Death to Dollo's Law!

Hint for science journalists: if the hook to get readers to pay attention to your story is to warn them to sit down because a 19th century "law" of evolution has been shown to be wrong, you're going to irritate scientists, who will then write rude blog posts sneering at your writing. That's the case with this story titled The Wrists of Birds Reveal Evolution Undoing Itself, which is also subtitled Contrary to earlier claims, a new study shows that evolution may be reversible.

It's describing a paper on the re-emergence of a wrist bone in birds, and it touts Dollo's Law.

The 19th-century biologist Louis Dollo taught that evolution is irreversible; once a structure is lost, that pathway is closed forever. It’s a principle now known as Dollo’s law.

It was called a law; it got bandied about in the literature extensively; and then, as we acquired a greater understanding of genes and molecules and the patterns of change in evolution, it was rendered obsolete and more or less wrong. What does it even mean to say that a single nucleotide change, for instance, has a "direction" or is irreversible? What we should be recognizing is that many morphological changes are a consequence of a whole series of accumulated mutations, and it's extremely unlikely that that a series of random changes would neatly reverse all of them.

I have a longish rant about Dollo's Law and it's continued promotion by biologists. I know that a key to getting your story heard is declaring that it breaks expectations, but really, people, breaking a dead, 125 year old "law" isn't news any more.

And it's a shame, because it's an interesting result. A wrist bone, the pisiform, disappeared in dinosaurs that had a relatively simpler forelimb; in their descendants, the birds, a new wrist bone appeared, called the ulnare. The research argues that the ulnare is the pisiform. Which wouldn't surprise me at all. But then a scientist has to go and say something stupid to the press.

But in analyzing the development of the ulnare, Vargas showed that it is, in fact, the re-emergence of the pisiform. “While the physical expression of a gene may be suppressed, it doesn’t mean the possibility of generating that structure has disappeared,” says Luis Chiappe, director of the Dinosaur Institute at the Natural History Museum of Los Angeles County. The gene is still there, it’s just dormant.

That last sentence? Stupid. No, the "gene" wasn't dormant. First of all, there is no "pisiform gene" -- there is a whole set of interacting genes that regulate how condensing bones in the wrist establish themselves and differentiate, and those genes are sensitive to the environment, both in the constraints on structure and in their epistatic relationship with other genes. Those genes were almost certainly functional in their entirety in birds and dinosaurs, but the dinosaur forelimb was an environment that did not promote the formation of a specific bone called the pisiform. Changes that promoted larger shape changes in the limb produced conditions under which those same genes would re-express themselves in a way that led to the formation of that bone.

Genes do not lie dormant for a hundred million years. Genes that do nothing accumulate mutations that are not removed by negative selection, and are reduced to pseudogenes, and eventually, after a very long time, are shredded by deletions and are lost.

Scientists who are not hobbled by an obsolete understanding of the evolutionary process that is not just pre-molecular biology, but pre-genetics, understand this.