How many of my readers remember when I reviewed Michael Behe's Edge of Evolution? One of Behe's main arguments is that two protein protein binding sites cannot evolve at the same time. As the argument goes, systems like the bacterial flagellum and cilium use many more than two protein protein binding sites, thus evolution by random mutation could not have occured.
When I reviewed the book, I sided with Jerry Coyne, and thought that Behe had made some unsound assumptions when calculating probability. Later on, Nick Matzke wrote a devastating review of Behe's book. Concernin Protein-Protein Binding sites, he said,
"Snake venom shows that even vertebrates with small populations can evolve huge gene families that specifically bind diverse proteins, with massive evidence of duplication, mutation, and selection as the mechanisms, and with intraspecific variation in regulation, sequence, and specificity. Is Someone actively designing rattlesnake (Sistrurus catenatus) venom in the American Midwest  and fine-tuning the specificity of black mamba (Dendroaspis polylepis) toxins for subtypes of mammalian muscarinic acetylcholine receptors ?"
Seeing as how we had evidence of binding sites evolving so recently, I was confident that Behe would soon be answered. Now, less than a year after The Edge of Evolution, the answer has come: Duplicated binding sites can be altered to produce brand new binding sites. There is a lot of evidence that many of our binding sites are actually duplicates of other binding sites. And better, we have observable, repeatable experiments to base this on.
I wonder how Behe will react? More arm waving? Finally admitting he is wrong? Fat Chance.