Selection & Evolution
Variation is the raw material of evolution. Individuals within a population differ in phenotype; some variation is heritable; heritable variation that improves survival and reproduction is selected for. Over many generations, allele frequencies shift — populations evolve. Given sufficient time and isolation, populations diverge until they can no longer interbreed: new species arise. Humans exploit the same mechanism through selective breeding, compressing evolutionary change from geological timescales to decades.
Variation
Members of the same species are not identical. Variation — differences in phenotype between individuals — arises from genetic and environmental sources. Only heritable (genetic) variation is passed to offspring and can be acted upon by natural selection.
Continuous and discontinuous variation
| Feature | Continuous variation | Discontinuous variation |
|---|---|---|
| Distribution | Normal distribution (bell curve); all values between the extremes possible | Discrete categories; no intermediates |
| Number of distinct phenotypes | Unlimited (a range) | Two or a few distinct classes |
| Genetic basis | Many genes (polygenic) and environmental modification | Usually one gene with two or a few alleles |
| Environmental influence | Significant — environment shapes the range | Little or none |
| Examples | Human height, skin colour, shoe size, milk yield in cattle | ABO blood groups, tongue-rolling ability, attached/free earlobes |
| Displayed on graph | Histogram with normal curve | Bar chart with separate bars |
Sources of variation
Random changes to DNA sequence (point mutations in Topic 6) or chromosome structure/number. The ultimate source of all new alleles — without mutation, all other mechanisms only reshuffle existing alleles. Mutations may be neutral, harmful, or rarely beneficial. The beneficial ones contribute to evolution when selected for.
Crossing over in prophase I creates new allele combinations on chromosomes. Independent assortment at metaphase I creates 2n possible chromosome combinations. Together with random fertilisation, these generate enormous genotypic diversity in offspring (Topic 16A).
Same genotype can produce different phenotypes in different environments. Examples: identical twins diverging in height depending on nutrition; plant height varying with soil quality; skin colour changing with sun exposure. Environmental variation is NOT heritable and cannot be selected for.
Phenotypic differences caused purely by environmental factors (e.g. a plant growing taller in fertile soil) cannot be inherited by offspring. Natural selection can only act on variation that is genetically encoded and passed to the next generation. Traits acquired during an organism's lifetime (Lamarckian evolution) are not passed on; this is why Lamarck's theory was discredited. The environment selects from existing genetic variation; it does not directly create heritable variation.
Using the t-test to compare variation between samples
The Student's t-test is used to determine whether the difference between the means of two samples is statistically significant. The 9700 syllabus requires you to apply the t-test and interpret the result. (LO 17.1.4)
- You have two groups with continuous data (e.g. leaf length, stem height, mass)
- You want to know whether the difference between the two means is likely a real biological difference, or just sampling chance
- Both samples should ideally be normally distributed; minimum n ≥ 5 per group recommended
- State the null hypothesis (H₀): “There is no significant difference between the means of [Group A] and [Group B].”
- Calculate t using the formula provided:
t = (x̄₁ − x̄₂) / √(s²₁/n₁ + s²₂/n₂)where x̄ = mean, s² = variance, n = sample size - Degrees of freedom (df): df = (n₁ − 1) + (n₂ − 1)
- Critical value: read from the t-table provided (at p = 0.05 with your df)
- Decision: if calculated t > critical value → reject H₀ (difference is significant, p < 0.05). If t < critical value → retain H₀ (difference not significant).
Pea plants grown in full light had a mean stem length of 42.3 cm (n = 15, s² = 18.4); plants grown in shade had a mean of 61.7 cm (n = 15, s² = 22.1). The calculated t-value is 12.4. The critical value at p = 0.05 with 28 df is 2.05.
(a) State the null hypothesis. [1]
(b) State whether the null hypothesis is accepted or rejected and justify your answer. [2]
(a) There is no significant difference between the mean stem length of pea plants grown in full light and those grown in shade. [1]
(b) The null hypothesis is rejected. The calculated t-value (12.4) exceeds the critical value (2.05) at p = 0.05, so the probability of obtaining this difference by chance alone is less than 5% — the difference is statistically significant. [2]
These concepts describe allele frequency changes due to random sampling in small populations. They are not core 9700 LOs but are sometimes taught as enrichment. For 9700 purposes, the primary mechanisms of allele frequency change are natural selection and mutation.
- Genetic drift: random change in allele frequency; more pronounced in small populations; not directional
- Founder effect: small subgroup with limited allele diversity establishes a new population
- Bottleneck: catastrophic reduction in population size; survivors are a random, reduced sample of original allele diversity
Useful context for understanding speciation, but do not present these as standard 9700 examination requirements.
Natural selection & speciation
Mechanism of natural selection
- Overproduction: organisms produce more offspring than can survive (populations have the potential for exponential growth)
- Variation: individuals within a population show heritable variation in phenotype
- Competition / struggle for existence: because resources (food, mates, space) are limited, individuals compete for survival; most offspring die before reproducing
- Differential survival and reproduction (selection): individuals with phenotypes better suited to the environment survive longer and reproduce more; they pass their advantageous alleles to offspring at higher frequency → allele frequencies change over generations
Over many generations, the accumulation of selected heritable changes produces evolution: gradual, heritable change in the characteristics of a population.
Three types of natural selection
Selection against both extremes of the phenotypic range. The intermediate phenotype is favoured. The mean of the population stays the same; variance decreases (the bell curve becomes narrower and taller).
Example: human birth weight. Very low birth weight (poor development) and very high birth weight (difficult delivery) are both disadvantageous; intermediate weights have the highest survival rate. The distribution has shifted little in the past centuries but has become more concentrated around the mean.
Selection favouring one extreme of the phenotypic range. The mean of the population shifts toward the favoured extreme; variance may decrease.
Example 1 — antibiotic resistance: before antibiotic use, bacteria with resistance mutations existed at low frequency. Antibiotic application (selection pressure) kills sensitive bacteria; resistant variants survive and reproduce. Over generations, resistant bacteria dominate. This is directional selection in action — and the most tested example in 9700 (synoptic with Topic 10).
Example 2 — Biston betularia (peppered moth): before industrialisation, pale moths were camouflaged on lichen-covered bark; dark (melanic) forms were rare. Industrial pollution killed lichens and darkened bark with soot; dark moths were now better camouflaged; pale moths were more visible to predators. Frequency of dark form increased in industrial areas. When clean-air laws reduced pollution, frequencies reversed.
Selection against the intermediate phenotype, favouring both extremes. The distribution splits into two peaks; variance increases; intermediate forms are rare.
Example: beak size in a bird species where both very small seeds (requiring small beaks) and large seeds (requiring large beaks) are abundant, but medium seeds are scarce. Birds with medium beaks are at a disadvantage; both extreme beak sizes are favoured.
Disruptive selection can drive a population toward two distinct subgroups and potentially contribute to speciation.
Speciation — how new species arise
A species is defined as a group of organisms that can interbreed to produce fertile offspring and are reproductively isolated from other groups. Speciation is the process by which one species splits into two or more distinct, reproductively isolated species.
- Geographic isolation: a physical barrier (mountain range, ocean, river, glacier) divides a population into two or more sub-populations that cannot interbreed
- Independent mutation and natural selection: each sub-population experiences different environments and different selective pressures; mutations occur independently in each; different alleles are selected for in each sub-population
- Genetic divergence: over many generations, allele frequencies in the two populations diverge until the gene pools are substantially different
- Reproductive isolation: eventually, individuals from the two populations would not be able to produce viable, fertile offspring even if they were reunited — they have become separate species
Reproductive isolation can be pre-zygotic (prevents mating or fertilisation) or post-zygotic (allows mating but offspring are not viable or fertile):
- Temporal isolation: the two populations breed at different times (seasons, times of day)
- Behavioural isolation: different courtship rituals, calls, or pheromones; neither population recognises the other's mating signals
- Mechanical isolation: incompatible anatomical structures prevent mating
- Post-zygotic isolation: hybrid offspring are inviable (die early) or infertile (e.g. mule = horse × donkey — viable but infertile)
Speciation can also occur without geographic isolation (sympatric speciation), e.g. through polyploidy in plants. When chromosome numbers double (allopolyploidy), the resulting individuals can no longer produce fertile offspring with the original species — instant reproductive isolation. This has been important in the evolution of many crop plants (wheat, cotton). The 9700 syllabus focuses on allopatric speciation as the main mechanism.
Antibiotic-resistant bacteria have become more common in hospitals over the past 50 years. Which type of selection best describes this change?
- A. Stabilising selection — the intermediate phenotype is favoured
- B. Directional selection — selection favouring resistance as a phenotypic extreme
- C. Disruptive selection — both extremes are favoured over intermediates
- D. Artificial selection — humans deliberately bred resistant bacteria
Describe how a new species could arise from a single ancestral population, using the concept of allopatric speciation. [8]
- A population of one species is divided into two sub-populations by a physical (geographic) barrier, e.g. a mountain range, sea, or river [1]
- The barrier prevents interbreeding (gene flow) between the two groups [1]
- Random mutations occur independently in each sub-population [1]
- Different environments (different climate, food sources, predators) exert different selection pressures on each sub-population [1]
- Different alleles are selected for (directional selection) in each sub-population; beneficial alleles increase in frequency in each group [1]
- Allele frequencies diverge over many generations as the gene pools change independently [1]
- Genetic drift may also change allele frequencies in small populations by chance [1]
- Reproductive isolation mechanisms accumulate: differences in behaviour, anatomy, physiology, or breeding season mean the two populations can no longer interbreed even if reunited [1]
- If reunited, hybrids would be non-viable or non-fertile (post-zygotic isolation) or mating simply would not occur (pre-zygotic isolation) [1]
- The two groups are now distinct species — they form separate, isolated gene pools [1]
Artificial selection
Artificial selection (selective breeding) is the deliberate selection by humans of individuals with desirable traits for breeding — exploiting the same mechanism as natural selection but directing it toward human-defined goals rather than survival and reproduction in the wild. It compresses evolutionary change from geological timescales to decades or generations.
Artificial selection — how it works
- Identify the desired trait (e.g. high milk yield, disease resistance, fast growth rate, drought tolerance)
- Select individuals from the existing population that show the strongest expression of that trait
- Allow only the selected individuals to breed together (and prevent breeding with non-selected individuals)
- Repeat over many generations — each generation, select again from the offspring
- Over time, allele frequencies shift toward alleles underlying the desired trait; the population's characteristics change
The underlying genetics is identical to natural selection: heritable variation exists, individuals with the favoured phenotype reproduce, allele frequencies shift. The difference is that humans define what is "favourable" rather than survival in a natural environment.
Examples of artificial selection
Cows with the highest milk yield are selected as breeding stock; their daughters tend to inherit high-yield genetics. Over decades of selective breeding, milk yield per cow has increased dramatically. Artificial insemination allows a single high-yield bull to sire thousands of offspring worldwide, rapidly spreading favoured alleles.
All domestic dog breeds derive from wolves. Thousands of years of selective breeding for different traits (speed in greyhounds, scent-tracking in bloodhounds, herding behaviour in collies) have produced extreme phenotypic diversity. Dogs also illustrate risks: extreme selection can cause health problems (e.g. breathing difficulties in brachycephalic breeds).
Wild grasses were selectively bred over thousands of years for higher seed yield, larger grains, shorter stems (to resist wind lodging), disease resistance, and drought tolerance. Modern wheat varieties would not survive in the wild without human cultivation — their seeds don't disperse naturally. This shows how far artificial selection can diverge from natural fitness.
All of the following are cultivars of the same species, Brassica oleracea, selected for different parts: broccoli (flower head), cauliflower (flower head), cabbage (leaves), kohlrabi (stem), kale (leaves), Brussels sprouts (lateral buds). Selective breeding of different structures from one ancestral plant demonstrates the power of artificial selection.
Natural vs artificial selection — comparison
| Feature | Natural selection | Artificial selection |
|---|---|---|
| Selection agent | Environment (predators, disease, climate, food availability) | Human breeder |
| Goal / criterion | Survival and reproductive success in the natural environment | A trait desired by humans (yield, appearance, temperament, etc.) |
| Speed | Slow: typically thousands to millions of generations | Fast: significant changes in decades or a few hundred generations |
| Direction | Unpredictable; changes with environment | Directed by human goals; consistent in one direction |
| Outcome | Increased fitness in natural environment; may lead to speciation | Increased expression of desired trait; may reduce fitness in nature |
| Genetic mechanism | Identical: heritable variation + differential reproduction → allele frequency change | Identical mechanism |
Inbreeding depression — a risk of artificial selection
Repeated selection from a small pool of related individuals increases homozygosity. When rare harmful recessive alleles are present in a population, inbreeding increases the chance of homozygous recessive offspring expressing deleterious traits. This is inbreeding depression:
- Reduced fertility, immune function, or viability in highly inbred strains
- Affects pedigree dog breeds, zoo populations, and highly selected crop varieties
- Countered by outcrossing (introducing individuals from a different breeding line) to restore heterozygosity
Topic 17 Practice — Comprehensive
Mixed practice across variation, natural selection, speciation, and artificial selection.
Human height shows a normal distribution in a population. Which best explains this pattern of variation?
- A. Height is controlled by one gene with two alleles, giving discrete height classes
- B. Height is polygenic (controlled by many genes) and also influenced by environment, producing a continuous normal distribution
- C. Height is entirely determined by the environment; the normal distribution reflects environmental variation only
- D. Height shows discontinuous variation because individuals either are tall or are short
Human birth weight data shows that babies with very low and very high birth weights have lower survival rates than babies of intermediate weight. Which type of selection does this illustrate?
- A. Stabilising selection — selection against both extremes, maintaining the intermediate
- B. Directional selection — selection favouring one extreme
- C. Disruptive selection — selection favouring both extremes
- D. Artificial selection — humans choose which babies survive
The frequency of antibiotic-resistant bacteria in hospitals has increased significantly over the past 50 years.
(a) Explain how resistance to an antibiotic arises and spreads through a bacterial population. Refer to mutation and natural selection. [5]
(b) Using Hardy-Weinberg principles, explain what the increase in resistant bacteria tells us about whether this bacterial population is in Hardy-Weinberg equilibrium. [2]
(c) Suggest TWO measures that could slow the spread of antibiotic resistance in hospitals. [2]
(a) Resistance via mutation and selection [5 marks]
- Random mutations in bacterial DNA occasionally produce variants with reduced sensitivity to the antibiotic (e.g. a mutation produces an enzyme that inactivates the antibiotic, or alters the antibiotic's target) [1]
- These mutations occur before antibiotic exposure — they are pre-existing variation, not induced by the antibiotic [1]
- When the antibiotic is applied, sensitive bacteria are killed; resistant variants survive and reproduce (differential reproduction) [1]
- The antibiotic acts as a selection pressure; resistant alleles increase in frequency in the population over successive generations [1]
- Resistance genes can also spread horizontally via plasmids to other bacteria, including unrelated species — accelerating the spread beyond vertical (parent-to-offspring) transmission [1]
(b) Hardy-Weinberg equilibrium [2 marks]
- Hardy-Weinberg equilibrium requires no natural selection; the population would be in equilibrium if all genotypes survived equally well [1]
- The increase in resistant bacteria shows that allele frequencies are changing — the H-W assumption of no selection is violated; this population is not in H-W equilibrium; it is evolving under directional selection [1]
(c) Two measures [2 marks]
- Prescribe antibiotics only when genuinely needed (bacterial infection confirmed) — reducing unnecessary selection pressure [1]
- Strict infection control (hand hygiene, isolating infected patients) — reduces transmission of resistant strains between patients [1]
- Complete prescribed antibiotic courses — ensures all bacteria are killed and does not leave a partially-resistant population [1]
- Restrict agricultural use of antibiotics [1]
Dairy cattle have been selectively bred for high milk yield over many generations.
(a) Outline the process of artificial selection used to increase milk yield. [3]
(b) Compare artificial selection with natural selection, giving TWO similarities and TWO differences. [4]
(a) Artificial selection for milk yield [3 marks]
- From the herd, identify cows with the highest milk yield (the desired phenotype) [1]
- Allow only these high-yield cows (and bulls whose daughters have high yield) to breed; prevent lower-yield individuals from breeding [1]
- Repeat selection and breeding over successive generations; milk yield gradually increases as alleles contributing to high yield become more frequent in the population [1]
(b) Similarities and differences [4 marks; 1 each]
Similarities (any two):
- Both rely on heritable variation in the population [1]
- Both work by differential reproduction: individuals with the favoured phenotype produce more offspring [1]
- Both result in allele frequency changes over generations [1]
Differences (any two):
- The selection agent: artificial selection = human breeders choosing specific traits; natural selection = environment selecting for survival and reproduction [1]
- Speed: artificial selection is much faster (decades vs thousands of generations) [1]
- Goal: artificial selection targets traits useful to humans (which may reduce fitness in the wild); natural selection increases fitness in the natural environment [1]
Topic 17 — Common Mistakes
- 🌿Saying the antibiotic "causes" resistance mutationsThis is Lamarckian thinking — the classic mistake. Antibiotic resistance mutations exist before antibiotic exposure. The antibiotic acts as a selection pressure that allows pre-existing resistant variants to survive while sensitive ones are killed. The antibiotic selects; it does not create. This is consistently the most-tested evolution concept in biology.
- 📊Confusing stabilising, directional, and disruptive selectionStabilising = against both extremes (variance ↓, mean unchanged). Directional = favouring one extreme (mean shifts, variance may ↓). Disruptive = against intermediate (variance ↑, population splits). The key: which phenotypes have higher reproductive success?
- 👥Saying environmental variation is heritable and can be selected forEnvironmental variation is NOT heritable. A plant that grows taller in fertile soil does not pass "tallness genes" to its offspring; it passes its existing genotype. Natural selection acts only on genetic variation. Saying "selection changes phenotype directly" without noting it must be genetically based loses marks.
- 🌐Confusing allopatric and sympatric speciationAllopatric speciation requires geographic separation (the main mechanism covered by 9700). Sympatric speciation occurs in the same geographic area (e.g. polyploidy in plants). Don't mix up the terms. The 9700 exam primarily tests allopatric speciation.
- 🔍Saying species are defined by appearanceSpecies are defined by reproductive isolation — the ability to interbreed and produce fertile offspring. Two populations may look very similar (cryptic species) but be reproductively isolated; or they may look different but still interbreed (e.g. ring species). Appearance alone is not the definition.
- 🌟Saying artificial selection is "unnatural" and therefore fundamentally differentThe underlying genetic mechanism is identical to natural selection: heritable variation exists, differential reproduction occurs, allele frequencies change. The only difference is the selection agent (human vs environment) and the criterion of fitness. Don't say "artificial selection doesn't involve natural allele frequencies" or "it creates new genes" — it does neither.
- 💥Saying "survival of the fittest" means the physically strongest survive"Fitness" in biology means reproductive success — the ability to survive and pass alleles to the next generation. The "fittest" individual is not necessarily the strongest or fastest; it is the one whose phenotype best matches the current environment and allows it to leave the most offspring. A camouflaged but slow organism can have higher fitness than a fast but visible one.
- ⚠Forgetting that speciation requires reproductive isolation, not just genetic divergenceTwo populations that have genetically diverged are not yet separate species unless they are reproductively isolated (cannot produce fertile offspring). Reproductive isolation is the defining criterion, not merely genetic difference. Allopatric speciation leads to speciation only when reproductive isolation mechanisms have accumulated to the point that gene flow between the groups is impossible.
Topic 17 is conceptually straightforward but requires precise language. Highest-yield items: four principles of natural selection (overproduction / variation / competition / differential reproduction), three selection types with correct distribution changes (stabilising = both extremes removed / directional = mean shifts / disruptive = intermediate removed), antibiotic resistance as directional selection with "mutations pre-exist, antibiotic selects", allopatric speciation 5-step sequence (geographic isolation → independent mutation → different selection → genetic divergence → reproductive isolation), reproductive isolation mechanisms, artificial selection procedure (identify → breed selected → repeat), 6-row natural vs artificial comparison, inbreeding depression. Synoptic links: Topic 10 (antibiotic resistance = lived example of directional selection), Topic 16A (crossing over + independent assortment = genetic variation), Topic 16B (H-W violation = selection occurring), Topic 6 (mutation = source of variation).