11.2 Selection, Gene Pools and Hardy–Weinberg

Stabilizing selection favors intermediate phenotypes and selects against both extremes, reducing variation around an established optimum. Directional selection favors one end of a distribution and shifts the population mean, often after environmental change. Disruptive selection favors both extremes over intermediates and can produce a bimodal distribution. These patterns describe reproductive success across phenotypes; they do not prove that a trait is controlled by one gene.

In artificial selection, humans choose which individuals reproduce because they express desired traits. Repeating that choice changes allele frequencies and has produced crop varieties, livestock breeds and domesticated forms that differ greatly from their wild ancestors. It demonstrates that substantial inherited change can accumulate over many generations, although the selecting agent is human preference rather than the unmanaged environment.

Strong selection from a limited breeding stock can narrow a gene pool. Desired alleles may be linked to harmful variants, and mating relatives raises homozygosity, exposing deleterious recessive alleles. Inbreeding depression is reduced population fitness associated with low diversity and increased homozygosity. Introducing unrelated individuals can restore alleles and improve breeding success, but only if a suitable external population remains.

A gene pool consists of all alleles present in an interbreeding population. Immigration can introduce alleles and emigration can remove them; together these movements cause gene flow. Gene flow often makes populations more similar because migrants reproduce across their former boundary. Mutation adds novel alleles, selection changes contributions non-randomly, and genetic drift changes frequencies through chance sampling.

Genetic drift is strongest in small populations. If only a few individuals survive a disaster, their alleles may be an unrepresentative sample of the original population: this is a bottleneck. If a few colonists establish a new population, their sample creates a founder effect. A rare allele can become common or disappear without improving adaptation. Drift can fix one allele and eliminate alternatives, reducing variation and causing isolated populations to diverge.

Selection and drift may operate simultaneously. An advantageous allele is more likely to increase, but when it is rare in a small population it can still disappear by chance before its advantage has much effect. Conversely, a neutral or mildly harmful allele may rise after a bottleneck. Explanations should match the evidence: consistent change associated with a measured pressure supports selection; unpredictable change among replicate small populations supports drift.

For a locus with two alleles, let p be the frequency of allele A and q the frequency of allele a. Because these are the only alternatives, p + q = 1. Under random mating, gamete union gives expected genotype frequencies p² for AA, 2pq for Aa and q² for aa. Their sum is one. The equation predicts genotype frequencies from allele frequencies; it does not state that dominant alleles must be common.

If a recessive phenotype is fully penetrant, its population frequency equals q² because only aa individuals express it. Take the square root to find q, calculate p = 1 − q, then use 2pq for the heterozygote frequency. A common error is to call the recessive phenotype frequency q. Another is to infer allele frequency directly from the dominant phenotype, which combines AA and Aa and therefore equals p² + 2pq.

Hardy–Weinberg equilibrium assumes a large population, random mating, no selection, no relevant mutation and no migration. It also assumes ordinary diploid inheritance for the modeled locus. Real populations rarely satisfy every condition perfectly; that is why the model is useful. A statistically meaningful departure between observed and expected genotype frequencies signals that one or more evolutionary or population-structure processes may be operating.