explain how environmental factors can act as stabilising, disruptive and directional forces of natural selection

Natural and Artificial Selection (Cambridge A‑Level Biology 9700)

Learning objective

Explain how environmental factors can act as stabilising, disruptive or directional forces of natural selection, relate each force to the underlying genetic changes in a population, and connect these ideas to inheritance, population‑genetic equations and speciation.

Key definitions

  • Natural selection: differential survival and reproduction of individuals because of differences in phenotype.
  • Artificial selection: human‑directed breeding that favours particular phenotypes.
  • Stabilising selection: favours intermediate phenotypes; reduces phenotypic variance.
  • Disruptive selection: favours extreme phenotypes at both ends of the distribution; increases variance.
  • Directional selection: favours one extreme phenotype; shifts the population mean.
  • Fitness (w): the relative contribution of an individual genotype to the next generation (often expressed as the average number of surviving offspring).
  • Selection coefficient (s): the proportional reduction in fitness of a less‑favoured genotype (0 ≤ s ≤ 1).
  • Selection differential (S): difference between the mean trait value of the reproducing individuals and the mean of the whole population.
  • Selection gradient (β): change in mean phenotype per unit of phenotypic variance (β = S/σ²).
  • Heritability (h²): proportion of phenotypic variance that is additive genetic (h² = VA / VP).

From genotype to phenotype to fitness (link to Topic 16 *Inheritance*)

Selection acts on phenotypes, but phenotypes are produced by genotypes. The chain is:

  1. Genotype (alleles) → determines genotypic value (G) (additive, dominance, epistatic components).
  2. Genotype + environment (E) → produces the phenotype (P) (P = G + E).
  3. Phenotype determines the individual's fitness (w) in a given environment.

Thus a change in allele frequency (Δp) occurs only when the environment creates a fitness differential among genotypes.

Sources of genetic variation

  • Mutation: creates new alleles.
  • Genetic recombination (cross‑over and independent assortment) during meiosis: produces new allele combinations.
  • Gene flow (migration): introduces alleles from other populations, potentially counteracting local selection.
  • Genetic drift: random changes in allele frequencies, especially in small populations; can mask or reinforce selection.

How environmental factors generate the three classic modes of selection

1. Stabilising selection

When the prevailing environment favours the average phenotype and penalises extremes.

  • Environmental example: In a temperate forest, medium‑sized beaks in a finch species optimise seed handling; very small or very large beaks reduce feeding efficiency.
  • Genetic consequence: Alleles that produce extreme phenotypes are removed, reducing additive genetic variance (V_A ↓) while the population mean (μ) remains unchanged.
  • Population‑genetic expression: If fitnesses are w₁ = w₂ = 1 (intermediate genotype) and w₃ = 1 − s (extremes), the change in allele frequency is Δp = (pq s)/\(\bar w\) with s > 0, driving p toward the intermediate allele.

2. Disruptive selection

When the environment contains two or more distinct niches, each rewarding a different extreme phenotype.

  • Environmental example: A lake with surface‑feeding insects and deep‑water crustaceans. Small fish exploit the surface, large fish hunt deeper; medium‑sized fish are outcompeted.
  • Genetic consequence: Frequencies of alleles for the two extremes increase, producing a bimodal phenotypic distribution and raising variance (V_A ↑). If assortative mating evolves, reproductive isolation can follow (ecological speciation).
  • Population‑genetic expression: With two fitness peaks, w₁ = w₃ = 1 + s (extremes) and w₂ = 1 − t (intermediate), Δp is positive for both extreme alleles, leading to a split in the gene pool.

3. Directional selection

When a change in the environment makes one extreme phenotype more advantageous.

  • Environmental example: Rising average temperatures select for lighter‑coloured beetles that reflect more solar radiation.
  • Genetic consequence: Alleles coding for the favoured extreme increase in frequency; the mean phenotype shifts toward that extreme (μ → higher or lower).
  • Population‑genetic expression: If genotype AA has fitness w₁ = 1 + s, Aa has w₂ = 1 + hs, and aa has w₃ = 1, the allele‑frequency change per generation is

    \$\Delta p = \frac{p q \big[ w{AA} - w{aa} \big]}{\bar w}\$

    where \(\bar w\) is the mean fitness.

Quantifying selection

  • Selection differential (S):

    \$S = \mu{\text{selected}} - \mu{\text{population}}\$

  • Selection gradient (β):

    \$\beta = \frac{S}{\sigma^{2}}\$

    (change in mean phenotype per unit phenotypic variance).

  • Breeder’s equation (quantitative genetics):

    \$R = h^{2} S\$

    where R is the response to selection (change in mean phenotype in the next generation).

  • Allele‑frequency change for a single locus (biallelic, random mating, no other forces):

    \$\Delta p = \frac{p q s}{\bar w}\$

    with p and q the frequencies of the two alleles, s the selection coefficient against the less‑fit genotype, and \(\bar w\) the mean fitness.

Hardy–Weinberg equilibrium and the effect of selection

In the absence of evolutionary forces, genotype frequencies remain constant:

\$p^{2} + 2pq + q^{2} = 1\$

Natural (or artificial) selection changes the relative fitnesses (w₁, w₂, w₃) of the three genotypes, perturbing this equilibrium. For example, if:

  • AA: w₁ = 1
  • Aa: w₂ = 1 − s
  • aa: w₃ = 1 − 2s

the next generation’s allele frequencies are given by the Δp formula above, moving the population away from Hardy–Weinberg expectations.

Interaction with other evolutionary forces

  • Gene flow can introduce alleles that are either favoured or disfavoured by local selection, altering the effective selection coefficient.
  • Genetic drift can randomise allele frequencies, especially in small or isolated populations, potentially overriding weak selection.
  • When selection, gene flow and drift act simultaneously, the net change in allele frequency is:

    \$\Delta p = \underbrace{\frac{p q s}{\bar w}}{\text{selection}} \;+\; \underbrace{m(p{m} - p)}{\text{gene flow}} \;+\; \underbrace{\text{drift term}}{\text{stochastic}}\$

    where m is the migration rate and \(p_{m}\) the migrant allele frequency.

Evidence for natural selection (classic empirical studies)

StudyEnvironmental changeObserved selectionKey outcome
Peppered moth (Biston betularia)Industrial soot darkening tree barkDirectional selection for the melanic formRapid increase in melanic frequency; reversal after pollution control
Darwin’s finches (Galápagos)Fluctuating seed size & availabilityDirectional selection on beak length/sizeMeasurable shifts in beak morphology over a few generations
Antibiotic‑resistant bacteriaIntroduction of antibioticsDirectional selection for resistance genesPopulation becomes dominated by resistant strains
Pesticide‑resistant insectsWidespread use of organophosphatesDirectional selection for detoxifying enzymesLoss of control efficacy; need for new chemicals

Artificial selection – a controlled analogue of natural selection

Artificial selection follows the same genetic principles, but the selective pressure is imposed by humans.

Quantitative framework

For a breeding programme the expected response is given by the breeder’s equation:

\$R = h^{2} S\$

where S is the selection differential imposed by the breeder and is the trait’s heritability.

Illustrative breeding programme

  • Species: Dairy cattle (Bos taurus).
  • Target trait: Milk yield (kg per lactation).
  • Heritability: h² ≈ 0.30 (moderate).
  • Selection differential: Top 10 % of bulls produce offspring averaging 500 kg more than the population mean (S = 500 kg).
  • Expected response: R = 0.30 × 500 = 150 kg increase in mean milk yield per generation.
  • Selection index: Combines milk yield with correlated traits (e.g., fertility, disease resistance) using weighting coefficients to avoid undesirable side‑effects.

Other artificial‑selection examples

  • Stabilising: Breed Labrador Retrievers to a standard size and temperament – reduces variance around the breed standard.
  • Disruptive: Select wheat lines that flower either very early or very late to create a bimodal flowering‑time distribution, useful for staggered harvests.
  • Directional: Choose broiler chickens with rapid growth rates; alleles for muscle development rise in frequency, producing a selective sweep detectable in genome scans.

Speciation and disruptive selection

Disruptive selection can initiate ecological speciation when:

  1. Two extreme phenotypes exploit different niches.
  2. Assortative mating evolves (e.g., size‑based mate choice), reducing gene flow between the extremes.
  3. Genetic divergence accumulates, eventually leading to reproductive isolation (sympatric speciation) or, if geographic separation later occurs, allopatric speciation.

Comparison of the three selection types

Selection typeTypical environmental contextEffect on trait distributionGenetic consequenceTypical evolutionary outcome
StabilisingStable, homogeneous habitatSingle peak at intermediate values; variance ↓Loss of extreme alleles; maintenance of existing allelesRetention of a well‑adapted phenotype
DisruptiveHeterogeneous habitat with multiple nichesTwo peaks at extremes; variance ↑Increase in frequency of extreme alleles; possible assortative matingPopulation divergence; potential ecological speciation
DirectionalChanging environment that favours one extremeShift of whole distribution toward the favoured extremeAlleles for the favoured extreme rise in frequency (Δp > 0)Evolution of a new adaptive phenotype

Suggested diagram: three overlapping bell‑shaped curves illustrating stabilising, disruptive and directional selection acting on a continuous trait.

Key points to remember

  1. Environmental factors create selective pressures that can be stabilising, disruptive or directional.
  2. Selection acts on phenotypes, but phenotypic variation originates from genotypic variation (mutation, recombination, gene flow).
  3. Stabilising selection reduces genetic variance, disruptive selection increases variance and can lead to reproductive isolation, and directional selection shifts the mean phenotype.
  4. Quantitative measures – selection differential (S), selection gradient (β), heritability (h²) and the breeder’s equation (R = h²S) – allow the strength and response to selection to be calculated.
  5. Natural or artificial selection perturbs Hardy–Weinberg equilibrium; the magnitude of change depends on fitness differences (w₁, w₂, w₃) and the selection coefficient (s).
  6. Gene flow and genetic drift modify the outcome of selection, especially in small or fragmented populations.
  7. Classic empirical examples (peppered moth, Darwin’s finches, antibiotic resistance) illustrate natural selection in real time, while artificial‑selection programmes (dairy cattle, wheat, broiler chickens) demonstrate the same principles under human control.
  8. Disruptive selection can initiate speciation when divergent ecological niches promote assortative mating and reproductive isolation.