Friday, June 16, 2006

Silly, silly creationists

It's one thing to be uninformed and to get things wrong. That's not stupid. It's ignorance, and that can be fixed. But there are people, truly stupid people, who will read a simple thing and instantly conclude that it says the exact opposite of what it actually says. Consider the IDolators of ARN who link to this creationist rant:
Wilson et al. examined environmental effects on wild sheep. The team studied, “for the first time, how variation in environmental quality simultaneously influences the strength of natural selection and the genetic basis of trait variability.” Unfortunately, the results were opposite than expected: “our results show that environmental heterogeneity induces a negative correlation between these two parameters.” The implications seem damaging for evolutionary theory in general
No, no, no. These results were not opposite than expected, and are not damaging to evolutionary theory in general.

As the paper explains:
The significant negative interaction between standardised birthweight and [environmental quality] indicates that the strength of the positive relationship between fitness and phenotype is reduced in better environments.
Duh! A good environment is, by definition, one that is less harmful. That mean's there's less death, and less death means less selection. This is exactly and precisely what you'd expect. Indeed, the authors refer to this result as a confirmation, not a refutation.

In good times, selection is reduced and genetic diversity increases. When bad times hit, selection kicks in, and a lot of that diversity is killed off. You have evolution.

What that means is that:
maternal genetic variance for birthweight increases with environmental quality, while the strength of selection decreases. This shared dependence on environmental quality results in a negative correlation between the strength of selection on birthweight and the amount of heritable variation.
At any point in time, evolutionary predictions are satisfied. When you average across timescales, you get a negative correlation between genetic variance and selection, but that stands to reason also. The more selection has happened, the less variation is left. It's often easiest to think of boundary conditions. If every sheep had a different allele, any mortality at all will reduce the genetic diversity. Any mortality in a genetically uniform population does nothing to the diversity. Bearing in mind that natural selection relies on differential survival and reproduction, it is impossible in a genetically uniform population.

When you understand that, you understand the basic principle of Gould and Eldredge's concept of punctuated equilibria. Most of the time a population is in some sort of equilibrium with its environment. There's some stabilizing selection, but most offspring survive and mortality due to predation is largely random or age related. Genetic diversity builds up, populations split, sometimes a new species forms, but not much happens, maybe for thousands of years.

Then something shakes up the environment. Lots of species can't survive at all, and those that do experience massive and selective mortality. The survivors are very different, and that difference passes on for generations. Lots of new species and lots of turmoil emerges from the change. In time, things settle down, populations grow, and genetic diversity increases.

Clearly, similar cycling happens at shorter temporal scales, too. Winter and summer bring different selective environments in the temperate zones. To survive a full year, you have to be able to handle both extremes. If you reproduce multiple times per year (and most species do), then each bout of reproduction will result from different selective environments through the year.

None of this is novel, none of it is surprising. I can't imagine why the IDolators think it's a problem.