Evolution and Natural Selection

Parent hub: Cell Biology, Genetics & Evolution | Prerequisites: Cell Division, Inheritance and Genetic Variation, Cell Organelles, Microscopy and Cell Size | Next logical topic (separate): Ecology and Ecosystems

Quick Definition

Evolution is change in the inherited characteristics of a population across generations. Natural selection is one mechanism of evolution: when heritable differences affect survival and reproduction, the variants that leave more descendants become more common in later generations. Evolution happens to populations, not to single individuals in their lifetimes.

Why It Matters

Evolution explains both unity and diversity in life: why all living things share DNA, cells and a common chemistry, yet differ so widely in form and behaviour. It also explains urgent modern problems — why bacteria become resistant to antibiotics, why insects become resistant to pesticides, and why a vaccine or drug that works today may need updating. If you understand selection, drift and gene flow, you can read those headlines as population changes, not as individual organisms "deciding" to change.

Learning Objectives

Define evolution, natural selection, adaptation and fitness in population terms.
Explain the four requirements for natural selection and apply them to an example.
Interpret a worked selection shift and a Hardy-Weinberg calculation, and state what each does and does not show.
Distinguish directional, stabilising and disruptive selection from graphs.
Summarise the main lines of evidence for evolution and what each line can and cannot prove on its own.
Explain speciation, genetic drift and gene flow, and how they differ from natural selection.
Identify common misconceptions about evolution, including Lamarckian inheritance and "evolution as progress".

Key Terms

Population — a group of individuals of the same species living in one place that can interbreed. Gene pool — all the alleles present in a population. Allele frequency — the proportion of all copies of a gene in a population that are a particular allele. Evolution — a change in allele frequencies (or in inherited characteristics) in a population over generations. Natural selection — differential survival and reproduction caused by heritable differences. Fitness — in evolution, an individual's relative contribution of genes to the next generation, not strength, size or health in everyday language. Adaptation — an inherited characteristic, shaped by selection, that improves fitness in a particular environment. Speciation — the formation of a new species, usually when populations become reproductively isolated. Genetic drift — random change in allele frequencies, strongest in small populations. Gene flow — movement of alleles between populations, for example by migration, pollen or seed dispersal.

Variation: The Raw Material

Natural selection cannot create variation; it can only sort what already exists. That variation comes from mutation, which creates new alleles, and from meiosis and fertilisation, which shuffle existing alleles into new combinations. The details of genes, alleles, crosses, linkage and mutation are taught in Inheritance and Genetic Variation, and are not re-taught here beyond that link: without heritable variation, selection has nothing to act on, and no evolution by selection can occur. An environmental effect on one individual's body — a tan, a muscle built by exercise, a leaf grown larger in shade — is not inherited unless it changes the DNA passed to offspring. That distinction is the whole difference between selection and the older Lamarckian idea discussed below.

Natural Selection in Four Steps

Four steps of natural selection: variation, heritable differences, differential survival and reproduction, and change in population frequencies — Figure 1. The four requirements for natural selection, in order: variation, heritability, differential reproduction, and a change in the population over generations.

Natural selection needs four conditions, and exam answers should name all four rather than writing only "survival of the fittest":

Variation exists in the population for a trait.
The variation is heritable — at least partly caused by genes passed to offspring.
More offspring are produced than the environment can support, so individuals compete for food, mates, space or safety, and not all survive or reproduce equally.
Differential reproduction follows — individuals with variants better suited to the current environment leave, on average, more descendants, so their alleles rise in frequency.

Figure 1 shows the sequence. Note the order: variation first, selection second, frequency change third. Selection does not cause the useful mutation to appear; it spreads a variant that is already present.

Fitness and Differential Reproduction

Fitness is measured in descendants, not in dominance displays. A large, strong animal that leaves no offspring has, in evolutionary terms, a fitness of zero for that generation. A smaller animal that leaves many surviving offspring has higher fitness in that environment. Fitness is also always relative to an environment: a pale moth that is well camouflaged on pale bark is poorly camouflaged on dark, soot-covered bark. Change the environment and the ranking of fitness can reverse — which is why a trait that is "good" in one place or time is not good in all places and times.

Differential reproduction is the key phrase. Selection acts through who reproduces, not only through who dies. A variant that makes an individual live longer but reproduce less may still lose, while a variant that shortens life but increases early reproduction may spread. That is why the definition used here always says "survival and reproduction", never survival alone.

Worked Example: A Simplified Selection Shift

This is a simplified, illustrative calculation, not data from a real population. Imagine a beetle population of 1,000 individuals on a light-coloured surface: 500 light beetles and 500 dark beetles, with colour heritable. Birds prey on them, and dark beetles stand out on the light surface.

Before predation: light 500 (50%), dark 500 (50%).
Survival: on a light surface, light beetles are camouflaged, so 80% of light beetles survive (500 × 0.80 = 400) and only 20% of dark beetles survive (500 × 0.20 = 100).
Survivors: 500 in total, of which light beetles are 400 out of 500 = 80% and dark beetles are 100 out of 500 = 20%.

The light phenotype has risen from 50% to 80% among survivors. If colour is heritable and survivors reproduce in proportion to their numbers, the next generation will contain a higher proportion of light beetles than the starting generation did. That shift — a change in the inherited composition of the population — is evolution by natural selection in miniature. Real selection is rarely this clean: survival is never 80% versus 20% for a single trait, many genes and the environment also affect the outcome, and one generation's shift is usually small.

Adaptation: What It Is and Is Not

An adaptation is an inherited feature that improves fitness in a particular environment because past selection favoured it: the thick fur of a polar bear in cold climates, the long neck that lets a giraffe reach high leaves, the shape of a finch beak that cracks particular seeds. Three cautions matter. First, an adaptation is good enough, not perfect: the human back, eye and birth canal all carry compromises from earlier evolutionary history. Second, an adaptation is tied to an environment: move the organism and the same feature may become a disadvantage. Third, not every feature is an adaptation: some features are by-products of other traits, some are leftovers of ancestry, and some persist by drift rather than by selection. Saying "this exists, therefore it must be adaptive" is a story, not evidence; the evidence is a demonstrated fitness advantage in that environment.

Types of Selection

Three schematic curves showing directional selection shifting the mean, stabilising selection narrowing variation, and disruptive selection favouring both extremes — Figure 2. Directional, stabilising and disruptive selection (schematic curves, not measured data): directional shifts the mean, stabilising narrows variation, disruptive favours both extremes.

Selection can act on a trait's distribution in three main ways, shown in Figure 2. All three are schematic curves, not measured data.

Directional selection favours one extreme: for example, larger body size when larger individuals survive cold better, or earlier flowering when a drought arrives sooner. The population mean shifts toward that extreme.
Stabilising selection favours the intermediate: for example, human birth weight, where very small and very large babies have historically had lower survival than intermediate babies. The mean stays roughly the same but variation narrows.
Disruptive selection favours both extremes over the intermediate: for example, finches with either large, strong beaks for large seeds or small, delicate beaks for small seeds, when medium seeds are scarce. Over time this can increase variation and, if the extremes stop interbreeding, contribute to speciation.

A fourth pattern, balancing selection, maintains more than one variant — for example, when heterozygotes have higher fitness than either homozygote, as in the commonly cited sickle-cell example in malaria regions. It is worth naming, but it should not be confused with the three distribution-shaping types above.

Evidence for Evolution

Fossil layers with older forms below and younger forms above, beside a branching tree from a common ancestor to two species — Figure 3. Two lines of evidence: fossil order in rock layers (older below, younger above) and branching from a common ancestor, with shared traits inherited.

No single line of evidence carries the whole case; the strength is that independent lines point the same way. Figure 3 shows two of them, fossil order and common ancestry.

Fossil record. Fossils appear in an order consistent with descent over time: simpler forms in older rocks, more recent forms in younger rocks, and transitional forms — such as early whales with hind-limb bones, or feathered dinosaurs — that combine features of earlier and later groups. The record is incomplete, because fossilisation is rare, so gaps alone do not overturn the pattern.
Comparative anatomy. Homologous structures share a basic plan from a common ancestor but serve different functions: the forelimbs of a human, a bat, a whale and a horse all contain the same set of bones arranged in the same pattern, used for grasping, flying, swimming and running. Analogous structures, such as bird and insect wings, look similar because they do similar jobs but have different origins; they are evidence of convergent evolution, not of close ancestry. Vestigial structures, such as the human appendix or the pelvic bones of whales, are reduced versions of structures that were functional in ancestors.
Molecular evidence. All living things use DNA, the same genetic code (with rare exceptions) and very similar core biochemistry. Species thought to be close relatives from anatomy also have more similar DNA and protein sequences than distantly related species. Shared "mistakes" are especially telling: for example, humans and other primates share the same broken gene for making vitamin C, which is most simply explained by inheritance from a common ancestor that already carried the broken copy.
Biogeography. Species on islands often resemble species on the nearest mainland more than species in similar climates elsewhere, as Darwin noted in the Galápagos. Marsupials are diverse in Australia, where they evolved in relative isolation, but rare elsewhere. Distribution patterns fit history and dispersal, not only climate.
Observed selection. Evolution has been watched in real time: bacteria evolving antibiotic resistance in hospitals and laboratories, insects evolving pesticide resistance, and finch beak sizes shifting after droughts in the Galápagos. These are small-scale changes, but they demonstrate the mechanism operating, which is what the larger claim requires.

Speciation

Speciation is the origin of a new species, and the usual definition of species used in exams is the biological species concept: a group of populations that can interbreed and produce fertile offspring, reproductively isolated from other such groups. The commonest route is allopatric speciation: a population is split by a geographical barrier — a river, a mountain range, a sea — and the two parts evolve independently under different selection, drift and mutation until they can no longer interbreed successfully even if they meet again. Sympatric speciation, without a geographical barrier, is rarer but documented, for example when different food sources or mating times split one population, or when polyploidy in plants creates instant reproductive isolation. Reproductive isolation can be prezygotic (different mating seasons, behaviours or gamete incompatibility, so mating or fertilisation fails) or postzygotic (hybrids are inviable or sterile, such as a mule). Speciation is usually gradual, but the fossil record sometimes shows long periods of little change punctuated by faster change, a pattern called punctuated equilibrium; it is still evolution by the same mechanisms, at different rates, not a different mechanism.

Genetic Drift and Gene Flow

Not all evolution is adaptive. Genetic drift is random change in allele frequencies from generation to generation, caused by chance in which individuals reproduce. It is weak in large populations and strong in small ones, where it can fix or eliminate an allele with no adaptive advantage at all. Two special cases appear in exams: the bottleneck effect, where a disaster sharply reduces a population so survivors carry only a random sample of the original alleles (cheetahs are commonly cited as a species with very low genetic diversity after a past bottleneck), and the founder effect, where a few colonists found a new population with allele frequencies unlike the source population. Gene flow is the opposite force in one sense: migration, pollen or seed movement carries alleles between populations, introducing variation and reducing differences between them, though it can also swamp local adaptation if incoming alleles are poorly suited to local conditions. Drift and gene flow change frequencies without selection, which is why a frequency change alone is not proof of selection.

Hardy-Weinberg: A Null Model

Bar chart of Hardy-Weinberg expected genotype frequencies for p equals 0.80 and q equals 0.20: AA 64 percent, Aa 32 percent, aa 4 percent — Figure 4. Hardy-Weinberg expectation for p = 0.80 and q = 0.20: 64% AA, 32% Aa and 4% aa — a worked example, not data from a real population.

The Hardy-Weinberg principle gives the genotype frequencies expected if a population is NOT evolving at a locus: no selection, no mutation, no gene flow, random mating, and an effectively infinite population so drift is negligible. It is a null model — a baseline to compare real data against, not a description of most real populations. For two alleles A and a with frequencies p and q, p + q = 1, and the expected genotype frequencies are p² (AA) + 2pq (Aa) + q² (aa) = 1.

Worked example (different numbers from the plant population-genetics page, which uses p = 0.60). Suppose p = 0.80 for allele A and q = 0.20 for allele a. Then expected frequencies are AA = 0.80² = 0.64, Aa = 2 × 0.80 × 0.20 = 0.32, and aa = 0.20² = 0.04. In a population of 1,000 individuals, that predicts 640 AA, 320 Aa and 40 aa (640 + 320 + 40 = 1,000). If real counts differ markedly, one of the assumptions may be violated — but a departure alone does not prove selection: non-random mating, population subdivision, sampling variation and genotyping error can also produce departures, and the departure must be investigated, not assumed.

History, in Brief

Charles Darwin is commonly credited with the theory of evolution by natural selection, published in On the Origin of Species in 1859, and Alfred Russel Wallace is commonly credited with independently reaching a similar idea. Darwin's argument drew on variation among domesticated breeds, on fossils and biogeography from his voyage, and on Malthus's observation that populations can grow faster than resources. The modern synthesis of the twentieth century joined Darwin's selection with Mendel's genetics, which Darwin did not have: it explains selection in terms of allele frequencies, which is the language used throughout this article. No historical portrait is used in this article; history is given only as "commonly credited to" wording.

Common Misconceptions

Misconception	Correction
Evolution means progress toward perfection.	Selection favours what works in the current environment; there is no goal, ladder or perfect endpoint.
A giraffe stretched its neck and passed the longer neck on.	That is Lamarckian inheritance of acquired characteristics, which is not the mechanism: acquired changes to the body are not written into DNA and passed on. Selection acts on inherited variation already present.
Humans evolved from living monkeys.	Humans and living monkeys share common ancestors; neither living group is the ancestor of the other.
Evolution is "just a theory", meaning a guess.	In science, a theory is a well-supported explanation of a large body of evidence, not a guess; a hypothesis is the testable guess.
If we cannot see a species change into another, evolution is unproven.	Speciation is usually too slow to watch directly, but its mechanism (selection, drift, gene flow) is observed, and fossil, anatomical, molecular and biogeographical evidence converge on common descent.

Quick Summary

Evolution is change in inherited characteristics of a population over generations; natural selection is differential survival and reproduction due to heritable differences.
Selection needs variation, heritability, overproduction with competition, and differential reproduction.
Fitness means relative genetic contribution to the next generation in a particular environment.
Directional selection shifts the mean, stabilising selection narrows variation around the mean, disruptive selection favours both extremes.
Fossils, homology, molecular similarities, biogeography and observed selection together support common descent; no single line alone carries the case.
Drift (including bottleneck and founder effects) and gene flow change allele frequencies without selection.
Hardy-Weinberg (p² + 2pq + q² = 1) is a null model; departure from it is a clue to investigate, not automatic proof of selection.
Individuals do not evolve; populations do. Evolution has no goal of perfection, and acquired characteristics are not inherited.

FAQ

Does natural selection make organisms perfect? No. It favours variants that work better than alternatives in the current environment. Trade-offs, historical constraints and the available variation all limit the result, so adaptations are good enough, not perfect.

Can a population evolve if all individuals are genetically identical? Not by natural selection: with no heritable variation, there is nothing for selection to sort, even if the environment is harsh. Mutation would first have to introduce variation.

Why do antibiotic-resistant bacteria appear so fast? Bacteria reproduce very quickly and in huge numbers, so rare resistant mutants are more likely to be present, and antibiotics then select strongly for them. The antibiotic does not cause the mutation; it selects the resistant variants already present or arising by chance.

Is genetic drift important in large populations? Very little, in most cases: chance fluctuations average out when the population is large. Drift matters most when populations are small, fragmented or newly founded.

If humans share ancestors with other primates, why do other primates still exist? Because evolution branches. One lineage can give rise to several descendants, and sister lineages can both survive; a living species is a cousin, not a parent, of another living species.

Sources (synthesised, not copied)

OpenStax Biology 2e, Chapters 18 (Evolution and the Origin of Species) and 19 (The Evolution of Populations); OpenStax Concepts of Biology, evolution chapters; Biology LibreTexts, natural selection and Hardy-Weinberg sections; NCBI Bookshelf, StatPearls and educational reviews on antibiotic resistance and sickle-cell balancing selection. History is limited to "commonly credited to" wording for Darwin (1859) and Wallace, and for the modern synthesis joining Darwin and Mendel; no historical portrait is used.

20 Explained MCQs — answers hidden until opened

Open each question to reveal its answer and explanation. These are the Part A MCQs.

1. Which statement best defines evolution?

A. A change in an individual's body during its lifetime B. The growth of a single organism from birth to death C. A sudden mutation in one individual D. A change in the inherited characteristics of a population across generations
Answer: D. Evolution is defined at population level, across generations. An individual's lifetime changes are development or acclimatisation, not evolution.

2. Natural selection can act only if a population has:

A. Heritable variation for a trait B. No predators at all C. Identical genotypes in every individual D. An unlimited food supply
Answer: A. Without heritable variation there is nothing for selection to sort, however harsh the environment.

3. In evolutionary terms, fitness means:

A. An individual's relative contribution of genes to the next generation B. Physical strength and size C. How long an individual lives, regardless of offspring D. How fast an individual can run
Answer: A. Fitness is measured in descendants. A strong animal that leaves no offspring has zero evolutionary fitness.

4. Which is an adaptation, as defined in the article?

A. A tan gained during a summer holiday B. A muscle built by exercise C. A leaf that grows larger in shade during one season D. Thick fur in a mammal living in a cold climate, shaped by past selection
Answer: D. An adaptation is inherited and shaped by selection. A tan, a trained muscle and shade-grown leaves are acquired in one lifetime and not inherited.

5. Under directional selection, the population's trait distribution:

A. Shifts toward one extreme B. Stays exactly the same C. Narrows around the same mean D. Splits into two extremes with few intermediates
Answer: A. Directional selection favours one end of the range, so the mean moves toward that extreme.

6. Human birth weight is used in the article as an example of:

A. Directional selection B. Disruptive selection C. Stabilising selection D. Genetic drift
Answer: C. Very small and very large babies have historically had lower survival than intermediate babies, so intermediates are favoured and variation narrows.

7. Disruptive selection favours:

A. Only the intermediate phenotype B. One extreme only C. No phenotype at all D. Both extremes over the intermediate phenotype
Answer: D. When the intermediate is disadvantaged and both extremes do well, the distribution splits toward the two ends.

8. The forelimbs of a human, a bat, a whale and a horse share the same basic bone plan. These are:

A. Analogous structures B. Homologous structures C. Vestigial structures D. Fossil structures
Answer: B. Homologous structures share a plan inherited from a common ancestor, even though their functions now differ.

9. Bird wings and insect wings look similar and do a similar job, but have different origins. They are:

A. Homologous structures B. Vestigial structures C. Analogous structures, showing convergent evolution D. Evidence of a recent common ancestor
Answer: C. Similar function with different origin is analogy, not homology. It shows convergent evolution, not close ancestry.

10. In undisturbed fossil layers, the general pattern is that:

A. The most complex forms are always at the bottom B. Simpler, earlier forms tend to occur in older, lower layers C. Fossils appear in random order with no pattern D. No transitional forms ever occur
Answer: B. Fossils appear in an order consistent with descent over time, with transitional forms combining earlier and later features.

11. Humans and other primates share the same broken gene for making vitamin C. The simplest explanation is:

A. They inherited the broken copy from a common ancestor B. Each group broke the gene independently by chance in exactly the same way C. Vitamin C is no longer needed by primates D. The gene was broken by diet in each generation
Answer: A. A shared, identical 'mistake' in DNA is strong evidence of inheritance from a common ancestor rather than independent identical breakage.

12. Species on islands tend to resemble species on the nearest mainland more than species in similar climates elsewhere. This pattern supports:

A. Evolution by chance alone B. The idea that climate alone determines every species C. Biogeographical evidence for common descent and dispersal D. Lamarckian inheritance
Answer: C. Distribution fits history and dispersal: island species are relatives of nearby mainland species, not simply climate matches.

13. Allopatric speciation usually begins when:

A. Two species mate in the same habitat B. A single mutation appears in one individual C. A population is split by a geographical barrier D. Pollinators change their behaviour in one season
Answer: C. Geographical separation lets the two parts evolve independently until reproductive isolation develops.

14. Two populations breed at different times of year, so they never mate. This is:

A. Postzygotic isolation B. Genetic drift C. Gene flow D. Prezygotic isolation
Answer: D. Isolation before fertilisation — here, different mating times — is prezygotic. Postzygotic isolation acts after fertilisation, on hybrids.

15. A disaster sharply reduces a population, and survivors carry only a random sample of the original alleles. This is the:

A. Founder effect B. Bottleneck effect C. Directional selection D. Hardy-Weinberg effect
Answer: B. A bottleneck is a sharp reduction followed by recovery from the surviving, unrepresentative sample.

16. A few colonists found a new island population whose allele frequencies differ from the source population. This is the:

A. Bottleneck effect B. Founder effect C. Stabilising selection D. Balancing selection
Answer: B. The founder effect is the unrepresentative allele sample carried by a small founding group.

17. Gene flow between populations generally:

A. Introduces alleles and reduces differences between populations B. Increases differences between populations C. Stops all evolution permanently D. Only affects very large populations
Answer: A. Migration, pollen or seed movement carries alleles between populations, mixing their gene pools, though it can also swamp local adaptation.

18. Which condition is required for Hardy-Weinberg equilibrium at a locus?

A. Strong natural selection favouring one allele B. Frequent mutation changing the alleles C. Random mating and no selection, mutation or gene flow D. A very small population
Answer: C. Hardy-Weinberg is a null model: it requires no selection, no mutation, no gene flow, random mating and a large population.

19. If allele frequencies are p = 0.80 and q = 0.20, the expected frequency of heterozygotes (Aa) is:

A. 0.04 B. 0.32 C. 0.16 D. 0.64
Answer: B. Heterozygote frequency is 2pq = 2 x 0.80 x 0.20 = 0.32. The values 0.64 and 0.04 are the two homozygote frequencies.

20. A student says: 'A blacksmith's strong arms are passed to his children.' The error is:

A. Strong arms are never useful B. Blacksmiths have no children C. Muscles are not made of protein D. Acquired characteristics are not inherited through DNA
Answer: D. This is Lamarckian inheritance. Changes acquired in a lifetime are not written into the DNA passed to offspring.

Separate Interactive Quiz — 20 different questions

These quiz questions are different from the MCQs above. Select an answer for each question. You will not see whether an answer is correct until you submit all 20 questions. Use Reset to try again.

Evolution and Natural Selection: How Populations Change