How Are Traits Inherited by Offspring?

Traits pass from parent to offspring primarily through DNA, the molecular instruction set packaged into chromosomes that gets copied and transmitted during reproduction. Each parent contributes roughly half of their genetic material, and the combination produces a new individual whose characteristics reflect that blend. But “reflect that blend” understates how varied and strange the process actually is. Some traits follow tidy inheritance rules first described over 150 years ago, while others involve thousands of genes, chemical tags layered on top of DNA, or genetic material that sits outside the cell nucleus entirely.

The Classic Pattern and Why It Only Goes So Far

The simplest version of inheritance involves a single gene with two variants: one dominant and one recessive. If you carry even one copy of the dominant version, that trait shows up in your body. You need two copies of the recessive version for that trait to appear instead. This framework comes from Gregor Mendel’s work with pea plants and explains a handful of human traits cleanly, such as whether you can roll your tongue or whether you have a widow’s peak hairline.

Underlying this system is a physical process during reproduction. When your body makes sperm or egg cells, paired chromosomes separate so that each reproductive cell gets only one copy of each chromosome instead of two. During that separation, chromosomes also swap segments with their partner, meaning individual pieces of chromosomes can sort independently as well.¹ This shuffling is why siblings who share the same two parents can look so different from one another. Each egg and each sperm carries a unique mix.

Dominance itself turns out to be more complicated than a simple on-off switch. At the molecular level, dominance can arise in several ways: a defective protein might poison a larger molecular complex it belongs to, a single working copy of a gene might not produce enough product to maintain normal function, or interactions between genes at different locations can shift which version wins out.² The upshot is that “dominant” and “recessive” are useful labels for simple cases, but they describe patterns you see in families rather than a single universal mechanism inside cells.

Polygenic Traits and the Thousands-of-Genes Problem

Most traits people care about, including height, skin color, susceptibility to heart disease, and intelligence, do not follow simple dominant-recessive rules. They are polygenic, meaning many genes contribute small nudges in one direction or another, and the combined effect produces a continuous range rather than two or three neat categories.

Height is the textbook example. It is highly heritable, meaning genetics plays a large role, yet no single “tall gene” exists. Genome-wide studies have identified over 12,000 independent genetic signals associated with height, and together they account for most of the common genetic variation in how tall people grow.³ Each individual variant has a tiny effect. Height is, in the language researchers use, a model polygenic trait: the additive effects of many small variants create a smooth, bell-curve distribution across a population.⁴

The idea that many tiny genetic effects can add up to explain the family resemblance seen in complex traits dates back over a century. The mathematical framework for reconciling single-gene inheritance with the smooth, graded variation you see in traits like height was proposed by R.A. Fisher and is sometimes called the infinitesimal model.⁵ It remains the foundation for how geneticists think about complex traits today. You inherit thousands of small-effect variants from each parent, and the sum total shapes where you land on the spectrum for a given trait.

When It Matters Which Parent a Gene Came From

For most genes, your body does not care whether a particular copy arrived via your mother or your father. Both copies get read and used. But a small set of genes, roughly a hundred or so in humans, break this rule through a phenomenon called genomic imprinting. For these genes, only the copy from one specific parent is active; the other is chemically silenced.⁶

Imprinting matters because it means a mutation in an imprinted gene can have completely different consequences depending on which parent it came from. If the active copy is the one with the mutation, you get a problem. If the silenced copy carries the mutation, you might never know. This parent-of-origin effect is the best-characterized cause of parent-specific trait inheritance and plays a role in growth disorders, some cancers, and certain neurological conditions.⁷ The silencing itself happens through chemical marks, mainly DNA methylation, that get applied differently during egg and sperm development.⁸

Beyond imprinting, mothers exert a separate kind of influence through maternal-effect genes. These are genes whose products, mostly RNA molecules and proteins, get stockpiled in the egg cell before fertilization. The embryo relies on these stored maternal supplies during its earliest cell divisions, before its own genome kicks in.⁹ If something goes wrong with these maternal stockpiles, the embryo can fail to develop normally even if its own DNA is perfectly fine.¹⁰ So the mother’s genetic contribution shapes the offspring in two distinct ways: through the chromosomal DNA she passes on and through the molecular environment she builds inside the egg.

Mitochondrial DNA Comes Only from Your Mother

Your cells contain a second, smaller genome that lives inside mitochondria, the structures that generate energy. This mitochondrial DNA is inherited exclusively from your mother. Studies going back to 1980 confirmed this by showing that children always carry their mother’s mitochondrial DNA pattern, never their father’s.¹¹

The mechanism behind this maternal-only inheritance was long debated, but recent research has clarified it in striking detail. During sperm development, a key protein called TFAM, which normally protects and maintains mitochondrial DNA, gets rerouted away from the mitochondria and sent to the sperm cell’s nucleus instead. This effectively strips sperm mitochondria of their DNA before fertilization even happens.¹² The result is that the fertilized egg receives its entire mitochondrial genome from the mother’s egg cell, with essentially zero paternal contribution.

Mitochondrial DNA mutations cause a range of diseases affecting energy-hungry tissues like muscles, the brain, and the heart. Because mitochondria reproduce within cells, a single cell can contain a mixture of normal and mutant mitochondrial DNA. Research in mice has shown that the body has a quality-control system involving a process called mitophagy, which selectively destroys mitochondria carrying mutant DNA. When this system is overwhelmed by a high burden of mutations, defective mitochondria slip through and get passed to the next generation.¹³ This explains why mitochondrial diseases can appear suddenly in a family line: the protective system keeps mutant DNA in check for generations until the balance tips.

Epigenetic Inheritance and the Question of Acquired Traits

One of the most debated frontiers in inheritance is whether experiences and exposures during a parent’s life can affect their children’s biology through mechanisms that do not alter DNA sequence. The answer, at least in animal models, is a qualified yes. Chemical modifications to DNA and associated proteins, collectively called epigenetic marks, can carry information from parent to offspring without changing any genetic code.

The strongest evidence comes from studies involving histone modifications, the chemical tags attached to the protein spools around which DNA is wrapped. In roundworms, researchers showed that specific histone marks on sperm chromosomes persist in offspring and influence which genes get turned on in the next generation’s germ cells. When sperm lack a particular repressive mark, the offspring’s germline abnormally activates genes meant for brain cells, pushing germ cells toward adopting a neuronal identity.¹⁴ In mice, experimentally disrupting histone marks in developing sperm led to severe developmental problems in offspring, and these defects persisted into subsequent generations even when the disruption was no longer present.¹⁵

Small RNA molecules in sperm are another channel. Mature sperm are loaded with tiny non-coding RNAs, and their composition changes depending on what the father experienced, particularly diet and stress.¹⁶ These sperm RNAs, especially fragments derived from transfer RNAs, can transmit metabolic effects. In one striking experiment, male mice fed a high-fat diet passed metabolic problems to their pups through sperm RNAs. Knocking out a specific enzyme that chemically modifies those RNAs blocked the transmission entirely.¹⁷ This suggests that RNA modifications serve as an additional layer of hereditary information, carried in sperm alongside DNA.¹⁸

How much of this applies to humans remains genuinely unclear. During the development of human eggs and sperm, the genome undergoes a massive epigenetic reset: DNA methylation levels drop below 5% genome-wide, erasing most parental marks.¹⁹ This reprogramming exists precisely to prevent the wholesale inheritance of epigenetic noise. Some marks clearly escape the reset, as imprinted genes demonstrate, but whether environmentally acquired epigenetic states regularly survive this erasure in humans is still a matter of active debate.²⁰ The animal experiments are compelling, but extrapolating them directly to human generations is premature.

The Environment Shapes What Genes Actually Do

Even when two organisms share the same DNA, the traits they display can differ dramatically depending on their environment. This is not a failure of genetics; it is a built-in feature. The same genotype can produce different outcomes under different conditions, a phenomenon called phenotypic plasticity.

In a controlled study using dozens of genetically distinct fruit fly lines, researchers found that roughly half of the variation in adult behavior could be attributed not to genetic differences alone, nor to environmental differences alone, but to the interaction between the two. The same fly genotype behaved differently depending on the conditions it was reared in, and different genotypes responded to environmental changes in different ways.²¹ This kind of gene-by-environment interaction is a confounding factor in genetic studies and a major reason why knowing someone’s DNA alone is often not enough to predict their traits with precision.

For humans, this means that inheriting a genetic predisposition for something, whether height, disease risk, or temperament, sets a range of possibilities rather than a fixed outcome. Nutrition, stress, toxin exposure, and countless other environmental variables determine where within that range you actually land. Two people carrying the same genetic variants for height will end up at different heights if one grew up malnourished and the other did not. The genes are inherited; the trait is built.

Why Not Every Mutation Gets Passed On

Your body accumulates DNA mutations throughout your lifetime as cells divide. But the vast majority of these mutations are irrelevant to your children because they occur in somatic cells, the cells that make up your skin, liver, brain, and everything else besides eggs and sperm. Only mutations in the germline, the cells that produce eggs and sperm, have any chance of reaching the next generation.

Research comparing mutation rates between these two cell types has confirmed that somatic cells accumulate mutations at significantly higher rates than germline cells.²² This makes biological sense: the body invests heavily in protecting the DNA that will be transmitted to offspring while treating the rest of the body as, in evolutionary terms, disposable. Germline cells have more robust DNA repair machinery and tighter error-checking during replication. A mutation in a skin cell might contribute to a mole or eventually a cancer in your body, but it dies with you. A mutation in a developing sperm cell can echo through every generation that follows.

How Organisms That Skip Sex Handle Inheritance

Everything discussed so far assumes sexual reproduction, where two parents combine genetic material. But plenty of organisms reproduce asexually, creating offspring that are essentially clones of the parent. Bacteria, many fungi, some plants, and a few animals produce offspring by copying their genome directly rather than shuffling two parents’ DNA together.

The tradeoff is significant. Sexual reproduction generates new genetic combinations every generation through chromosome shuffling and segment swapping, which gives populations the raw variation needed to adapt to changing conditions. Asexual lineages lose that advantage. Without recombination, harmful mutations tend to pile up over generations with no way to purge them, and beneficial combinations cannot be assembled as quickly.²³ This is one reason most complex multicellular organisms reproduce sexually.

Fungi blur the line in interesting ways. Roughly one-fifth of known fungal species were historically thought to be purely asexual, but genetic studies reveal that many of them are actually recombining their DNA, just through cryptic or rare sexual cycles rather than the obvious ones seen in animals and plants.²⁴ Pure clonal inheritance is rarer than it looks on the surface.

Inheriting More Than Genes Alone

Genetic inheritance usually refers to DNA, but organisms also inherit other biological entities from their parents, most prominently microbes. Many insects, for example, depend on intracellular bacteria that provide essential nutrients or protect against parasites. These bacterial symbionts are transmitted from mother to offspring through the eggs, a vertical inheritance route that mirrors the way mitochondrial DNA is passed down.²⁵

This mode of transmission has consequences for both the host and the symbiont. Maternal transmission acts as a filter: symbionts that harm the host tend to get eliminated over evolutionary time because they reduce the host’s reproductive success and thus their own transmission. Beneficial symbionts, by contrast, spread to fixation in a population.²⁶ In aphids, a defensive bacterial symbiont that protects against parasitic wasps has been experimentally shown to reinfect developing embryos after being introduced to uninfected adults, and these infections persist across subsequent generations through the egg.²⁷

On an even larger evolutionary scale, entire stretches of DNA can jump between distantly related species. Transposable elements, sometimes called jumping genes, have been documented transferring horizontally between vertebrate lineages that diverged over 120 million years ago. A large-scale screen of over 300 vertebrate genomes found evidence for nearly a thousand such transfer events, though they are overwhelmingly concentrated in fish and rare in mammals and birds.²⁸ Viruses appear to act as vectors in some cases, and horizontally acquired sequences have occasionally been co-opted into new functions by the host genome.²⁹ This is not inheritance in the parent-to-child sense, but it is a real route by which organisms acquire heritable genetic material.

Predicting Traits from DNA

Given how many layers influence trait inheritance, you might wonder how well modern genetics can actually predict what a person will be like from their DNA. For single-gene conditions like sickle cell disease or cystic fibrosis, genetic testing is quite reliable because one gene with a known variant drives the outcome. For polygenic traits, the picture is murkier.

Researchers build polygenic risk scores by adding up the effects of thousands of genetic variants to estimate a person’s predisposition to a trait or disease. These scores have grown more accurate as studies get larger, but the pace of improvement has slowed. Simply recruiting bigger study populations is yielding only modest gains in predictive power at this point.³⁰ Increasing the range of genetic variants included in the score, by using whole-genome sequencing data rather than common-variant chips, appears to be the more promising path forward. Even so, polygenic scores currently work better for identifying statistical associations across large groups than for predicting any individual person’s outcome.³¹

The gap between group-level prediction and individual-level certainty reflects everything covered in this article. Gene-by-environment interactions, epigenetic variation, mitochondrial genetics, imprinting effects, and sheer randomness in development all shape traits in ways that DNA sequence alone cannot fully capture. Inheritance gives you a set of biological starting materials. What gets built from them is another story.