Skip to main content
SWStudyWalks Search

Evolutionary Processes

Part 3 of 4: The Forces That Change a Population. This unit runs past the 40-minute ceiling, so it ships in 4 lectures.

The narration between entries paraphrases; it never reads an entry aloud before the entry plays. Every entry in this lecture plays exactly as it does on its own page — same file, same speed, nothing laid over it.

What you can do after this

  1. Say which fraction of variation selection can act on
  2. Tell drift from selection, and name two special cases of drift
  3. Tell the three modes of selection apart by their effect on variation
  4. Say why a sexually selected trait can reduce survival
  5. Say why evolution produces no perfect organism

What it covers

19 entries in 7 topics. Each number links to that entry on its own.

  1. 1. Variation, and what selection can reach

  2. 2. Chance

  3. 3. Three more forces

  4. 4. Fitness, and three modes of selection

  5. 5. Two cases where the rules bend

  6. 6. What selection cannot do

  7. 7. One mutation, followed to fixation

Check yourself

Which fraction of a population's variation can selection act on?

Only the heritable fraction — the portion of phenotypic variation attributable to genetic differences rather than to environmental factors. Variation caused by the environment can be highly visible and still be invisible to selection: skin tone shaped by sun exposure, or a reptile's sex set by incubation temperature, are not passed to offspring and so cannot be selected for or against. The limit narrows further still, because only genetic variation in the germline — the cells producing sperm or eggs — can reach the next generation at all. A mutation in a skin cell affects that cell's descendants and stops there, which is why a somatic mutation spreading through a tissue is not population evolution.

How does drift differ from selection, and what are its two special cases?

Genetic drift is random change in allele frequencies between generations, driven by chance rather than by any effect on fitness — an allele can become more common because its carriers happened to reproduce more, with no advantage involved. Drift and selection are not alternatives: both can act on a population at the same time. Two circumstances magnify drift. The founder effect occurs when a small group splitting off does not, by chance, reflect the parent population's allele frequencies, and its descendants inherit that skew. The bottleneck effect occurs when a sudden event drastically cuts population size, so the survivors' frequencies define the population afterward — and survival in a bottleneck is indiscriminate with respect to genotype.

How do the three modes of selection differ?

By what each does to the population's phenotypic variation. Stabilising selection favours individuals near the average and selects against both extremes, so variation narrows around an average that does not move. Directional selection favours one extreme of the existing range, shifting the population's average toward it — the peppered moth shifting darker as soot darkened tree bark — though it works on an already-present extreme rather than creating a new phenotype. Disruptive selection favours two or more extremes simultaneously while selecting against the intermediate, so variation increases; beach mice matching either sand or grass outsurvive intermediates that match neither.

Why can a sexually selected trait reduce survival?

Because mating success and survival are two separate contributions to fitness, and selection acts on the total rather than on either alone. Sexual selection favours traits that increase mating success specifically, so a trait can spread even while making its bearer more vulnerable — a peacock's elaborate tail is favoured by female choice despite reducing its escape speed from predators. Because such selection acts unevenly on the sexes, it produces sexual dimorphism, visible physical difference between them. Two hypotheses explain why a costly ornament would signal quality — the handicap principle and the good genes hypothesis — and the course declares no verdict between them.

Why does evolution produce no perfect organism?

Because selection is constrained in several ways at once. It acts only on variation that already exists, so it cannot reach a better solution that no individual happens to carry. Genes sitting close together on a chromosome are inherited together, so a beneficial allele can drag a harmful linked allele along with it. And selection is not the only force changing allele frequencies: drift in particular changes a population's fit without improving it, as do gene flow and nonrandom mating. Evolution has no goal it is working toward — of the several forces acting, only natural selection consistently produces adaptation at all.

Lecture transcript

The narration as recorded, in order. Where an entry plays, the transcript links to it rather than reprinting it — the entry plays from its own file, unaltered, and says the same thing on its own page.

This is Unit Eight of Biology, Part Three: the forces that change a population. The part opens on a limit. Selection acts only on the heritable fraction of a population's variation — the part attributable to genetic differences. Variation caused by the environment, however visible, does not respond to selection at all, because it is not passed on. That limit narrows further. Only variation in the germline, the cells producing sperm or eggs, can reach the next generation. A mutation in a skin cell affects that cell's descendants and stops there — which is why a somatic mutation spreading through a tissue is not population evolution. Then five forces. Drift changes frequencies by chance alone and can act at the same time as selection rather than instead of it. Gene flow moves alleles between populations. Mutation is the ultimate source of every new allele and a slow force on frequencies. Nonrandom mating changes genotype frequencies without changing allele frequencies. Then selection proper, in three modes distinguished by what each does to variation — narrowing it, shifting it, or widening it. And the part closes on what selection cannot do. By the end you will be able to do five things. One: say what selection can reach. Two: tell drift from selection. Three: tell the three modes apart. Four: say why an ornament can cost survival. Five: say why there is no perfect organism.

Four entries narrowing what counts. A population needs genetic variation to be polymorphic, selection reaches only the heritable fraction of variation, and only germline variation is heritable at all. Each entry tightens the previous one.

Distinct variants.

Bio 1749 Polymorphism

A polymorphic population displays two or more distinct phenotypic variants of a characteristic among its individuals, reflecting genetic variation at that locus — a litter of kittens with several coat-colour patterns. A trait is polymorphic only where the variation has a genetic basis; diet or environment alone does not make a population polymorphic.

How much is genetic.

Bio 1751 Heritability of Phenotypic Variation and Environmental Variation

Heritability is the fraction of a population's phenotypic variation attributable to genetic differences, with the rest coming from environmental factors — skin tone shaped by sun exposure, reptile sex set by incubation temperature. Selection acts only on the heritable fraction; environmentally caused variation, however visible, does not respond to it.

The displaced idea.

Bio 1753 "Acquired Traits Are Inherited"

Traits an organism acquires during its own lifetime are not passed to offspring, because only genetic variation in the germline is heritable — an athlete's gym-built muscle does not appear in an offspring's genome, since building muscle alters no germline D N A. This is the inheritance of acquired characteristics that natural selection displaced.

Which cells count.

Bio 1755 Only Germline Variation Is Heritable

Only genetic variation in an organism's germline — the cells producing sperm or eggs — can pass to offspring and be subject to selection. A skin-cell mutation affects that cell's descendants but never reaches the next generation. A somatic mutation spreading through one tissue is not population evolution; it never entered the gene pool.

Three entries on frequency change that owes nothing to fitness. The first defines it, and the two after are circumstances where chance dominates — a small founding group, and a sudden crash. Both are drift rather than selection.

Random change.

Bio 1757 Genetic Drift

Genetic drift is a random change in allele frequencies between generations, driven by chance rather than any effect on fitness — an allele's frequency shifting because its carriers happened to reproduce more. Drift and natural selection can act on a population at once; one being present does not rule out the other.

A small start.

Bio 1759 The Founder Effect

The founder effect occurs when a small group splitting off does not, by chance, reflect the larger population's allele frequencies, and its descendants inherit the skew — a colonising group heavy in one allele founding a population where it stays unusually common. It is drift specific to a founding moment, not selection.

A sudden crash.

Bio 1761 The Bottleneck Effect

The bottleneck effect occurs when a sudden random event drastically reduces population size, so the survivors' allele frequencies become the population's structure afterward. A bottleneck kills indiscriminately with respect to genotype — surviving is chance, not selection for any particular allele.

Three entries completing the list. Alleles moving between populations, the only process that creates a genuinely new allele, and mating that is not random — which changes something narrower than the others do.

Between populations.

Bio 1763 Gene Flow and Clines

Gene flow is the movement of alleles between populations through migrants or gametes, and a cline is gradual variation across an ecological gradient — pollen carried on the wind, or a species growing larger toward the poles. Ongoing flow keeps a cline's variation gradual; restricted flow sharpens differences and can eventually drive speciation.

The only source.

Bio 1765 Mutation as an Evolutionary Force, Population-Level

At the population level, mutation is the ultimate source of every new allele, though any single event changes allele frequency only slightly — one mutation introduces one new allele copy into the gene pool for the first time. Mutation supplies novelty slowly; most frequency change comes from selection, drift or gene flow acting on variation already present.

Not at random.

Bio 1767 Nonrandom Mating

Nonrandom mating changes genotype frequencies when individuals do not pair at random, whether from mate preference or unequal access to partners — assortative mating, pairing with phenotypically similar partners, is one common form. It changes genotype frequencies specifically, and does not by itself change underlying allele frequencies.

Four entries. What fitness actually measures — a comparison within one population rather than an absolute — and then three selection patterns. Sort the three by what each does to the population's variation and they stay separate.

Contribution, compared.

Bio 1769 Fitness and Relative Fitness

Fitness is an individual's genetic contribution to the next generation, and relative fitness compares that against others in the same population rather than any absolute standard — an individual whose offspring outperform the population average has higher relative fitness whatever its absolute output. The same individual could have a different value elsewhere.

Narrowing.

Bio 1771 Stabilizing Selection

Stabilising selection favours individuals near a population's average phenotype and selects against both extremes, reducing overall variation — mice whose fur matches the forest floor's average shade surviving better than lighter or darker ones. It decreases variation around a stable average; it does not shift that average in any direction.

Shifting.

Bio 1773 Directional Selection

Directional selection favours one extreme of an existing phenotypic range, shifting the population's average toward it, typically after an environmental change — peppered moths shifting darker as industrial soot darkened tree bark. It shifts an existing distribution toward an already-present extreme; it does not create a new phenotype.

Widening.

Bio 1775 Disruptive (Diversifying) Selection

Disruptive selection favours two or more phenotypic extremes at once while selecting against the intermediate, increasing variation — beach mice matching either sand or grass surviving better than intermediates that blend with neither. Favouring extremes together is the opposite outcome from stabilising selection's single favoured average.

Three entries. One where a trait's advantage depends on how common it is, and two on traits favoured for mating success at a cost to survival — the second of which leaves a live dispute unsettled.

Advantage from rarity.

Bio 1777 Frequency-Dependent Selection

Frequency-dependent selection favours a phenotype based on how common or rare it currently is, rather than on any fixed advantage of the trait — rare throat-colour morphs in side-blotched lizards gaining an advantage precisely because they are rare. The advantage can reverse as the frequency changes, so the favoured phenotype cycles rather than settling.

Mating, not surviving.

Bio 1779 Sexual Selection and Sexual Dimorphism

Sexual selection favours traits increasing mating success rather than survival, and acting unevenly on the sexes it produces sexual dimorphism — a peacock's tail favoured by female choice despite reducing escape speed. A sexually selected trait can reduce survival while increasing mating success; the two fitness contributions are not the same thing.

An open dispute.

Bio 1781 The Handicap Principle and the Good Genes Hypothesis

The handicap principle and the good genes hypothesis both propose that a costly ornament signals male quality — the first holding that the cost itself proves fitness, the second that the ornament directly signals disease resistance. Both are stated in strongest form with no verdict declared; costly signalling remains an open question.

One entry limiting everything the part has claimed. Selection works on what is already there, other forces change populations without improving them, and nothing here has a direction it is aiming at.

No perfection.

Bio 1783 Selection's Limits: No Perfect Organism, Not All Evolution Adaptive

Selection acts only on existing variation and cannot produce a perfect organism, and other forces — drift especially — change a population's fit without improving it. Genes close together are inherited together, so a beneficial allele can drag a harmful linked one along. Evolution has no goal; only natural selection consistently produces adaptation.

One capstone running a single new allele through the whole part. Its warning is the corrective to a story that otherwise sounds inevitable — a beneficial mutation can be lost by chance before selection ever acts on it.

The capstone. Mutation to fixation.

Bio 1785 Tracing a Population's Adaptive Evolution from Mutation Through Fixation

A population sits near Hardy-Weinberg equilibrium until one new mutation appears in a single individual, introducing a novel allele whose fate is not settled by its origin. Existing in only a few copies, it is dominated by drift — chance alone could erase it within a few generations. Suppose the environment then shifts to favour its carriers: directional selection begins, the frequency climbs, and Hardy-Weinberg's no-selection assumption is violated, which is exactly what the model's departure reveals. Sustained long enough, the allele reaches fixation, and the population has evolved by this course's own definition.

Five things to carry out of Part Three. Selection reaches only the heritable fraction of variation — the part attributable to genetic differences. Environmentally caused variation does not respond to it. And within genetic variation, only what sits in the germline can pass on, which is why a somatic mutation spreading through a tissue is not population evolution. Drift changes allele frequencies by chance, independent of fitness, and it acts alongside selection rather than instead of it. The founder effect is drift at a founding moment, when a small splinter group does not reflect the parent population. A bottleneck is drift after a crash, with survival indiscriminate with respect to genotype. The three selection modes differ by what they do to variation. Stabilising selection narrows it, favouring the average against both extremes without moving that average. Directional selection shifts the average toward one already-present extreme. Disruptive selection widens it, favouring both extremes against the middle. A sexually selected trait can reduce survival while increasing mating success, because those are two separate contributions to fitness. A peacock's tail costs escape speed and gains mating opportunity, and selection weighs the total rather than either half. And there is no perfect organism, because selection acts only on variation that already exists, linked genes drag harmful alleles along with beneficial ones, and forces like drift change populations without improving them. Evolution has no goal — only natural selection consistently produces adaptation at all. Part Four turns to phylogeny, classification, and how the tree of life is reconstructed.

This is free, and stays free. If it helped, here is what keeps it running.