Sunday, April 10, 2005

Why we call him Dumbski

Billy Dembski writes at his new blog Overwhelming Evidence? – Quantity vs. Quality of Evidence:
Evolutionists cannot predict major biological innovations or provide detailed testable models of how such innovations occurred.
If he'd read an evolution textbook, or even a biology text in the last 15 years (and I can't really say what happened before then) he'd know that there are lots of predictions that evolution makes and important testable models. Or check through the Evolution Project.

Among the non-obvious predictions are sex ratios at sexual maturity, in contrast to at birth. One might expect that most species should have equal numbers of male and female offspring, given how the sex chromosomes work. In fact, most species aren't 50-50 at birth. Humans have a male biased sex ratio at birth. By the time a male-biased cohort reaches their teen years, the ratio has evened out. Male children are more likely to injure themselves and to die of diseases.

Simple theoretical arguments show that, in general, the sex ratio of reproductive individuals should be 50-50, which is what we find. In a population which is skewed toward females (for instance) each male has more offspring (on average) than each female. The children of an individual with a genetic propensity to have more male offspring would pass more genes on to their children (the grandchildren of the first mutant). That mutation will be genetically favored until the sex ratio becomes male biased, at which point individuals with a tendency to produce female offspring will be favored.

This is not an obvious result without careful evolutionary reasoning. In order to ensure an even adult sex ratio, it's necessary to bias the ratio at birth. In fact, different human populations have different sex ratios at birth.

In addition, there are circumstances where female bias might be preferred. Certain wasps are fertilized at birth. The mother lays her eggs in a fig. The males hatch first, and go to each female egg and fertilize them before they hatch. In this case, all the offspring share the same mother, and there's no benefit to an even sex ratio; one male could fertilize each egg. In fact the sex ratio in these wasps is very far from even, as predicted.

Interestingly, this pattern occurs in both Old World and New World figs and the wasps which pollinate them.

This is textbook stuff. Great, classic science. Part of its fascination is that the ideas aren't obvious until you work through the evolutionary logic.

Does this disprove non-evolutionary models? No. But it does disprove the claim that we can't provide testable models, verified by evidence, of how major innovations evolved.