Organisms evolve because their DNA changes over time, and those changes get filtered by the environment. Some genetic changes help an organism survive and reproduce, so they spread through the population. Others are neutral or harmful and may persist or disappear depending on chance. Evolution isn’t a choice organisms make. It’s the inevitable result of imperfect DNA copying, environmental pressure, and the math of who leaves behind more offspring.
How Natural Selection Works
Natural selection is the most familiar engine of evolution, and it operates on a simple principle: organisms with traits that help them survive and reproduce in their environment leave behind more offspring than those without. Over generations, the genes behind those helpful traits become more common in the population. The key insight is that natural selection is always relative. It doesn’t matter how well an organism performs in absolute terms. What matters is whether it performs better or worse than others in the same population. Evolution tracks the difference between winners and losers.
Four conditions have to be met for natural selection to work. First, individuals in a population must vary in their traits. Second, some of that variation must affect survival or reproduction. Third, the variation must be at least partly heritable, meaning it has a genetic basis. And fourth, the environment must favor certain variants over others. If any one of these conditions is missing, natural selection stalls. A trait that improves survival but isn’t genetic, for instance, can’t be passed to the next generation and won’t drive evolutionary change.
When population geneticists model this process, the path of a beneficial gene spreading through a population follows a predictable S-shaped curve. It starts slow when the gene is rare, accelerates as it becomes more common, and levels off as it approaches saturation. The speed of this process depends entirely on how much of a reproductive edge the gene provides.
Where Genetic Variation Comes From
Evolution can’t happen without raw material, and that raw material is genetic variation. New variations enter a population primarily through mutations: errors that occur when DNA is copied during cell division. Many mutations are point mutations, where a single “letter” of the genetic code gets swapped for another. Others involve small insertions or deletions of genetic material. These errors happen because DNA copying, while remarkably accurate, isn’t perfect. The molecular machinery that builds new DNA strands has a built-in proofreading system, but some mistakes slip through.
Mutations also arise from external damage. Ultraviolet radiation, certain chemicals, and other environmental agents can alter the structure of DNA in ways that change how genes are read. These chemically induced mutations add to the pool of variation that natural selection, or other evolutionary forces, can act on.
Beyond small-scale mutations, larger genetic reshuffling happens through recombination. During sexual reproduction, chromosomes exchange segments with each other, creating new combinations of existing genes. Mobile genetic elements can also jump from one location in the genome to another, or even between chromosomes. These rearrangements don’t create new genetic “letters,” but they shuffle the deck in ways that can produce traits neither parent had.
Evolution Without Natural Selection
Not all evolution is adaptive. Genetic drift, the random fluctuation of gene frequencies from one generation to the next, can change a population’s makeup without any trait being better or worse than another. Drift is especially powerful in small populations, where chance events have outsized effects.
Two scenarios make drift particularly dramatic. A population bottleneck occurs when a population’s size crashes, even briefly. The survivors carry only a fraction of the original genetic diversity, and some gene variants are lost forever simply because their carriers didn’t make it. The population that rebounds may look genetically quite different from the one that existed before, not because of any adaptive advantage, but because of luck.
A founder effect is similar but happens when a small group splits off to colonize a new area. That small group carries a random sample of the original population’s genes, which may not be representative. The Afrikaner population of South Africa, descended largely from a small group of Dutch colonists, has an unusually high frequency of the gene that causes Huntington’s disease. Those original colonists just happened to carry that gene at higher-than-normal rates. No selective advantage was involved.
Reduced genetic variation from bottlenecks or founder effects can have long-term consequences. A population with less diversity may struggle to adapt when conditions change, because the genetic variants that selection would need to work with may already be gone.
Gene Flow Between Populations
When individuals migrate between populations and breed, they carry genes with them. This gene flow acts as a connecting thread, keeping separated populations genetically similar. When migration rates are high (roughly four or more migrants per generation), populations stay genetically unified even at genes that have no particular adaptive function. When migration drops below about one migrant per generation, populations begin to diverge as drift fixes different gene variants in each group.
Gene flow interacts with natural selection in two opposing ways. It can prevent local adaptation by swamping a population with genes suited to a different environment. If the migration rate exceeds the survival advantage of a locally beneficial trait, that trait can’t get established. But gene flow also spreads beneficial mutations. Even very low levels of migration are enough to carry a highly advantageous gene across an entire network of populations. This means that species with limited migration can still evolve together at their most important genes, while diverging at genes that don’t matter much.
Sexual Selection and Costly Traits
Some evolved traits actually decrease an organism’s chances of survival. The peacock’s tail is the classic example: it’s heavy, conspicuous to predators, and expensive to maintain. It exists because it increases mating success enough to outweigh the survival cost.
Sexual selection operates through two mechanisms. In intrasexual selection, members of the same sex (usually males) compete directly with each other for access to mates. This drives the evolution of weapons and large body size: deer antlers, beetle horns, and the sheer bulk of elephant seals. Individuals who can exclude competitors get more mating opportunities and father more offspring. In intersexual selection, one sex (usually females) chooses mates based on specific traits. This drives the evolution of elaborate displays, bright coloration, and complex courtship behaviors. Females tend to be choosier because they typically invest more in each offspring, so picking the right mate matters more to their reproductive success.
What Pushes Organisms to Adapt
The pressures that drive adaptation fall into two broad categories. Biotic pressures come from other living things: predators, parasites, competitors, bacterial and viral infections. Animals have evolved sophisticated immune systems largely in response to the constant threat of pathogens. The “arms race” between hosts and parasites is one of the fastest-moving areas of evolution, with each side continuously adapting to the other’s latest defenses or attacks.
Abiotic pressures come from the physical environment: temperature extremes, drought, high salinity in soil, UV radiation, changes in atmospheric chemistry. Plants face these pressures acutely because they can’t move. High salinity, chilling, and drought are among the most common abiotic stresses that constrain plant growth and have driven the evolution of specialized tolerance mechanisms over millions of years. When these pressures shift, as they do during climate changes, populations either adapt, migrate, or go extinct.
How New Species Form
Evolution doesn’t just change existing species. It creates new ones. Speciation most often begins when a physical barrier, like a mountain range, a river, or an ocean, splits a population in two. This is allopatric speciation. With gene flow cut off, the separated groups accumulate different mutations, adapt to their local environments independently, and eventually diverge so much that they can no longer interbreed even if the barrier disappears.
Speciation can also happen without complete geographic separation. In some cases, populations living in the same area adapt to different habitats or food sources so strongly that they begin mating preferentially with others who share their specialization. Periwinkle snails on rocky shores, for example, have evolved distinct forms adapted to different zones of the shoreline. The local adaptation itself creates a barrier to gene exchange, and additional barriers like habitat preference and mate choice reinforce the split. Hybrids between the forms show reduced fitness, further cementing the divergence.
Evolution You Can Watch Happen
Evolution isn’t always a slow grind across millennia. Bacteria can evolve resistance to antibiotics over the course of a single infection. In laboratory experiments, bacteria exposed to gradually increasing doses of an antibiotic evolved significant resistance in just 12 days, with the drug dose doubling every 72 hours as resistance emerged. This speed is possible because bacterial populations are enormous, generation times are measured in minutes, and the selective pressure is extreme: adapt or die.
Humans are still evolving too, though on longer timescales. One of the clearest recent examples is lactose tolerance. Most mammals lose the ability to digest milk sugar after weaning, and most humans historically did the same. But within the last 7,000 to 10,000 years, populations that domesticated dairy animals evolved the ability to keep producing the enzyme that breaks down lactose throughout adulthood. A computer simulation estimated that the key genetic variant first came under selection about 7,500 years ago among dairy farmers in central Europe. Independent mutations producing the same effect arose in African pastoral populations around the same period. This is evolution responding to a cultural change (dairy farming) with a biological adaptation, all within a timeframe that is remarkably short by evolutionary standards.
Inheritance Beyond DNA
The traditional view of evolution focuses on changes to DNA sequences, but there’s growing evidence that chemical modifications to DNA and its associated proteins can also be inherited across generations. These epigenetic changes don’t alter the genetic code itself. Instead, they affect which genes are turned on or off, changing an organism’s traits without changing its underlying blueprint.
In plants and invertebrate animals, transgenerational epigenetic inheritance is well documented. Chemical tags on DNA can persist across multiple generations even after the environmental exposure that triggered them is gone. Some of these modifications show surprising permanence, with certain genes retaining their altered activity status generation after generation. In mammals, the evidence is more limited. Most epigenetic marks get wiped clean between generations, though some appear to escape this reset. Whether epigenetic inheritance plays a significant role in long-term mammalian evolution remains an open question, but in plants and invertebrates, it likely contributes meaningfully to how populations change over time.