Vestigial organs are commonly cited as evidence for evolution; indeed, I wrote most of a chapter about them in WEIT. And although some of these organs may have a function (the appendix, for example, is now thought to play a small role in the immune system), the fact of functionality doesn’t detract from their vestigiality — or their usefulness as evidence for common ancestry. The fact that the penguin’s “wings” now serve as flippers doesn’t mean that they say nothing about penguins’ ancestry from birds that could fly. (My own definition of a vestigial trait is loosely “a trait that now serves a function completely different from what impelled its initial evolution, and for which traces of its original structure still remain.”)
And, anyway, some vestigial organs appear to have no function at all. You’d be hard pressed to claim that there was a function for the tiny nub-like wings of the kiwi bird, buried deep within the feathers on its flanks.
But you’d be even harder pressed to argue that dead genes have a function.
One of the predictions of evolution is that if vestigial organs exist, then surely vestigial genes must exist, for traits that shrink or disappear must surely be based, at least sometimes, on genes that disappear. But genes aren’t just snipped out of the DNA when they’re no longer needed: natural selection will inactivate them, usually by favoring a mutation that removes one or a few DNA bases (a deletion), putting in a “stop codon” (a change in the DNA that prematurely terminates the protein being read from that DNA’s code, making a “frameshift” mutation (changing the triplet code so that the entire coding frame of a DNA sequence is thrown off), or changing the regulatory region of a gene so its protein is no longer made. But the remnants of the gene will remain in the DNA, testifying to its ancestry from a gene that was active in an ancestor. These dead genes are called pseudogenes.
Now that genome sequencing is routine, evolutionists can do large-scale searches for dead genes. And, as predicted, they’re all over the place — in nearly every species that has been examined. Some of the “dead” genes I discuss in WEIT include human genes that used to make vitamin C or egg yolk in our ancestors, but have now been rendered mute. It would be hard to argue that these genes still have a “function,” for they produce no protein at all. Their existence is a powerful argument in favor of evolution and against creationism.
A new paper in PLoS Genetics continues the search for predicted dead genes — this time for genes that once made tooth enamel — and finds a lot of these wrecks. They’re exactly where you expect to find them — in toothless animals long thought to have descended from animals with teeth. So the “theory” of evolution is once again confirmed, although we hardly need further confirmation. But this paper goes beyond a mere redudant proof of common ancestry. The authors also make models of how the “enamel” genes degenerated, and, by calculating when this degeneration happened, predict what the teeth of common ancestors should look like. This prediction is in principle testable by finding the relevant fossils and looking at their teeth.,
There are two kinds of mammals that lack tooth enamel: those that are completely toothless (e.g., armadillos, pangolins, aardvarks, baleen whales), and those that have teeth that lack enamel (e.g., dwarf sperm whales, two-toed sloth). From other evidence, including fossils and comparative morphology, scientists have confidently predicted that every one of these species descended from ancestors that had enameled teeth. The researchers sequenced, in many mammals species, the critical gene enamelin (ENAM), which helps deposit hydroxyapetite into the tooth. If ENAM is knocked out in mice, enamel doesn’t form.
Sure enough, in every species lacking teeth or enameled teeth, ENAM was rendered nonfunctional, either by the accumulation of frameshift mutations or stop codons. The authors give a very nice graph of how the genes have degenerated in each lineage, which I show below (the caption is at the bottom of this post if you’re a geneticist).

Figure 1. Enamelin gene phylogeny for mammal species with no teeth, un-enameled teeth, or fully enameled teeth (tooth type illustrated at right, before species name). Caption for this figure, showing the various genetic changes that have inactivated ENAM, is at the bottom of this post.
Well, that’s evidence for evolution and common ancestry, and it’s neat. But we hardly need more evidence of this type to support Darwinism, because there are many such reports of pseudobenes. What makes the paper unique is the authors’ model of how and when ENAM was inactivated in several lineages.
The details are complicated, but in essence the authors modeled a two-step process of gene evolution: before the mutation inactivating ENAM (whose occurrence can be estimated from the gene tree), and thereafter. Before inactivation the gene was assumed to evolve slowly since gene changes were largely deleterious. After inactivation, the gene evolved faster — since it was no longer making a useful product, changes were assumed to accumulate “neutrally” (that is, all changes carried neither selective penalty nor advantage). This analysis made several predictions. One of them is that ancestral xenarthans (armadillos, sloths, and anteaters), had teeth with enamel, even though living representatives don’t. This is a testable prediction: find early ‘basal” xenarthans, and look at their teeth.
You may have noticed that after the gene was inactivated, it was assumed to degenerate through the accumulation of random, neutal mutations. This is a valid assumption for gene inactivation, but not necessarily for the disappearance or shrinkage of entire traits. If a trait is no longer useful, there are actually three ways it can degnerate:
1. As posited for genes, the accumulation of inactivating mutations, which carry no penalty since they affect a trait that makes no contribution to reproduction.
2. Traits can degenerate through positive natural selection. One way is simply that an animal (or plant) can be seen as an economic compromise among biological materials, and if you don’t need a feature, selection can favor rerouting its building blocks to other features. The material used to build the wing of a flightless bird, for example, could be diverted to making or strengthening bones in other parts of the body.
3. Likewise, traits that are no longer useful might be easily injured. Selection would then favor degeneration of that trait as a way of preventing injury. This may, for example, explain the reduced eyes of cave fauna or burrowing animals like moles.
It isn’t easy to distinguish between these hypotheses. One way is to find a bunch of genes involved in the trait’s degeneration and, by sequencing them, see if the DNA has changed by natural selection since the trait was no longer useful. (There are statistical tests for this, though they have some problems.)
_________________
Meredith RW, Gatesy J, Murphy WJ, Ryder OA, Springer MS (2009). Molecular decay of the tooth gene enamelin (ENAM) mirrors the loss of enamel in the fossil record of placental mammals. PLoS Genet 5(9): e1000634. doi:10.1371/journal.pgen.1000634
Caption for Fig. 1 (above), from Meredith et al.: Figure 1. Species tree with frameshift mutations and dN/dS branch coding. Symbols next to taxon names denote taxa having teeth with enamel, taxa having teeth without enamel, and edentulous taxa. Branches are functional (black), pre-mutation (blue), mixed (purple), and pseudogenic (red). Vertical bars on branches represent frameshift mutations (see Table S1). Frameshifts that map unambiguously onto branches are shown in black. Frameshifts shown in white are unique, but occur in regions where sequences are missing for one or more taxa (Figure S7) and were arbitrarily mapped onto the youngest possible branch. Homoplastic frameshifts (deltran optimization) are marked by numbers. Numbers after taxon names indicate the minimum number of stop codons in the sequence (before slashes) and the length of the sequence (after slashes).