December 17, 2004
Shopping for gulls
An interesting topic bubbled up around an article which appeared a couple days ago in the NY Times. I won't bother linking it because one has to subscribe to get into it. Anyway, the article was about "barcoding" life. A layman friendly way of describing the process of cataloging DNA in various species as a means of identifying them. Not a new idea.
The Times article was based on a paper published online at Plosbiology and also available from the authors at their website.
Most of the birder discussion has centered around the 4 "new" species that pop out using this technique. These 4 would presumably be splits involving already described subspecies although the paper is not clear about whether the splits would actually fall along already described morphological boundaries.
What I found more interesting were the species that, if barcodes alone were used, would get lumped. Some like Northwestern Crow deserve to be lumped, but seven species of gulls (Herring, Mew, Ring-billed, Iceland, Glaucous, Great Black-backed and Thayer's Gulls) showed low divergence. In fact, the variations between individuals within a given species were as great as between species.
I can identify gulls when I have to, but I admit openly that I'm not the kind of lariphile that spends hours studying the nuances at a garbage dump full of gulls. I am disinclined to dive into the gull debates on ID-frontiers. I would not shed many tears if all these North American gulls got lumped. But I don't believe for a minute that Mew Gulls and Great Black-backed Gulls are the same species.
So I have to ask: what good is a barcode that doesn't seem to work on 5% of the species? The authors blame recent divergence and frequent hybridization. They further state that finding more genetic markers will probably solve this problem. That would be like, to push the analogy, needing two different scanners at the store to buy groceries. Clearly, this is not a magic bullet.
A critique of the barcode process takes these concerns even farther, suggesting that the phylogeography of North American species is less complex than the tropics and that the rules for a barcode methodology for tropical species may prove to be even squishier.
Posted by mbalame at December 17, 2004 3:51 AM