Gene flow and evolutionary dynamics::two bird studies


Garant, Dany, Loeske E. B. Kruuk, Teddy A. Wilkin, Robin H. McCleery, and Ben C. Sheldon. 2005. "Evolution driven by differential dispersal within a wild bird population." Nature 433:60--65.

Postma, Erik and Arie J. van Noordwijk. 2005. "Gene flow maintains a large genetic difference in clutch size at a small spatial scale." Nature 433:65--68.

Here are two papers about evolution in field populations of wild birds, specifically great tits (Parus major). The first paper considers a single widespread population occupying a continuous woodland in England over 35 years. They measured nestling body mass, determined the genetic variance behind it, and discovered that the genetic variance is spatially variable among the eight sectors of the woodland. They found that this genetic variation among spatial locations is maintained by patterned dispersal. Specifically, smaller birds tended to migrate into a densely occupied habitat in the eastern part of the wood, while larger birds tended to migrate into the sparsely-occupied north. The authors speculate that larger birds were more successful in moving to more favored habitat where there is less mating competition. It is not known whether food availability or other ecological factors cause the difference in local population density, but the difference in mass in the migrating birds is not itself initiated by the new environment because the size differences are also reflected by their nestling mass in other parts of the wood. The interesting part of this study is that the authors were able to determine that birds with different genetic characteristics migrated to different places, and that differential dispersal partly causes the genetic differentiation of different subpopulations.

The second paper examines the clutch sizes of great tits on an island off the coast of the Netherlands. The island population is spatially structured, and the authors examined two subpopulations. While island birds tend to have smaller clutches, mainland birds have larger clutches. This difference is primarily due to genetic differences between island and mainland birds. The study observed the number of immigrant birds and their effects on the two island populations, which themselves differed significantly in clutch size. The interesting part of the study is that the authors were able to measure the fitness of birds with different clutch sizes, and thereby show that the level of selection in each of the island study populations was the same. Because the level of selection was the same, they can show that the difference in clutch size between the two island populations is entirely due to different levels of gene flow from the mainland.

Either of these studies may have interesting implications for human evolution. The idea that animals move in accordance with their physical characteristics is not entirely unexpected--since for example animals that tolerate cold better will be more likely to migrate to colder places. But the continuation of such genetically structured movement over substantial periods of time implies that genes may become spatially autocorrelated for adaptive reasons.

Why is this interesting? Think about Neandertals. They were clearly adapted differently from contemporary fossil humans, an adaptation that likely included different tolerance for cold, different energetic requirements and different habitat preferences. Traditionally, people have thought about these differences in terms of partial or complete reproductive isolation between Europeans and other populations. But an alternative is that the differences were maintained in the face of genetic contacts by structured dispersal. Consider that human hunter-gatherers live in small exogamous groups, with considerable contact with surrounding groups. In this circumstance, a dispersing individual may exercise substantial choice of new group, based on his or her (or others') knowledge. If there is a tendency toward assortative mating, or if individuals try to match their new group in some other way, then this population structure may induce the kind of structured dispersal that would maintain phenotypic gradients across space.

The second paper is less suggestive, but only because the joint effects of selection and gene flow have often been considered in thinking about the distribution of variation among ancient humans. Nevertheless, it is a good reminder that the effects of selection depend on the quantity of gene flow, and the effects of gene flow depend on the level of selection.