Incomplete dominance, codominance, and the classical laws Gregor Mendel described in the 1860s are not three competing systems of inheritance. They are different facets of the same underlying process. Mendel’s laws govern how alleles are passed from parent to offspring; incomplete dominance and codominance describe how those alleles interact once they land in the same organism. The confusion between them is one of the most persistent in genetics education, and a 2024 paper in the journal Genetics argued that widely used learning resources are partly to blame.
What Mendel’s Laws Actually Say
Mendel’s three laws deal with the mechanics of inheritance, not with what a trait looks like. The Law of Dominance says that one allele can mask the expression of another. The Law of Segregation says that allele pairs split apart during the formation of reproductive cells, so each egg or sperm carries only one allele for a given gene. The Law of Independent Assortment says that alleles for different genes are sorted into reproductive cells independently of one another.
1PubMed. Genetics: Inheritance PatternsThese laws describe the choreography of alleles as they move between generations. They do not dictate the specific outcome when two different alleles meet inside a cell. That outcome, the visible trait the organism ends up with, depends on the molecular details of the alleles involved: how much protein each one produces, whether those proteins interfere with each other, and how the cell’s machinery reads them. This is where incomplete dominance and codominance come in, and why they do not contradict Mendel at all.
Complete Dominance and Why Mendel Saw It First
In complete dominance, one allele’s effect entirely masks the other. A pea plant with one allele for purple flowers and one for white flowers looks purple, full stop. The white allele is still there, still gets passed to offspring, still follows segregation perfectly. It just does not produce enough of whatever protein it encodes, or its product is nonfunctional, so the purple allele runs the show.
Mendel happened to study traits in peas that mostly behave this way, which made the patterns clean and countable. It was partly good experimental design and partly good luck. Many traits in nature do not divide so neatly. The key insight, though, is that the rules about how alleles are distributed to offspring have nothing to do with whether the resulting trait looks cleanly dominant, blended, or dual. The distribution rules still hold.
Incomplete Dominance Produces a Blend
In incomplete dominance, neither allele fully masks the other. Instead, the organism with one copy of each allele shows an intermediate trait, something between the two extremes. The textbook example is the snapdragon: cross a red-flowered plant with a white-flowered plant, and you get pink offspring. The pink flowers are not some third, independent trait. They are the result of having half the dose of pigment that a red-flowered plant produces.
Research on morning glories has shown this mechanism at the molecular level. In the common morning glory Ipomoea purpurea, flowers that carry one copy of a particular pigment gene produce roughly half the pigment-related protein and about half the anthocyanin pigment found in flowers with two copies of that gene. The intermediate color comes directly from getting half the molecular raw material.
2PubMed. The molecular basis of incomplete dominance at the A locus of CHS-D in the common morning glory, Ipomoea purpureaThis dose-dependent logic explains why incomplete dominance is common for traits tied to pigment, enzyme activity, or structural protein quantity. When having two working copies of a gene gives you twice the output, having one copy gives you something visibly in between. The alleles still segregate according to Mendel’s rules. Cross two pink snapdragons and you get roughly one-quarter red, one-half pink, and one-quarter white offspring, a ratio that reflects textbook Mendelian segregation at the genetic level even though the appearance of the offspring does not follow the classic 3:1 phenotypic ratio.
Codominance Means Both Alleles Show Up Fully
Codominance is different from incomplete dominance in a specific way: instead of a blended middle ground, both alleles express their products simultaneously and independently, and you can detect each one. There is no averaging. Both are fully present.
The most familiar human example is the ABO blood group. A person who inherits one A allele and one B allele does not end up with some intermediate blood type. They have type AB blood, meaning their red blood cells carry both A-type and B-type surface molecules at the same time. A blood test can pick up each one separately. That is the hallmark of codominance: both allele products coexist without diluting or overriding each other.
At the molecular level, codominance was demonstrated in early research on tissue-compatibility genes in mice. In cells that carry two different versions of the H-2 gene, researchers detected separate glycoprotein molecules from each version, whereas cells with two identical copies only produced one type. Both parental alleles were expressed as distinct proteins on the cell surface.
3PubMed Central. The molecular basis of codominant expression of the histocompatibility-2 genetic regionThis pattern has also been found in single-celled organisms. In the ciliate Euplotes crassus, cells with two different mating-type alleles secrete distinct pheromones from each allele simultaneously, rather than producing one dominant pheromone or a hybrid. Neither allele takes priority over the other.
4PubMed. Evidence for gene duplication and allelic codominance (not hierarchical dominance) at the mating-type locus of the ciliate, Euplotes crassusThe Practical Difference Between Incomplete Dominance and Codominance
If you are staring at an organism that has two different alleles and wondering which pattern you are looking at, the question to ask is straightforward: can you tell the two allele products apart, or do you see a single blended result?
- Blended trait: The organism shows an intermediate phenotype, and you cannot separately identify either allele’s individual contribution. A pink flower from a red-white cross is pink everywhere; there is no way to point to “the red part” and “the white part.” This is incomplete dominance.
- Both traits visible: The organism displays both allele products side by side, and you can identify each one independently. AB blood shows both A and B markers on every red blood cell. A roan cow has distinct red hairs and white hairs intermixed, not a uniform pink coat. This is codominance.
The distinction sometimes collapses depending on what level you examine. A roan horse, for instance, looks codominant at the level of individual hairs (each hair is either red or white) but might look incompletely dominant from a distance (the coat appears pinkish). The label you use depends on the resolution of the observation, which is why genetics textbooks occasionally disagree on borderline cases.
Why Calling These “Non-Mendelian” Is Misleading
Many online study guides and even some textbooks list incomplete dominance and codominance as examples of “non-Mendelian inheritance.” A 2024 perspective in the journal Genetics argued that this is flatly wrong and urged educators and content creators to correct it. The paper pointed out that while these patterns produce phenotypic ratios that differ from the classic 3:1 ratio Mendel observed, the alleles involved still segregate during meiosis and still produce Mendelian genotypic ratios in offspring.
5PubMed Central. Clarifying Mendelian vs non-Mendelian inheritance – Section: Extensions of Mendel’s laws and variations in phenotypic ratiosThe confusion comes from conflating “Mendelian ratios” with “Mendelian inheritance.” Mendel’s core principles are about how alleles segregate and assort. When you cross two pink snapdragons, you get a 1:2:1 phenotypic ratio instead of 3:1, but at the genotype level, allele segregation follows Mendel’s law exactly. The offspring are one-quarter homozygous for one allele, one-half heterozygous, and one-quarter homozygous for the other. That is textbook Mendelian segregation. It is only the mapping from genotype to phenotype that differs.
Truly non-Mendelian inheritance involves mechanisms that violate the segregation or assortment rules themselves. Examples include mitochondrial inheritance, where genetic material passes exclusively through the mother and does not segregate in the Mendelian sense, or genomic imprinting, where the expression of an allele depends on which parent it came from. These genuinely break the symmetry Mendel’s laws assume.
6Human Molecular Genetics. Mechanisms of non-Mendelian inheritance in genetic diseaseWhy Dominance Exists in the First Place
It might seem arbitrary that some alleles are dominant, some are codominant, and some show incomplete dominance. But there is a biochemical logic to it. In metabolic pathways, enzymes often work with more capacity than they strictly need. If an organism has two copies of an enzyme gene and one copy stops working, the remaining copy can still produce enough enzyme to keep the pathway running at near-normal speed. The organism looks fine, so the working allele appears “dominant.” This excess capacity, sometimes called enzymatic buffering, is a natural feature of metabolic networks.
7PubMed Central. Evolution of dominance in metabolic pathwaysWhen an enzyme operates well below its maximum capacity, losing one gene copy does not reduce the pathway’s output much, and complete dominance results. When the enzyme is already working near capacity, losing one copy cuts output noticeably, and you see incomplete dominance. Codominance tends to arise when the two alleles produce structurally distinct proteins that function independently, as with the A and B blood-type enzymes, each of which adds a different sugar to the cell surface regardless of what the other is doing.
The dominance relationship is also not always fixed. In some cases, whether an allele behaves as dominant can be influenced by the organism’s genetic background or environmental conditions. In agricultural crops, for instance, the effects of genes on traits are commonly modulated by environmental conditions, so that the same cross can yield slightly different dominance patterns in different growing environments.
8PubMed Central. A reaction norm model for genomic selection using high-dimensional genomic and environmental dataHaploinsufficiency and When One Working Copy Is Not Enough
The discussion above assumes that one working allele usually compensates for one broken allele. For many genes, that is true. But for some, a single working copy simply cannot produce enough protein to keep things running normally. This is called haploinsufficiency, and it makes a loss-of-function mutation behave as dominant: carriers of just one defective copy show symptoms.
9PubMed. Mechanisms of Mendelian dominanceHaploinsufficiency is why some genetic conditions follow a dominant inheritance pattern even though the mutation itself is a simple loss of function. Marfan syndrome, for example, results from having just one faulty copy of the gene for the connective-tissue protein fibrillin-1. The remaining good copy does not produce enough fibrillin-1 to maintain normal tissue structure, so carriers develop symptoms. Understanding whether a gene is haploinsufficient matters for genetic counseling: it determines whether a person with one copy of a mutation is likely to be affected or merely a silent carrier.
Penetrance, Expressivity, and Why Families Surprise Geneticists
Even when you know the dominance relationship of an allele and its inheritance pattern, predicting the actual outcome in a specific person is harder than it sounds. Two additional phenomena complicate real-world genetics.
Penetrance refers to whether a genotype produces any observable effect at all. An allele can be dominant in theory, but not every person carrying it develops the associated trait. When some carriers show the trait and others do not, the allele has incomplete penetrance. Variable expressivity is a related but distinct concept: among people who do show the trait, the severity or specific features can vary widely. The same genetic variant can cause anything from no discernible clinical effect to severe disease, even within the same family.
10PubMed Central. Incomplete Penetrance and Variable Expressivity: From Clinical Studies to Population CohortsThese phenomena are well documented in families carrying chromosome abnormalities. A parent may pass on the same unbalanced chromosomal change to several children, yet those children can develop vastly different symptoms or no symptoms at all.
11PubMed. Incomplete penetrance, variable expressivity, or dosage insensitivity in four families with directly transmitted unbalanced chromosome abnormalitiesPenetrance and expressivity are shaped by the rest of a person’s genome, by environmental factors, and sometimes by chance events during development. They are the reason that knowing someone’s genotype at one location does not always tell you what their body will do with it. Mendel’s peas were conveniently high-penetrance and low-variability, which is part of why his results were so clean. Most human traits are messier.
Genomic Imprinting as a Genuinely Non-Mendelian Wrinkle
If incomplete dominance and codominance follow Mendel’s laws despite altering the expected phenotypic ratios, what does genuinely non-Mendelian inheritance look like? Genomic imprinting is one of the clearest examples. In imprinted genes, only the copy from one specific parent is active; the other copy is chemically silenced regardless of what allele it carries. This means the same genotype can produce different outcomes depending on whether the active allele came from the mother or the father.
Research in mice selected for divergent body weight has quantified the relative contributions of different genetic effects. Across 18 identified loci on 13 chromosomes, straightforward additive effects (the kind Mendel would recognize) accounted for most of the variation in body weight throughout development. Imprinting effects were consistently small.
12PubMed Central. Relative contribution of additive, dominance, and imprinting effects to phenotypic variation in body size and growth between divergent selection lines of miceThat finding is reassuring in a practical sense: for most quantitative traits, Mendelian-style effects are still doing the heavy lifting. Imprinting, while real and medically important in specific disorders, does not usually override the bigger picture. This is another reason geneticists push back against lumping incomplete dominance and codominance with truly non-Mendelian mechanisms. The scale of the departure from Mendel’s framework is completely different.
Why Balancing Selection Keeps Multiple Alleles Around
You might wonder why codominance and incomplete dominance persist in populations rather than one allele simply winning out over evolutionary time. Part of the answer is balancing selection, a set of evolutionary forces that actively maintain genetic diversity. When carrying two different alleles provides a survival or reproductive advantage, or when the benefit of an allele depends on how common it is in the population, natural selection works to keep multiple alleles in circulation rather than fixing one and discarding the rest.
13PubMed Central. On the importance of balancing selection in plantsThe sickle-cell allele in humans is a well-known example from medicine. People with two copies of the sickle-cell allele develop severe disease, but those with one sickle-cell allele and one normal allele gain some resistance to malaria. In regions where malaria is prevalent, this heterozygote advantage keeps both alleles in the population. The dominance relationship here is itself a mix: for red blood cell shape, the sickle allele shows incomplete dominance under low-oxygen conditions, but for malaria resistance, having one copy confers a clear benefit. The “dominance” of an allele is not a fixed property of the allele itself but of the trait you are measuring.
Implications for Plant Breeding and Medicine
Understanding whether a trait follows complete dominance, incomplete dominance, or codominance has direct practical consequences. In plant breeding, a trait governed by complete dominance can be tricky to work with because you cannot visually distinguish an organism with two dominant alleles from one with only one dominant allele; both look the same. With incomplete dominance, breeders can identify heterozygous plants by their intermediate phenotype, making selection easier. Knowing the dominance pattern informs decisions about crossing schemes and how many generations of selection it takes to stabilize a variety.
In medicine, the distinction matters for genetic counseling. If a disease allele is fully recessive, carriers show no symptoms and may not know they are carriers without testing. If the allele shows incomplete dominance, carriers may have mild or subclinical features that a clinician can catch. If two alleles are codominant, as with the hemoglobin variants underlying sickle-cell trait, a simple lab test can detect both gene products and identify carriers definitively. Each pattern demands a different counseling and screening strategy.
Crop improvement programs are increasingly interested in phenomena beyond simple Mendelian dominance. Mechanisms like cytoplasmic inheritance and epigenetic modification do not follow Mendel’s segregation rules, and exploiting them requires entirely different breeding approaches than those designed around nuclear-gene dominance.
14PubMed Central. Unveiling the Mysteries of Non-Mendelian Heredity in Plant BreedingWhen the Label Depends on the Instrument
One last detail that rarely makes it into introductory courses: whether you call an allele interaction “dominant,” “codominant,” or “incompletely dominant” can depend on how you measure the trait. Consider the sickle-cell example again. If you look at a person’s overall health, the normal hemoglobin allele appears dominant because carriers are generally healthy. If you look at their red blood cells under a microscope after stressing them with low oxygen, you can see both normal and sickle-shaped cells, which looks like codominance. If you run a hemoglobin electrophoresis, you detect both hemoglobin A and hemoglobin S as distinct bands, which is unambiguous codominance at the molecular level.
The alleles have not changed. The inheritance has not changed. What changed is the resolution of your observation. This is why professional geneticists are cautious about labeling dominance relationships in absolute terms. The relationship between alleles is real, but the label we give it is partly an artifact of our measurement tools and the level of biological organization we choose to examine. For any given gene, asking “is it dominant?” is less useful than asking “dominant with respect to which phenotype, measured how?”