Natural Selection Acting on Variation

Variation is the raw material, but without a force to sort through it, no species would ever change. That force is natural selection, and understanding how it filters, shapes, and redirects variation is central to grasping the entire process of evolution.

Darwin’s Mechanism: From Variation to Evolution

In his 1859 book Origin of Species by Means of Natural Selection, Darwin laid out a straightforward chain of logic. Not all organisms reproduce at the same rate, and mating is not random. This differential reproduction and non-random mating mean that certain traits get passed on more frequently than others. Nature effectively “selects” the traits that give organisms a better fit with their environment. Over time, these selected traits become more common in the population, which shifts the overall gene frequency (the proportion of particular gene variants in the gene pool). That shift in gene frequency is, at its core, organic evolution.

The process moves through five connected steps:

Differential reproduction and non-random mating occur in the population
Certain traits are eliminated or selected depending on their usefulness
Selected traits lead to better adaptation to the environment
The population’s gene frequency shifts as favoured genes become more common
These cumulative shifts produce organic evolution

So during the struggle for existence, only those individuals survive whose variations prove beneficial in the face of environmental pressure. But natural selection does not operate in just one pattern. It takes different forms depending on which part of the variation spectrum is favoured.

Three Forms of Selection

Stabilizing Selection: Keeping Things in the Middle

Stabilizing selection is the most conservative form. It works by removing individuals at both extremes of a trait and preserving those near the average. Think of it as a narrowing force: the bell curve of trait distribution stays centred but becomes tighter over time.

A well-known example involves newborn birth weight. Babies that are very light or very heavy face higher health risks and lower survival rates. The highest survival rate belongs to babies of intermediate weight. Over generations, this pattern keeps the population clustered around a moderate birth weight. Extreme values on either side are selected against.

Directional Selection: Shifting the Curve

Directional selection pushes the population’s trait distribution towards one end of the spectrum. Instead of trimming both extremes, it favours one extreme over the other. The bell curve effectively slides sideways.

Two clear examples show this at work:

Body proportions in hot humid regions — Populations in equatorial climates have been shaped by directional selection towards wider noses (which help cool inhaled air more efficiently) and longer extremities (which increase the body’s surface-area-to-volume ratio, helping with heat dissipation). The selection pressure consistently pulls traits in one direction.
Alcohol metabolism in East Asian populations — There has been directional selection favouring the fast allele (gene variant) for alcohol metabolism in Oriental (East Asian) regions. This means the population has shifted toward faster processing of alcohol, a clear one-directional genetic change.

Disruptive Selection: Splitting the Population

Disruptive selection is the opposite of stabilizing selection. Rather than favouring the middle, it favours both extremes while putting individuals with intermediate traits at a disadvantage. The result is that a single population splits into two distinct groups.

A classic example is the African swallowtail butterfly Papilio dardanus. This species uses Batesian mimicry (a strategy where a harmless species copies the appearance of a distasteful or dangerous one) to avoid predators. Some individuals closely mimic the distasteful model species and gain protection, while others retain the original appearance of their own species. Butterflies with intermediate appearances, those that look neither quite like the model nor the original, get no protection from either strategy. Predators recognise them as neither the real distasteful species nor something familiar, so they attack. Over time, these intermediate forms are weeded out, and the population splits into two groups: the original form and the mimetic form.

Magnitude of Selection: How Population Dynamics Shape Outcomes

Beyond the three forms, natural selection also operates at different magnitudes depending on population density, the frequency of traits, and group dynamics.

Density Dependent Selection

This type of selection kicks in when the relative abundance of a species affects whether a trait remains advantageous. The Batesian mimicry case from above illustrates this perfectly.

The mimetic form of Papilio dardanus gains protection only when it is rare compared to the genuinely distasteful species. Predators encounter the distasteful species often enough to learn to avoid anything that looks similar, so the rare mimic benefits. But as the mimetic form’s population grows, predators start encountering it more frequently. They begin to discover that these particular butterflies are actually palatable. At that point, the mimicry stops working, and the mimetic form faces a selective disadvantage.

This creates a self-limiting dynamic: the mimetic population can never grow too large because its advantage evaporates with increasing numbers.

Frequency Dependent Selection

Here, the frequency (commonness) of a particular trait in the population determines its selective value. The research of Dolinger et al. on plant defences provides a clear example.

Plants produce alkaloids (chemical compounds that deter pests). When one particular type of plant alkaloid, say alkaloid “A”, becomes the most widely distributed in a plant population, pests face constant exposure to it. Over time, pests evolve resistance specifically against that most common alkaloid. Once resistance develops, alkaloid A loses its protective edge, and plants carrying it become vulnerable. Meanwhile, rarer alkaloid variants (like alkaloid “B”) remain effective precisely because pests have had less exposure to them.

The takeaway is that being common can itself become a disadvantage, because the environment (in this case, pests) adapts to whatever is most frequent.

Group Selection

Sometimes selection operates not on individuals but on the group as a whole. The classic example comes from baboon social behaviour.

In a baboon troop, the dominant male often leads the group and confronts predators head-on when danger appears. This dominant male frequently gets killed in these encounters. From an individual selection standpoint, this makes no sense: the animal with arguably the best genotype (strongest, most capable) is the first to die. But from a group perspective, the logic is clear. By sacrificing itself, the leader allows the rest of the troop, many of whom carry genes similar to his, to survive and reproduce. Selection is acting at the group level, preserving the broader gene pool even at the cost of the single best individual.

Ecological Basis of Selection: r-Selection vs K-Selection

The ecological context in which a species lives also shapes how selection operates. Ecologists distinguish two broad strategies:

Feature	r-Selection	K-Selection
Core strategy	Establish the population as rapidly as possible with high reproductive output	Compete effectively within the constraints of a full environment
Named after	’r’, the intrinsic rate of population increase	’K’, the carrying capacity (maximum population an environment can sustain)
Reproduction	High: many offspring produced	Low: few offspring produced
Individuation	Low: minimal parental investment per offspring	High: significant parental care and resources per offspring

r-selected species thrive in environments where resources are abundant and population density is low. Their strategy is to reproduce quickly and in large numbers, with little investment in any single offspring.

K-selected species live in environments that are near full capacity. Resources are scarce and competition is fierce, so they produce fewer offspring but invest heavily in each one, maximising the chance that each individual survives.

A Social Perspective on r and K Strategies

This ecological framework maps surprisingly well onto human societies:

Farming societies in rural areas resemble r-selection conditions. Food and space are relatively abundant, which supports high reproduction rates with lower individuation (less investment per child).
Urban societies resemble K-selection conditions. Food and space are scarce, pushing families toward fewer children with higher individuation (more resources, education, and attention devoted to each child).

This is not a value judgement but an ecological observation: the availability of resources shapes reproductive strategy in humans just as it does in other species.

Natural Selection in Context: Not the Only Force

While natural selection plays a central role in evolution, it does not work in isolation. Genetic mutation introduces new variation into the gene pool. Isolation prevents gene flow between populations, allowing them to diverge. Genetic drift causes random shifts in gene frequency, especially in small populations. These forces operate alongside natural selection, and together they drive the full spectrum of evolutionary change. No single mechanism can explain the complexity of how species transform over time.