants.
Then I don’t understand why you said “Yes. Yes, there is.” That’s about as far from what Chronos was asking as you can get.
