A proposal that butterfly caterpillars are descended from the mating of an ancient insect with a type of worm is comprehensively rejected by a new paper to be published this week in the journal Proceedings of the National Academy of Sciences.
The author of the original paper, Donald Williamson of the University of Liverpool, England, appealed to the scientific community to test his hypothesis. Rick Grosberg, professor in the Department of Evolution and Ecology and the Center for Population Biology at UC Davis, and Mike Hart of Simon Fraser University, British Columbia took up the challenge, showing that Williamson’s hypothesis can be disproven using data already published and readily available.
“There’s not a shred of genetic evidence supporting a hybridization event, which there absolutely should be,” Grosberg said. “All the predictions that you would make, beyond the most basic, do not hold up.”
Williamson proposed that caterpillars are descended from an onychophoran, an animal sometimes called a “velvet worm.” These worms have a segmented body and simple legs. In the distant past, he proposed, such a worm mated with an ancestral insect. The resulting hybrid had two genomes, one from each parent. The worm-like genome went on to form the modern caterpillar, while the insect genome formed the adult butterfly. The two genomes would have continued side-by-side to the present day, with each controlling different stages of the life cycle.
Among Williamson’s predictions are that modern butterflies and other insects with a caterpillar stage will have a larger genome than either modern onychophorans or insects that do not have a caterpillar.
Data readily available online show that this is not true, Hart and Grosberg point out. In fact, butterflies have a smaller genome than either onychophorans or other insects.
Several insect genomes have been sequenced, although the genome of the onychophoran Priapulus is not yet complete. But a published study of 149 genes from insects, onychophorans, scorpions, crustaceans and other arthropods does not show that onychophorans are particularly closely related to the butterflies, or that butterflies have a mix of insect and velvet worm genes.
The original paper, published in the same journal earlier this year, was met with widespread incredulity. Many scientists questioned how the paper came to be published. One researcher quoted by the media compared the journal to a supermarket tabloid.
PNAS has two routes for accepting scientific papers. Most follow a similar route to other scientific journals. They are submitted to the editorial office, which assigns other scientists to carry out an independent, confidential peer review. But in the case of PNAS, members of the National Academies can also “communicate” a limited number of papers per year to the journal and manage the peer review process themselves.
The Williamson paper was submitted by this latter route, communicated by Lynn Margulis, professor at the University of Massachussetts, Amherst.
Grosberg and Hart opted not to address the issue of how the paper came to be published.
“We wanted to point out that the data and tests exist, and deal with it on scientific grounds,” Grosberg said.
Grosberg noted that Williamson has been advocating similar ideas since 1989, claiming that sea squirts in his laboratory could interbreed with sea urchins, a different phylum of marine animal. Other scientists, including Hart, have not been able to reproduce that work, and shown that the actual larvae that Williamson raised contained only sea urchin DNA
“The good side of this is that this idea might have lain around and resurfaced in minor journals for decades. Now it’s been front and center,” Grosberg said.
“The scientific process worked in the end, but a bunch of people went to a lot of trouble to refute an extremely implausible idea.”