The theory of evolution has two main planks: 1) organisms (and species) are related by common descent, and 2) variation is generated by mutation, natural selection, and chance (e.g. genetic drift). In a couple of places, John has now cast aspersions on mutation as a source of variation. Note that this doesn’t touch the first plank, common descent – that’s the one I’ve been pushing him on (e.g. here and here). No answers from him yet on that plank.

I’m happy to take up the topic of mutation, though.

First, in his “Trees and Kinds” post, John asserts without argument (and without defining his terms) that mutation can’t produce new kinds, because complex “information” always comes from intelligence. That’s demonstrably false on any reasonable interpretation of “information”. Karl Sims’ evolved virtual creatures in this cool movie and many other kinds of evolved artificial life (e.g. here and here) are clear counterexamples. In these examples, the programmer is like a designer who creates the first life, and then lets evolution do the rest. It’s clear that the increase in “information” as these creatures evolve comes not from the designer, but from the evolutionary mechanism. Indeed, the researchers involved often have little idea of how the creatures are even managing to do these things!

In the same post, John claims that biologists believe that mutation produces variation in organisms “by blind faith.” (The skunk complains of the rose’s stench.) Here are some powerful reasons why biologists accept mutation as the fundamental source of variation upon which natural selection acts:

a) Mutation-based evolution has been observed (despite John’s claim to the contrary):
• In Lenski’s meticulous experiment on E. coli, small populations (by bacterial standards) have been reproducing in controlled conditions (which limit evolution’s scope) for years, and Lenski’s team has tracked all the mutations. Some of these mutations have led to new capabilities, for instance the consumption of citrate in the presence of oxygen (a capability that is used to distinguish normal E. coli from Salmonella).
• Kubinak et al. have observed viruses mutate to resist particular parts in mouse immune systems.
• Peter and Rosemary Grant observed mutation-based evolution occurring over the course of decades in Darwin’s finches. (More on finches in a moment.)
• Castro-Nallar and colleagues have shown how, over the last hundred years alone, the HIV virus has mutated from a single type to hundreds. Many of these mutations are beneficial to the virus (though not to us!); for example, resistance to AZT.
• In the oral debate, John stated that antibiotic resistance in bacteria results only from genetic variation that is already present in the original population of bacteria. This is false – many novel mutations have been studied that lead to antibiotic resistance in various types of bacteria. See the review, Fitness effects of mutations in bacteria, or here (quote: “Spontaneous mutations, in particular, are pivotal in the emergence of novel resistances.”)
I could go on an on. But of course John won’t actually comment on any of those cases. He’ll just repeat his mantra, “But you haven’t shown me any evidence, yet, Dan! Where is it?” Maybe he thinks “evidence” means “bananas.” I’d have to agree with him that none of my posts contain bananas.

b) Mutation has been observed in the lab, of all sorts of different kinds (see my next post for some details). When we look at DNA sequences of modern related organisms (whether two human cousins or mice and bears) and compare them, we can work out the changes that would have led from a common ancestor’s DNA sequence to the sequences of those modern organisms. (Sometimes we even have the ancient DNA sequence, like with this polar bear.) Those transformations fit very nicely with the types of mutations we’ve observed in the lab.

We can even work out roughly how long the transformations must have taken, based on observed mutation rates. Then we can go to the fossil record, and see if that prediction matches. (See the amazing example of Foraminifera, in note 2 here.) Time and again, the genetic data fits with the fossil record. This is stunning evidence that mutation is the fundamental source of variation.

I should note that evolution acknowledges other sources of variation besides mutation, e.g. symbiosis (mitochondria and chloroplasts), sex, and recombination. (You might include here the insertion of viral genomes too, but I’m just counting that as a type of mutation.) Mutation doesn’t have to do all the work.

Finally, let me say something about John’s mistake about the finches, in his fifth paragraph here. He cites a Nature article (which is volume 442 by the way, not 445 as he states), claiming that it shows finch variations are not caused by mutations. He says “The genetic information for all finch beak shapes is built in and they can flip flop beak shape almost at will.”

First, flip-flopping is irrelevant – if the selection pressures reverse direction (as the Grants observed), so will beak shape. Evolution doesn’t have a “target”, though this is a popular misconception. Evolution is just about change whatever the direction.

Second, the study John cites does not show that beak shape variations are not caused by mutations. In fact, the article assumes that they are. What the article is showing is that there are separate mechanisms that determine different aspects of beak shape (e.g. length vs. width). This means that mutations can affect length and width independently, which makes it easier for finches to evolve new beak shapes. Of course, John would say: God designed that mechanism for easier evolvability; biologists would say that it’s a product of evolution – same old dispute. What’s not in dispute is that it’s mutation that causes the variation. (Indeed, Petren et al. work out the history of those genetic changes in “A phylogeny of Darwin’s Finches based on microsatellite DNA length variation”, Proc. R. Soc. Lond. B (1999) 266, 321-329.)