Mailbag: Chance and recent selection

3 minute read

Our work on recent selection was featured in Discover magazine this month. I’ll link to that later. In the meantime, I’ve been getting some thoughtful letters from readers of the article. I thought I would post some of these letters with answers, because they really illustrate a cross-section of interest in the work. Here’s the first:

I am just a private citizen, but I have thought about something for some time. Let's take 1,000 sets of identical twins and divide them into two groups. The groups will be put far apart and never come in contact with each for 1000 years. The two areas the two groups go to are identical and the number of them stays at 1,000.
After the 1000 years they are brought back together. I will wager $10,000 the two groups will have evolved differently, even though they lived in identical climates. Is anyone willing to take me up on the bet?
I am guessing if we took one set of identical twins who could live for 1000 years, separate them for the 1000 years in identical climates, the same thing would happen.
My thinking is that climate is part of the evolution process, but the more complicated life gets means the body has to adjust to the changes. If one group's environment doesn't change, then they only have to adjust to the environment once. If the other group starts to invent things and are constantly improving on things, then they have to keep evolving to the constant changes. I say living in one area that keeps changing for 1,000 years will change the person.

There’s a bit more to this letter, which I may include later. In the meantime, here’s what I wrote in response:

Thank you for your letter. In fact you are entirely correct; if you separate two sets of identical people for 1000 years, the two populations will evolve differently. This will be much more so if you separate 100 people instead of 1000, as the chance element of evolution is the largest factor in these small populations.
Now, on the other hand, if you separate two groups of 100,000 or 1,000,000 people for 1000 years, we will see very little genetic change at all in either group. Except to the extent that their genes are subject to selection.
But as you mention, the genes of these large populations may evolve differently even if their phenotypes are subject to the same environment. For example, Europeans and north Asians have both evolved lighter skin color in the last 20,000 years. But in these two populations we see very different genetic changes. Europeans have a high frequency of new genetic changes in genes like SLC24A5, OCA2, and Mc1r; Asians also have a change in Mc1r but lack the others; instead they have changes in other genes like DCT.
Again and again in recent human evolution, we see that chance element influencing the variation that selection has to work with. In maybe the most famous example, different populations have come to suffer from falciparum malaria in the last 5000 years. In these populations, we see diverse mutations that help to resist the malaria parasite. The sickle cell trait became common in West Africa and north India; West Africa also got hemoglobin C and G6PD deficiency. Hemoglobin E arose in southeast Asia; alpha- and beta-thalassemia appeared in the eastern Mediterranean and elsewhere; ovalocytosis in New Guinea. These different genetic adaptations have different values and different costs. If humans had always lived in a single population exposed to malaria, some of these adaptations may not have proliferated. But the long distances and slow movement between these populations means that new adaptations can grow in numbers faster than they spread to different parts of the globe.

I certainly wouldn’t take up his wager. But who knows, there may be a profit opportunity here – there are a lot of folks studying human genetics who don’t think evolution in recent humans is possible!