Genetic drift is not the only force that causes populations to diverge from one another. Environments are not homogeneous, and therefore different genotypes will be selected for in different parts of a species' range; this is known as local adaptation. A commonly cited example of local adaptation is heavy metal tolerance in plants and fungi. Suilloid ectomycorrhizal fungi (Suillus luteus and S. bovinus), for example, were found to be highly zinc-tolerant when collected within 5 km of a zinc smelter, whereas those collected at least 15 km away from the pollution source were zinc-sensitive (Colpaert et al., 2004). As with genetic drift, adaptation of populations to contrasting environmental conditions is a diverging force but, unlike drift, adaptation can occur even when gene flow is relatively high. The co-existence of these diverging and homogenizing forces has led to one of the most challenging questions in ecological genetics: how can adaptation occur in the face of gene flow?
The answer to this question involves a trade-off between the amount of gene flow and the strength of selection pressure acting on the trait in question. Selection pressures can be quantified on the basis of relative fitness levels. The phenotype (and its underlying genotype) that leaves the most descendants within a population is considered to have a fitness value (w) of one. Since all other genotypes have a relatively lower fitness, their fitness values must be less than one. Consider a hypothetical population in which genotype A has the highest fitness (w = 1) and the relative fitness of genotype a is 0.75. This means that for every 100 A genotypes that survive, only 75 a genotypes will survive. A related measure is the selection coefficient, which is the corollary of w and a measurement of the reduced probability of survival. It is calculated as:
The selection coefficient against genotype a is calculated as s = 1-0.75 = 0.25, i.e. survival of a genotypes is predicted to be 25 per cent lower than A genotypes in each generation. For any given genotype, the higher the selection coefficient, the stronger the selection pressure against that particular genotype will be, and the more likely that genotype is to become extinct. One of the best-known examples of selection coefficients is industrial melanism in the peppered moth Biston betularia. Although most populations were light coloured before the industrial revolution, dark forms of the moth became favoured in polluted areas because they are better camouflaged against blackened trees. The selection coefficient against light-coloured moths in these regions has been estimated to be 0.32, making their chances of survival approximately 32 per cent lower than dark-coloured moths (Haldane, 1924).
If selection pressure is high enough, local adaptations may persist despite high levels of gene flow (Endler, 1977). This has been demonstrated in a number of species, including the Dominican anole (Anolis oculatus), which is a morphologically variable member of the family Iguanidae that is endemic to Dominica. Morphological diversity in this species, which includes variable tail depth, scale size and body size, reflects adaptation to environmental variables such as rainfall and type of vegetation, both of which vary markedly across the island. Populations therefore show substantial levels of morphological differentiation, but at the same time microsatellite data revealed surprisingly high levels of gene flow, with only 7.8 per cent of population pairs revealing FST values that were significantly greater than zero (Stenson, Halhotra and Thorpe, 2002). In another example, populations of Galapagos lava lizards (Microlophus albemarlensis) showed no evidence of genetic differentiation at six out of seven microsatellite loci, despite having differences in anti-predator behaviour that most likely reflect local adaptation (Jordan, Snell and Jordan, 2005). The anole and lava lizard studies provide just two of the numerous examples in which local adaptation has been maintained despite high levels of ongoing gene flow.
Since gene flow can impede local adaptation only if the migration rate (m) exceeds the strength of selection (s), it follows that, if gene flow is low, even relatively weak selection pressures can accelerate the divergence of two populations from one another. This, however, is only part of the story; we must also remember the potential contribution of genetic drift to population differentiation. So how do drift and selection interact? Key here is the size of populations (Ne). If selection is strong relative to the population size, i.e. if s is much greater than 1/(4Ne), then the effects of drift will be negligible. If, however, s is much less than 1/(4Ne), then changes in allele frequencies will be largely attributable to genetic drift, although mutations also are likely to make a contribution. The interactions of gene flow, drift and selection are summarized in Figure 4.13.
There are a number of reasons why we may wish to know whether drift or selection is a more important driver of population differentiation. For one thing,
Minimal population differentiation from genetic drift
Does population experience appreciable levels of gene flow?
Is migration (m) > strength of selection (s)?
Minimal population differentiation from selection
Populations will differentiate following selection
Was this article helpful?