Perhaps the most persistent criticism of evolutionary biology is that natural selection isn’t sufficient to explain life’s complexity. This of course is the mantra of intelligent design, which claims that in principle natural selection can’t explain “irreducibly complex” traits (for a refutation of this see the discussion at TalkOrigins).
But there’s also criticism that there is not much evidence for natural selection. I’ve discussed this in WEIT and in my recent review of two books (and of natural selection) in The Nation. The refutations of this claim are easy: we have lots of evidence for selection occurring in nature, including the many studies cited by John Endler in his book Natural Selection in the Wild, and more recent studies including the famous work by Peter Grant and his colleagues on selection for break size in a Darwin’s finch (for a summary of recent studies go here).
Evidence for natural selection also comes from the many observations of animals and plants responding to human-induced changes in the environment. These include the famous studies of wing color in the peppered moth, heavy-metal tolerance in plants, insecticide resistance in insects, size changes in animals due to overharvesting, and of course antibiotic resistance in bacteria.
Some creationists object to using these cases as evidence for selection, since the selection pressures come from humans, making these (so they say) analogous to artificial selection. This objection is misguided. Yes, the selection pressures result from human activity, but far from trying to produce a given response, humans do not want a response. Who wants bacteria to become resistant to penicillin?
Further, unlike animal and plant breeding, these responses to anthropogenic changes in the environment are not directed to a particular end by human desire. When a plant evolves resistance to lead or copper on mine tailings, it’s drawing on a naturally-occurring pool of genetic variation—exactly as if the plant were responding to high concentrations of metals that occur naturally on serpentine soils. I see all genetic responses to human meddling with the environment as evidence for natural selection, and we’re going to see more as the climate gets hotter. Evolution is going on around us all the time, but it’s often invisible.
In the past week there have been two new reports of natural selection causing evolutionary change as a response to human activity. The first is in today’s New York Times: a discussion of how weeds in the U.S. are becoming resistant to the herbicide Roundup. The Roundup story is complex, but the basics are that the herbicide’s active ingredient is glyphosate, an amino acid analog that kills plants by interfering with the synthesis of some amino acids in the growing points. It’s really effective at wiping out weeds, but can also kill growing crop plants if it’s sprayed onto fields after planting. In the 90s Monsanto also developed “Roundup Ready” varieties of transgenic crops (which now include corn, soybeans, cotton, and alfalfa), in which they introduced into plants a gene from bacteria that allowed amino-acid synthesis to proceed normally in the presence of Roundup. The combination of Roundup and Roundup Ready crops made millions for Monsanto, and farmers liked the convenience of not having to till the soil before planting to get rid of weeds. But there were lots of objections to the use of transgenic crops and to the dependence of farmers on corporations who made both herbicides and herbicide-resistant crops.
But now, as reported in the Times, weeds are evolving resistance to Roundup throughout the world. As one soybean farmer observed, “We’re back to where we were 20 years ago.” So far only about 5% of crops are infested with resistant weeds, but I can confidently predict that natural selection will make things worse.
I haven’t followed the Roundup controversy closely, but the Times reports that Monsanto “once argued that resistance would not become a major problem”. That’s insane. Any evolutionist will tell you that you can’t predict stuff like that. More often than not, natural selection finds a way around these things, just as bacteria have devised insidious ways to combat antibiotics, and the antibiotics that replaced those antibiotics, and so on to the point where in some cases (like TB), we have bacteria that resist all known antibiotics.
The problem will get worse, and companies will have to develop new herbicides, putting farmers into the same spiral that doctors have been in with antibiotics for decades.
Crop plants are afflicted with insects as well as weeds, and those too can adapt to human-induced selection. A new report in The Proceedings of the Royal Society of London by Vincent Calcagno et al. shows that one crop pest, the European corn borer (Ostrinia nubilalis), seems to have responded to corn harvesting by changing its behavior, making the larvae less likely to be killed during the harvest.
The corn borer appears to have evolved from the sister species Ostrinia scapulalis about 500 years ago when corn was introduced to Europe. O. scapulalis lives on the non-crop plant mugwort (Artemisia vulgaris). Both species undergo a winter “diapause”—a period of arrested development—in the stems of their plants. (We are talking about the caterpillars here, though both develop into moths.)
Calcagno found that, unlike the mugwort-infesting species, larvae of the corn borer move down from the tops of the plant before they diapause. This was shown both in the wild, by putting both species on either corn or mugwort, and in the lab by making artificial “stems” out of food-filled plastic tubes. The down-movement (“negative geotaxis”) of the corn pest did not depend on living in corn, but was also seen when the corn borer was forced to live in mugwort. In contrast, O. scapulalis did not move down at all before diapause, but tended to overwinter right where it was in the plant.
Calcago et al. speculate that this difference was due to selection associated with the harvesting of corn. That harvesting now involves lopping off the top 15-40 cm (about 6″-16″) of the plant to get the ears. The truncated stalks are then left standing in the field over winter. This would impose strong selection on corn borers to diapause further down in the stalk so they wouldn’t get lopped and killed. Calcagno estimate that this selection is pretty strong, eliminating roughly half of the corn borers. One would certainly expect a response to that selection, and it seems to have appeared in the last few centuries. This of course makes things worse for corn farmers, because many of the pests who would be killed during the harvest now survive.
Now their case isn’t airtight, for they haven’t actually nailed down the evolutionary cause of the species difference in movement. All we see is a difference between species and a pretty plausible story that is supported by experiments. One way to test their idea would be to see what happens in populations of corn borers that aren’t harvested—presumably they would revert to the ancestral behavior. (Farmers wouldn’t want to do this test, of course, but it could be done in an agricultural station.) Calcagno et al. also mention another species of corn borer, O. furnacalis, that might have evolved independently in Asia. If that species showed the same down-moving behavior it would strengthen Calcagno et al.’s case for the evolution of a corn-harvesting-associated behavior.
What all this shows is that our control over nature isn’t as strong as we think. As the volcano in Iceland abruptly reminded us, we can’t fool Mother Nature. And we can’t fool natural selection, either.

Fig. 1. The European corn borer doing its job. It’s a pest because these infestations can stunt the plant or make the ears drop prematurely. (Photo from Iowa State University.)
UPDATE: Carl Zimmer goes into more detail about Roundup at The Loom.
h/t: Greg Mayer
_________
Calcagno, V., V. Bonhomme, Y. Thomas, M. C. Singer and D. Bourguet. 2010. Divergence in behavior between the European corn borer, Ostrinia nubilalis, and its sibling species Ostrinia scapulalis: adaptation to human harvesting? Proc. Roy. Soc. Lond. B:doi: 10.1098/rspb.2010.0433 (online).