D4.1 Natural selection

Evolution is a change in the heritable characteristics of a population over generations (see A4.1 Evolution and speciation). Natural selection is the main mechanism that causes it. It operates all the time, in every population, and has been operating since life began more than 3.5 billion years ago. Acting over that immense span of time, together with the steady supply of new variation, it has produced the biodiversity of life on Earth: millions of species, each adapted to its own way of life.

Natural selection follows from a few observations that are easy to check:

Darwin replaced Lamarckism: a paradigm shift

By the early 1800s many naturalists accepted that species change over time, but nobody had a convincing mechanism. The most popular idea was Lamarckism: organisms change during their lives through use and disuse of organs, and these acquired characteristics are passed to their offspring. Charles Darwin, in On the Origin of Species (1859), proposed instead that random heritable variation is sorted by natural selection. Alfred Russel Wallace reached the same idea independently, and their papers were presented together in 1858.

Darwin’s theory explained a vast range of observations and gradually displaced Lamarckism. Replacing one explanatory framework with another in this way is called a paradigm shift: a change in the basic framework of a scientific field, in which a new theory replaces the old one and changes the questions scientists ask. Later discoveries, such as genes carried in gametes and the base sequence of DNA, showed why acquired characteristics cannot be inherited and strengthened Darwin’s theory.

Natural selection can only “choose” between variants that already exist, so it depends on heritable variation: differences between individuals that are caused by differences in their genes.

• Mutation is a random change in the base sequence of DNA. A mutation in a gene creates a new allele. It is the original source of all genetic variation. Mutations are not caused by the organism’s needs: they arise at random, and whether a new allele is harmful, neutral or beneficial depends on the environment.
• Sexual reproduction generates new combinations of alleles in three ways:
  • Crossing over in prophase I of meiosis swaps sections between homologous chromosomes, producing chromatids that carry new combinations of linked alleles.
  • Random orientation of homologous pairs in metaphase I (independent assortment) means each gamete receives a random mix of maternal and paternal chromosomes.
  • Fertilization combines the alleles of two gametes from different parents, so new mutations that arose in separate family lines can come together.

Together these processes mean that, apart from identical twins, every individual produced by sexual reproduction is genetically unique. Asexually reproducing organisms such as bacteria rely mostly on mutation, although bacteria can also gain genes by horizontal gene transfer (see C3.2 Defence against disease).

Most species produce far more offspring than can survive. An oak tree can drop tens of thousands of acorns in a good year, and a large female cod releases millions of eggs in a breeding season, yet the numbers of oaks and cod stay roughly constant from one decade to the next. Even slow-breeding species have this potential: Darwin calculated that if every calf survived, a single pair of elephants, among the slowest-breeding animals, would have nearly 19 million descendants after about 750 years.

This overproduction of offspring matters because resources are limited. The maximum population size that an environment can support is its carrying capacity (see C4.1 Populations and communities). When more young are produced than the carrying capacity allows, they must compete for the resources in short supply, and many die before they can reproduce.

Competition between members of the same species (intraspecific competition) is especially intense, because they need exactly the same resources. Overproduction does not cause evolution by itself; it creates the “struggle for existence” in which small heritable differences decide who survives.

A selection pressure is any factor that makes some individuals more likely than others to survive and reproduce. Selection pressures can be:

• Biotic: caused by other organisms, such as predators, pathogens, parasites, competitors and potential mates.
• Abiotic: caused by non-living features of the environment, such as temperature, water availability, salinity, pH, fire or flooding.

Many abiotic pressures are density-independent: their effect on each individual does not depend on how crowded the population is. A severe frost kills an individual that cannot tolerate it whether the population is large or small. Density-dependent factors such as competition and disease, by contrast, act more strongly in crowded populations.

Abiotic extremes still cause selection, because individuals differ in their heritable tolerance:

• Low temperature. After an unusually severe cold snap in Texas in 2014, surviving green anole lizards in southern populations tolerated cold better than lizards sampled there before the event.
• Cold, wet weather and food shortage. A long spell of cold rain in Nebraska in 1996 stopped flying insects, the food of cliff swallows, and killed about half of the birds. The survivors were, on average, larger, probably because bigger birds hold more fat reserves.
• High temperature and drought. Heatwaves kill plants and animals that cannot cope with heat or water loss; individuals with greater heat or drought tolerance survive and reproduce.
• Fire. Trees with thicker, more fire-resistant bark, or the ability to resprout, survive bushfires that kill others.

The key point is that the abiotic factor does not create the tolerant variants. It removes the less tolerant ones, so the heritable traits of the survivors become more common in the next generation.

Individuals in a population differ in how well adapted they are: how well their heritable traits suit them to their environment. Better-adapted individuals tend to survive longer and to have more offspring. This difference in survival and reproduction is the basis of natural selection.

Biologists describe it with the idea of fitness: how well a genotype survives and how much it can reproduce (its reproductive potential) in a given environment, measured by its contribution of offspring to the next generation compared with other genotypes. Fitness is not strength, size or health. An individual that survives but never reproduces has a fitness of zero, and a trait that is advantageous in one environment can be a disadvantage in another.

Selection is strongest through intraspecific competition. Members of the same species compete for the same food, space and mates, so even small heritable differences matter: a slightly better-camouflaged moth is a little less likely to be eaten, and over many generations that small advantage changes the population.

Things to try in the simulation:

1. Keep the default settings and press Run 10 generations a few times. The histogram shows that variation exists first; selection then removes moths that contrast with the bark, and the dark allele frequency rises towards the value that matches the bark.
2. Set selection pressure to 0. Which moths die is now a matter of chance, so allele frequencies only drift slightly.
3. Move the bark shade to 0 (pale, lichen-covered bark). The direction of selection reverses. The same allele can raise or lower fitness, depending on the environment.

For natural selection to change a population, the traits that improve survival and reproduction must be heritable: coded in the base sequence of genes, so that offspring inherit them through the gametes.

Many characteristics are acquired during an individual’s life because of its environment or its own activity:

• muscles enlarged by exercise, or calluses on the hands
• a scar, a lost limb or a docked tail
• skin darkened by exposure to sunlight
• learned skills and knowledge
• a plant stunted by poor soil

None of these changes the base sequence of genes in the cells that produce gametes, so none of them is inherited and none can contribute to evolution. In the 1880s August Weismann cut the tails off mice for five successive generations; every generation was still born with a normal tail. His conclusion, that the germ line (the cells that make gametes) is kept separate from the body cells, explains why Lamarck’s mechanism does not work.

In the simulation above, tick Survivors darken during adult life. The adults look darker, but the young are born with shades set only by the alleles they inherit, and the dark allele frequency follows exactly the same course as before.

Some chemical marks on DNA (epigenetic changes, see D2.2 Gene expression) can occasionally pass to the next generation, but they do not alter the base sequence and are usually reset within a generation or two.

Sexual selection is selection for traits that increase an individual’s success in getting a mate. It happens in two main ways:

• Mate choice (usually by females): members of one sex prefer mates with particular physical or behavioural traits.
• Contests between rivals of the same sex: for example, red deer stags fight with their antlers for control of groups of females.

Females typically invest more in each offspring (large eggs, pregnancy, parental care), so they are choosier, while males compete for matings. Traits that females prefer can therefore spread even if they make survival harder.

Birds of paradise

Birds of paradise are a family of about 45 species of forest birds, most of them living in New Guinea. In many species the males have extraordinary plumage: long tail streamers, iridescent breast shields, capes and head plumes, combined with elaborate courtship dances and calls. In most species the females are drab brown, and they alone build the nest and raise the young. Males of some species display together at a traditional site (a lek), and the females visit, watch and choose.

For example, a male superb bird-of-paradise spreads a velvety black cape into an oval around his head, shows an iridescent blue-green breast shield and hops around the female in a rhythmic dance. Females mate with the males whose display they prefer, and a few males obtain most of the matings.

Why would such costly traits evolve? Bright, elaborate plumage takes energy to grow, makes a male easier for predators to spot and can hinder flying. Only a male in good condition, with genes that let him find food, resist parasites and avoid predators despite these handicaps, can produce the most impressive display. The display is therefore a signal of overall fitness. Females that choose such males produce offspring that inherit the good genes, and sons that inherit the attractive traits. Over many generations, sexual selection has made the plumage and display more and more elaborate.

Guppies (Poecilia reticulata) are small fish from the streams of Trinidad and north-east South America. Males are covered with spots of orange, black, blue and iridescent colours; females are drab. In Trinidad’s mountain streams, waterfalls separate stretches with different predators. Downstream, the pike cichlid (Crenicichla) eats many adult guppies; upstream, above waterfalls that it cannot climb, the main predator is a small killifish, Rivulus hartii, which mostly takes juvenile guppies. John Endler noticed that males are much more colourful in the low-predation streams, and tested whether predation was the cause.

Two selection pressures act on the colour of male guppies in opposite directions:

• Sexual selection by females, who prefer to mate with males that have more, larger and brighter spots.
• Natural selection by visual predators, which spot and catch conspicuous males more easily, favouring duller males.

In the greenhouse experiment, males in the pike cichlid pools ended up with fewer and smaller spots, while males in the Rivulus and predator-free pools became more colourful. In the field transplant, the descendants of fish moved from a high-predation site to a stream with only Rivulus became more colourful within about two years. Because guppies breed so quickly, this was roughly fifteen generations, showing that evolution can be observed over a few years when selection pressures change.

Modelling selection by controlling the selection pressure

Endler’s work is an example of modelling natural and sexual selection with an experiment. He held other variables constant (the same stock, pools of the same size, gravel and lighting), changed only the independent variable (type of predator), and measured the dependent variable (spots per male). The simulation is a second kind of model: you can switch off female preference or predation and see which pressure produces each change.