Selection and Evolution is Cambridge International Biology 9700 Topic 17. It connects genetic and environmental variation to statistical comparison, natural and artificial selection, allele-frequency change, Hardy-Weinberg calculations, DNA-sequence evidence and speciation. Theory owns population mechanisms, calculations and inference; measurement design and sample collection remain in the dedicated practical hub.
1. Sources of phenotypic variation
Phenotypic variation can arise from genetic factors, environmental factors or their interaction. Genetic differences include alleles, mutation, meiosis and fertilisation. Environmental effects include nutrition, temperature, light, disease, activity and experience.
Some traits are largely genetic, such as an ABO blood group. Others are strongly environmental within a genotype, such as a plant's growth under different mineral supplies. Many traits, including height and body mass, reflect both inherited potential and environment.
An environmental effect on phenotype does not automatically alter inherited DNA. A genetic predisposition also does not guarantee one phenotype under every environment.
2. Discontinuous and continuous variation
Discontinuous variation places individuals into distinct categories with no intermediates, such as ABO blood groups. It is commonly controlled by one or a few genes, so allele combinations produce separate outcomes. Environmental influence is often relatively limited.
Continuous variation spans a range with many intermediate values, such as height. It is usually polygenic: many loci each contribute a small effect. Environmental factors further broaden the distribution. Large samples commonly form an approximately normal distribution when many small influences combine.
The distinction concerns the observed pattern and genetic basis. Measuring a discontinuous category with numbers does not turn it into continuous variation.
3. Comparing two sample means with a t-test
A t-test evaluates whether the difference between two sample means is large relative to within-sample variation and sample size. Begin with a null hypothesis that the population means do not differ and any observed difference is due to chance sampling.
Choose the specified test for paired or independent samples and check its sampling and distribution assumptions. Equal-variance and unequal-variance tests use different calculations. Use the appropriate formula and degrees of freedom. For a two-sided test, reject the null hypothesis when the absolute t value exceeds the positive critical value for the chosen significance level, commonly 0.05. For a one-sided test, use the stated direction and its corresponding critical region.
Check this topic from memory
Attempt the matching topic bank before reopening the notes. Use each missed idea to decide what to review next.
A significant result supports a difference under the test conditions; it does not prove the tested factor alone caused it. Sampling, controls and confounding variables still matter. Theory owns the inference, while reliable measurement and sampling remain practical responsibilities.
4. Natural selection
Populations can produce more offspring than available resources can support. Individuals therefore compete in a struggle for existence. Heritable phenotypic variation means some individuals are better suited to the current environment.
Better-adapted individuals are more likely to survive and reproduce. They pass their alleles to offspring, so advantageous alleles tend to become more frequent over generations. Chance can still affect their fate, especially in small populations. Populations evolve; individual organisms do not change their inherited alleles because they need to adapt.
Fitness is relative reproductive success in a stated environment. A phenotype advantageous in one environment may be disadvantageous when conditions change.
5. Stabilising, directional and disruptive selection
Stabilising selection favours intermediate phenotypes and acts against extremes. The mean may remain similar while variation narrows. It occurs when established conditions make the existing intermediate well suited.
Directional selection favours one phenotypic extreme. The distribution mean shifts over generations, as when a changed environment consistently favours greater or smaller values.
Disruptive selection favours both extremes over intermediates. Variation can become bimodal. If extremes also become reproductively isolated, disruptive selection can contribute to divergence, but selection alone does not automatically establish new species.
6. Genetic drift, founder and bottleneck effects
Genetic drift is random change in allele frequencies due to chance sampling of which individuals reproduce. It is strongest in small populations, where each individual represents a larger fraction of the gene pool.
The founder effect occurs when a small number of individuals establish a new population. Their allele frequencies may differ by chance from the source population, and rare alleles can become unusually common or absent.
A bottleneck effect follows a sharp population reduction. The survivors carry only a sample of the original variation. Even if population size recovers, genetic diversity can remain low. The drift component is random with respect to fitness. A population reduction or founding event can also occur alongside selection; not every survival difference during such an event is random.
7. Antibiotic resistance as natural selection
A bacterial population contains heritable variation due to mutation and gene acquisition. An antibiotic kills or inhibits susceptible bacteria. Resistant bacteria survive, reproduce and pass resistance vertically or transfer genes horizontally.
Antibiotic exposure can favour resistant variants over susceptible bacteria, increasing the frequency of resistance genes. Unnecessary antibiotic use adds selection pressure, while transmission and gene transfer can spread resistance.
The antibiotic does not instruct bacteria to mutate in the needed direction. Resistance variants arise before or during population growth, and exposure selects them.
8. Hardy-Weinberg principle
The Hardy-Weinberg principle relates allele and expected genotype frequencies under an idealised inheritance model. Here, consider a diploid population and an autosomal locus with two alleles, A and a. Their frequencies p and q sum to 1. Under the model, AA, Aa and aa have expected frequencies p squared, 2pq and q squared respectively.
Assume random mating at this locus, negligible drift in a very large population, and no selection, mutation or migration changing its allele frequencies. Non-random mating can change genotype proportions without changing allele frequencies. Observed departures require further investigation, including sampling and measurement error; they do not identify a particular evolutionary force by themselves.
With complete recessivity, full penetrance and reliable phenotype classification, the recessive phenotype identifies aa. Under the model, its frequency estimates q squared. Take the square root to estimate q, obtain p from 1 minus q, and calculate expected heterozygotes as 2pq. These steps depend on the stated assumptions.
9. Principles of selective breeding
Artificial selection begins by identifying individuals with desirable heritable phenotypes. Breeders select them as parents, cross them, evaluate offspring and repeatedly choose the best descendants. Controlled mating increases frequencies of desired alleles over generations.
Selection can reduce genetic diversity and increase harmful recessive homozygosity. Breeding therefore balances gain in a target trait against fertility, disease vulnerability and other performance traits.
The environment still affects measured phenotype, so reliable records and comparable conditions are needed to identify genetic merit.
10. Disease resistance in wheat and rice
Breeders can cross a productive variety with a line carrying resistance to a pathogen. Offspring are screened for resistance and desirable yield or quality traits. Repeated selection and crossing combine resistance alleles with the commercial background.
Resistance can reduce crop loss and the need for pesticide treatment, but pathogen populations can evolve. Maintaining multiple resistance sources and genetic diversity can make protection more durable than relying on one genotype everywhere.
11. Inbreeding, hybridisation and maize
Repeated self-fertilisation or close breeding produces homozygous inbred maize lines. Each line becomes genetically uniform but can lose vigour as deleterious recessive alleles become homozygous.
Crossing sufficiently homozygous selected inbred lines produces genetically uniform F1 hybrid seed. Heterozygosity can restore or increase vigour, called hybrid vigour, although not every cross is superior. To reproduce the intended uniform F1, repeat the defined parental cross: saved F2 seed segregates and does not retain the same uniformity.
Inbreeding creates stable parental lines; hybridisation combines them. These steps have different purposes and should not be treated as synonyms.
12. Improving dairy-cattle milk yield
Cattle with high milk yield and other desirable records are selected as parents. Bulls can be evaluated through performance of many daughters because males do not express milk yield directly. Genomic predictions can also inform selection of young bulls before they have milking daughters. Artificial insemination can distribute semen from selected males widely.
Repeated selection increases favourable alleles, but breeders should also monitor health, fertility, lifespan and inbreeding. A high-yield phenotype reflects feeding and management as well as genotype, so comparisons require standardised records.
13. Evolution and changing gene pools
Evolution is change in the genetic composition of populations across generations. Selection, mutation, migration and genetic drift alter gene pools. Accumulated changes can cause populations to diverge and eventually form new species from pre-existing species.
Evolution has no planned endpoint. It describes population change under historical environments, chance and inheritance. Modern species are not necessarily steps toward one predetermined ideal.
14. DNA sequence evidence
Homologous DNA sequences can be aligned between species. A greater proportion of matching bases usually supports a more recent common ancestor, while more accumulated differences generally indicate longer divergence.
Multiple genes or genomic regions provide stronger evidence than one short sequence. Mutation rates vary, selection acts differently across genes, and horizontal transfer or incomplete lineage sorting can complicate one locus.
Sequence evidence should be combined with other data and interpreted through a model of inheritance and change. Similarity supports relationship; it does not mean one living species directly descended from the other.
15. Allopatric speciation
In allopatric speciation, a geographical barrier divides a population and reduces gene flow. Different mutations, selection pressures and drift change allele frequencies independently.
Over many generations, genetic differences accumulate. If the populations become reproductively isolated and can no longer exchange genes successfully when contact resumes, they are separate species under the biological species concept.
Geographical separation starts the process but does not by itself prove speciation. Reproductive isolation is the key outcome.
16. Sympatric speciation
Sympatric speciation occurs without a geographical barrier. Ecological separation can arise when groups use different habitats, hosts or resources in the same region. Behavioural separation can occur through different courtship signals or mating times.
Assortative mating reduces gene flow between groups. Selection, mutation and drift then permit divergence. When reproductive isolation becomes established, separate species can form.
This overview focuses on ecological and behavioural separation, without establishing current assessment boundaries. A complete explanation must connect separation to reduced gene flow and divergent gene pools.
Worked application: Hardy-Weinberg calculation and its limits
In a hypothetical sample from a large diploid population, 9 percent show a fully penetrant recessive phenotype at a two-allele autosomal locus. Assume reliable classification and the Hardy-Weinberg conditions. Treat 0.09 as an estimate of aa frequency: q squared is 0.09, q is 0.30 and p is 0.70. The expected heterozygote frequency is 2 multiplied by 0.70 multiplied by 0.30, giving 0.42 or 42 percent. This conditional estimate does not establish equilibrium. Non-random mating can alter genotype proportions without changing allele frequencies. Evidence of allele-frequency change would still need to distinguish selection, drift, migration and mutation, with sampling uncertainty considered.
Common misconceptions and corrections
Calling all phenotypic variation genetic. Environment and interaction can contribute.
Calling continuous variation a set of categories. It spans a range.
Saying one gene usually controls continuous variation. It is commonly polygenic.
Using a t-test to prove causation. It tests evidence for a mean difference.
Saying non-significant means identical populations. Evidence was insufficient at the threshold.
Saying individuals evolve during their lives. Populations change across generations.
Calling the environment's action purposeful. Selection has no foresight.
Saying stabilising selection shifts the mean to one extreme. Directional selection does.
Saying disruptive selection favours intermediates. It favours extremes.
Calling genetic drift non-random adaptation. It is chance allele-frequency change.
Saying drift is strongest in huge populations. Its effect is stronger in small ones.
Calling founder and bottleneck effects the same event. One establishes a population; one sharply reduces it.
Saying antibiotics create directed resistance mutations. They select heritable variants.
Using Hardy-Weinberg without checking assumptions. The model has five key conditions.
Equating q with q squared. q is an allele frequency; q squared is the expected aa frequency under the model.
Calling 2pq the dominant phenotype frequency. It is the expected heterozygote frequency under the model.
Saying selective breeding has no diversity cost. Repeated selection can narrow variation.
Calling inbreeding hybridisation. They serve distinct maize-breeding steps.
Selecting dairy bulls from their own milk yield. Use genetic evaluations informed by relatives, progeny and/or genomic data.
Calling evolution progress toward perfection. It is gene-pool change over time.
Saying identical sequence means no evolutionary relationship needs inference. Evidence still needs context.
Saying one sequence always gives the true species tree. Loci can differ.
Calling geographical separation complete speciation. Reproductive isolation must develop.
Saying sympatric speciation requires a physical barrier. Ecological or behavioural separation can reduce gene flow.
Assessment guidance
Separate the source of variation from the shape of its distribution. A t-test answer needs a null hypothesis, correct critical comparison and a bounded conclusion. Natural-selection explanations should track heritable variation, competition, differential reproduction and allele-frequency change; keep drift explicitly random. For Hardy-Weinberg calculations, state assumptions and distinguish allele, genotype and phenotype frequencies. Artificial-selection answers should name selection, crossing and repeated evaluation, then address trade-offs. Evolutionary evidence should compare homologous sequence data with limitations. Speciation answers must connect the isolating mechanism to reduced gene flow, divergence and reproductive isolation.
Retrieval practice
Classify examples as genetic, environmental or interactive and as continuous or discontinuous. Complete a t-test decision from supplied statistics. Draw stabilising, directional and disruptive selection outcomes and compare them with founder, bottleneck and drift effects. Solve three Hardy-Weinberg problems after checking assumptions. Reconstruct the three named breeding programmes. Finish by evaluating a DNA alignment and writing parallel causal chains for allopatric and sympatric speciation.
