Can we ever be sure about what happened long ago?

Often when I lecture about evolutionary biology, a member of the audience will say, "You don't have a time machine, so you can never be sure about what happened long ago, this is all this speculation."

Well, it is hard work, and often no experiment can conclusively differentiate the hypotheses. However, sometimes one can be absolutely sure.

Paleontologists have long recognized a major decline inlarge mammals about 35,000 years ago in the vicinity of the land bridge between Alaska and Siberia. The cause has now been identified.

Firestone and colleagues at Berkeley hypothesized that a meteor impact killed off the mammals. To test this, they examined 8000 mammoth tusks for the presence of metal particles. Seven tusks and a bison skull had magnetic particles embedded, and the levels of nickel and titanium were substantially different from those found in terrestrial iron samples. One more remarkable observation is the clincher. In all of the tusks with multiple particles, the particles were all on the same side of the tusk.

So, can we be sure? I find it convincing. But let's see the publication and give it a year for comments.

Ten Ways to Apply Evolution in Medicine

I have always found it somewhat confusing that evolutionary principles can be applied to medicine in so many ways. From the start of my work with George Williams, it has seemed clear that our attempts to ask why natural selection left the body vulnerable to so many diseases are fundamentally different from applications of population genetics or phylogenetic trees. Finally, in preparing an article with Stephen Stearns, I came up with a framework that seems helpful.

First it is essential to be clear about the kind of question:
1. What is the phylogeny of the trait?
2. How has the trait given a selective advantage?

Second, it is important to distinguish five different objects of explanation:
1. Human (or other focus species) phenotypic trait
2. Human (or other focus species) gene
3. Pathogen phenotypic trait
4. Pathogen gene
5. Somatic cell lines such as cancer or immune cells

This fleshes out into a rather nice table that defines ten ways evolutionary biology can be applied to medicine.

The full article is available free from Evolutionary Applications.

Nesse, R. M., & Stearns, S. C. (2008). The great opportunity: Evolutionary applications to medicine and public health. Evolutionary Applications 1(1), 28-48.