A recent post to this blog was regarding the article in PNAS positing a hybrid origin for the larval stage of lepidopterans (the insect order containing moths and butterflies). More specifically, Williamson (2009) proposed that modern lepidopterans, with their highly complex larval life stage and metamorphosis, acquired it not by the traditional neo-Darwinian means of descent with modification, but instead by hybridogenesis with adult onychophorans – the latter being an ancient, distantly related group thought to be the sister group to Arthropoda (the animal phylum of lepidopterans and all other insects, as well as groups as varied as crabs, millipedes, and barnacles).
Considerable negative attention has been directed towards Williamson’s kooky idea since it was published – and even more to the fact that it was published in one of the more prestigious journals of our field – and still more to the (possibly unscrupulous) manner that the article was handled by the communicating National Academy member, Lynn Margulis. Much of this rapid-fire negative attention came from the blogosphere, most notably from Jerry Coyne who dubbed the article the “worst paper of the year” on his popular “Why Evolution is True” blog.
A more formal scientific response was sure to follow. First, Gonzalo Giribet of Harvard responded with a short letter to PNAS entitled “On velvet worms and caterpillars: Science, fiction, or science fiction?” in which he points out that, in making his claims to the hybrid origin of lepidopteran larvae, Williamson has gratuitously overlooked a large body of phylogenetic evidence showing that onycophorans are the sister group to arthropods, and not closely related to either larval or adult lepidopterans.
More recently, Michael Hart and Richard Grosberg went farther, publishing a very thorough and deliberate rebuttal to Williamson (2009). In a "Mythbusters" style article uncreatively (but fittingly) entitled “Caterpillars did not evolve from onychophorans by hybridogenesis” (emphasis mine) they systematically review virtually all of the predictions of Williamson’s hypothesis. . . and just as systematically reject them! For instance, Williamson predicts that under his hybridogenesis theory holometabolous (metamorphic) insects will have larger genomes than non-metamorphic insects. Hart and Grosberg point out that C-value (whole genome size) information exists for a wide range of insect species in metamorphic and non-metamorphic groups. Analysis of these data has revealed that, in fact, species in holometabolous insect orders have genome sizes that are on average smaller (not larger) than non-metamorphic insects, although the ranges of genome size in both groups overlap broadly. Numerous other predictions from Williamson (2009) are similarly tested and rejected.
Since I originally blogged on this subject over two months ago several additional comments about the article, the “crackpot index” of the author, or the submission and review process at PNAS have appeared online (e.g., here, here, and here). Horrifyingly (but predictably) the “Institute of Creation Research” has even picked up on the controversy, labeling it as evidence for “deep disagreement” among evolutionary biologists (I’m loath to link to the article here, lest that in so doing I increase the page rank of the ICR website). My interpretation is to the contrary. Controversy abounds in our field, as it should in any vital, active area of science. However, rarely have I seen a group of evolutionary biologists as unified as they have been against this contrary theory and furthermore its questionable publication in a respected journal of our field!
Reporting on the Reptiles of Redonda
19 hours ago