COLLEGE BIOLOGY • INHERITANCE & GENETICS

Non-Mendelian Genetics

How incomplete dominance, codominance, epistasis, and polygenic inheritance reveal that most traits defy simple dominant–recessive rules.

Historical Context & Motivation

Gregor Mendel's meticulous pea-plant experiments, published in 1866, established the foundational principles of heredity: the laws of segregation and independent assortment. These principles predicted clear-cut, either-or phenotypic ratios — tall or short, round or wrinkled — governed by pairs of alleles exhibiting complete dominance. When Mendel's work was rediscovered at the turn of the twentieth century by de Vries, Correns, and von Tschermak, geneticists immediately began testing its predictions across a broader range of organisms. They quickly discovered that many traits did not sort neatly into Mendelian ratios. Flower colors blended, coat patterns displayed both parental phenotypes simultaneously, and skin pigmentation varied along a continuous gradient. These observations signaled that Mendel's framework, while correct in its core logic, was incomplete. The study of non-Mendelian genetics emerged to explain the myriad inheritance patterns that lie beyond simple dominance.

1900
Rediscovery of Mendel's Work
Hugo de Vries, Carl Correns, and Erich von Tschermak independently rediscover Mendel's principles, spurring systematic genetic research across plant and animal species and revealing the first deviations from expected ratios.
1906
Bateson & Punnett: Gene Interaction
William Bateson and Reginald Punnett observe unexpected phenotypic ratios in sweet pea flower color, providing the first evidence of epistasis — the interaction between genes at different loci.
1908
Nilsson-Ehle: Polygenic Inheritance
Herman Nilsson-Ehle demonstrates that wheat kernel color is controlled by multiple genes, each contributing additively to the phenotype — the foundation of polygenic inheritance.
1910
Morgan: Sex-Linked Inheritance
Thomas Hunt Morgan's white-eyed Drosophila experiments reveal that certain genes reside on sex chromosomes, producing inheritance patterns that differ between males and females — a clear violation of Mendel's assumption of autosomal loci.
1940s–1960s
Quantitative Genetics Matures
R.A. Fisher, Sewall Wright, and J.B.S. Haldane develop the statistical framework connecting Mendelian inheritance to continuous phenotypic variation, bridging the gap between classical genetics and Darwinian evolution.

The central question that unites all non-Mendelian phenomena is straightforward: if Mendel's laws accurately describe allele segregation and assortment, why do so many crosses produce phenotypic ratios that deviate from 3:1 or 9:3:3:1? The answer lies in the molecular and cellular complexity of gene expression — from allelic interactions within a single locus to multi-gene networks, cytoplasmic inheritance, and environmental modulation. Understanding these mechanisms is essential for interpreting genetic data in clinical diagnostics, agriculture, and evolutionary biology.

Core Principles of Non-Mendelian Inheritance

Non-Mendelian genetics encompasses a diverse set of inheritance patterns that share one feature: the phenotypic outcome cannot be predicted by simple dominant–recessive logic at a single autosomal locus. While the underlying Mendelian mechanics of meiosis — chromosome segregation and independent assortment — remain intact, the relationship between genotype and phenotype is modified by molecular events at the level of protein function, gene regulation, or organellar inheritance. The following core concepts form the foundation for analyzing non-Mendelian patterns.

1

Incomplete Dominance

The heterozygote displays a blended intermediate phenotype between the two homozygous forms. Neither allele fully masks the other; the single functional allele produces insufficient protein to achieve the homozygous dominant phenotype. Classic example: red × white snapdragons yield pink F₁ offspring.
2

Codominance

Both alleles are fully and independently expressed in the heterozygote; the phenotype is not a blend but a simultaneous display of both allelic products. Human ABO blood type illustrates this: the IAIB genotype produces both A and B surface antigens.
3

Epistasis

The expression of one gene is modified or masked by another gene at a different locus. This generates dihybrid ratios that deviate from the classic 9:3:3:1 — for example, 9:3:4 in recessive epistasis or 12:3:1 in dominant epistasis.
4

Polygenic Inheritance

Multiple genes contribute additively to a single continuous trait. The resulting phenotypic distribution approximates a bell curve rather than discrete classes. Human height, skin color, and blood pressure are polygenic traits influenced by dozens to hundreds of loci.
5

Sex-Linked & Extranuclear Inheritance

Genes on the X or Y chromosome produce sex-dependent inheritance patterns, while mitochondrial and chloroplast DNA follow strictly maternal transmission. These patterns violate Mendel's assumption that reciprocal crosses yield identical results.
KEY TAKEAWAY
Think of Mendelian genetics as a light switch — the trait is either on or off. Non-Mendelian genetics is more like a mixing board in a recording studio: multiple channels (genes) at different levels (allele dosages) combine, interact, and sometimes override each other to produce the final sound (phenotype). The underlying wiring (meiosis and segregation) hasn't changed; the complexity of the output has simply increased.

Visual Explanation: Allelic Interaction Spectrum

The distinction between complete dominance, incomplete dominance, and codominance is best understood as a spectrum of allelic interaction. All three phenomena involve the same Mendelian cross structure — a monohybrid cross between two homozygous parents — but they differ in how the heterozygous F₁ genotype maps to phenotype. The diagram below illustrates this progression using a classic flower-color cross, highlighting how the F₂ phenotypic ratios shift from 3:1 (complete dominance) to 1:2:1 (incomplete dominance) while the genotypic ratio remains 1:2:1 in every case.

All three allelic interaction types begin with the same cross structure and produce identical genotypic ratios (1:2:1) in the F₂ generation. The critical difference lies in how the heterozygous genotype maps to phenotype: masked in complete dominance, blended in incomplete dominance, and simultaneously expressed in codominance.

Notice that in all three panels, the genotypic ratio of the F₂ generation is 1:2:1, because allele segregation during meiosis is identical regardless of the dominance relationship. What changes is the phenotypic ratio: in complete dominance, the heterozygote is indistinguishable from the homozygous dominant, collapsing the ratio to 3:1. In incomplete dominance and codominance, the heterozygote produces a distinct third phenotype — whether blended or dual-expressed — preserving the 1:2:1 ratio at the phenotypic level. This insight is fundamental: non-Mendelian inheritance does not contradict Mendel's law of segregation; it modifies only the genotype-to-phenotype mapping.

Molecular Mechanisms & Mathematical Framework

Non-Mendelian inheritance patterns arise from specific molecular mechanisms that alter the simple one-gene, one-trait, complete-dominance model. Understanding these mechanisms not only explains why phenotypic ratios deviate from Mendelian predictions but also provides the mathematical tools for predicting offspring outcomes in more complex genetic scenarios.

Multiple Alleles & ABO Blood Typing

While any individual organism carries at most two alleles for a given autosomal locus, a population may harbor more than two allelic variants. The ABO blood group system is the classic example: the I locus has three alleles — IA, IB, and i. The IA and IB alleles encode glycosyltransferases that add distinct sugar residues to the H antigen on red blood cells, whereas the i allele is a loss-of-function variant. IA and IB are codominant to each other (producing type AB in heterozygotes), while both are completely dominant over i.

ABO Blood Group Genotypes, Phenotypes, and Serum Antibodies
GenotypeAntigens PresentBlood TypeAntibodies in Serum
IAIA or IAiA antigenType AAnti-B
IBIB or IBiB antigenType BAnti-A
IAIBA and B antigensType ABNeither
iiNeither (H antigen only)Type OAnti-A and Anti-B

Quantitative Framework for Polygenic Inheritance

For polygenic traits, each contributing locus typically has two alleles — one that adds to the phenotypic value and one that does not. If we assume additive alleles with equal and interchangeable effects, we can model the number of phenotypic classes and their distribution.

PHENOTYPIC CLASSES
Number of phenotypic classes = 2n + 1
where n = number of gene loci contributing to the trait. For example, 3 loci yield 2(3) + 1 = 7 phenotypic classes.
FRACTION WITH EXTREME PHENOTYPE
P(extreme) = (1/4)ⁿ
The probability of obtaining either extreme phenotype (all contributing alleles or none) in a cross between heterozygotes at all n loci. For 3 loci: (1/4)³ = 1/64.

Epistatic Ratios

In a standard dihybrid cross (AaBb × AaBb), the expected Mendelian F₂ phenotypic ratio is 9:3:3:1. Epistasis modifies this ratio by collapsing certain phenotypic categories. The specific collapsed ratio reveals the type of epistasis at work. Recessive epistasis (e.g., Labrador coat color) yields 9:3:4 because the homozygous recessive genotype at the epistatic locus masks all expression at the hypostatic locus. Dominant epistasis yields 12:3:1, and duplicate gene interaction produces a 15:1 ratio. In each case, the total still sums to 16 parts, reflecting the 4 × 4 Punnett square.

EPISTATIC RATIO VERIFICATION
Sum of ratio parts = 4ⁿ (where n = number of loci)
For a dihybrid cross (n = 2), any epistatic ratio must sum to 4² = 16. This serves as a quick check: 9 + 3 + 4 = 16 ✓, 12 + 3 + 1 = 16 ✓.

Epistasis: Classification & Modified Dihybrid Ratios

Epistasis encompasses several distinct interaction patterns, each producing a characteristic deviation from the standard 9:3:3:1 dihybrid ratio. Correctly identifying the type of epistasis from observed phenotypic data is a core skill in genetics. The diagram and table below systematize the major epistatic categories, showing how specific genotypic classes merge to produce modified ratios.

The top section shows how the standard 9:3:3:1 Mendelian ratio branches into four common epistatic modifications. The bottom Punnett square illustrates recessive epistasis in Labrador retrievers: the 3 aaB_ and 1 aabb classes (pink) all express the same epistatic phenotype (yellow), collapsing into a combined class of 4.
Summary of Common Epistatic Interactions and Their Modified Dihybrid Ratios
Epistasis TypeModified RatioMechanismClassic Example
Recessive9:3:4Homozygous recessive at epistatic locus masks hypostatic locusLabrador coat color (TYRP1 + MC1R pathway)
Dominant12:3:1Dominant allele at epistatic locus masks hypostatic locusSquash fruit color
Complementary9:7Functional alleles at both loci required for phenotype expressionSweet pea flower pigmentation
Duplicate gene15:1Dominant allele at either locus sufficient to produce phenotypeWheat kernel shape
Duplicate recessive9:3:3:1 → 9:6:1A_ and B_ produce same phenotype independently; aabb is distinctShepherd's purse seed capsule shape

Worked Example: Predicting Offspring from a Codominant Cross with Multiple Alleles

A woman with type A blood (whose father was type O) marries a man with type AB blood. Determine the probability that their first child will have type B blood.

ABO Blood Type Cross
1
Step 1 — Determine Parental GenotypesThe woman has type A blood, so her genotype is either IAIA or IAi. Since her father was type O (genotype ii), he must have contributed an i allele to her. Therefore, the woman's genotype is IAi. The man with type AB blood has genotype IAIB (the only genotype producing type AB).
Mother: IAi × Father: IAIB
2
Step 2 — Set Up the Punnett SquareThe mother produces gametes IA and i, each with probability 1/2. The father produces gametes IA and IB, each with probability 1/2. The four possible offspring genotypes are: IAIA (Type A), IAIB (Type AB), IAi (Type A), and IBi (Type B).
3
Step 3 — Identify Phenotypic ProbabilitiesFrom the Punnett square: Type A = IAIA (1/4) + IAi (1/4) = 1/2. Type AB = IAIB = 1/4. Type B = IBi = 1/4. Type O = 0 (no ii genotype possible).
P(Type B) = 1/4 = 25%
4
Step 4 — Verify the ResultCheck: 1/2 (Type A) + 1/4 (Type AB) + 1/4 (Type B) + 0 (Type O) = 1.0 ✓. The probabilities sum to 1, confirming our Punnett square is complete. Notice that this cross cannot produce a type O child because the father carries no i allele; every offspring inherits at least one dominant allele (IA or IB) from the father.
The probability of their first child having type B blood is 1/4 (25%).

Mendelian vs. Non-Mendelian: Strengths & Limitations

Classical Mendelian genetics provides an elegant and powerful framework for predicting inheritance of traits governed by single genes with complete dominance. However, most phenotypic variation in natural populations cannot be explained by this model alone. The following comparison highlights where each framework excels and where it falls short, helping geneticists choose the appropriate analytical tools for different scenarios.

Comparison of Mendelian and Non-Mendelian Genetic Frameworks
FeatureMendelian GeneticsNon-Mendelian Genetics
Dominance relationshipComplete dominance only; heterozygote = homozygous dominantIncomplete dominance, codominance, overdominance
Number of allelesTwo alleles per locus in the populationMultiple alleles per locus (e.g., ABO with three)
Gene interactionsEach gene acts independently (independent assortment)Epistasis, complementation, suppression between loci
Phenotypic distributionDiscrete classes (tall/short, round/wrinkled)Continuous distributions for polygenic traits
Chromosome locationAutosomal loci assumedX-linked, Y-linked, and organellar DNA inheritance
Environmental influenceNot considered; genotype → phenotype is deterministicPleiotropy, penetrance, expressivity, and norm of reaction
Predictive powerExcellent for simple single-gene traits (e.g., CF, Huntington's)Required for complex traits (height, disease susceptibility)
KEY TAKEAWAY
Mendelian genetics and non-Mendelian genetics are not competing theories — they form a nested hierarchy. Mendel's laws describe the behavior of chromosomes during meiosis, which remains universally valid. Non-Mendelian inheritance patterns arise from molecular complexity layered on top of that chromosomal framework. Think of it like civil engineering: Mendel described the structural steel beams (chromosome mechanics), while non-Mendelian genetics accounts for the plumbing, wiring, and HVAC systems (protein interactions, regulatory networks, and environmental inputs) that determine how the building actually functions.

Connections to Quantitative Genetics & Genomics

Non-Mendelian inheritance forms the conceptual bridge between classical genetics and the modern disciplines of quantitative genetics and genomics. While this lesson has focused on identifiable loci with discrete allelic effects, most medically and agriculturally important traits are influenced by hundreds or thousands of loci, each with small effect sizes, interacting with environmental variables. Genome-wide association studies (GWAS) have revealed that traits such as height, body mass index, and susceptibility to type 2 diabetes involve extensive polygenic architecture combined with epistatic interactions that cannot be captured by single-locus models.

From Classical Non-Mendelian Concepts to Modern Genomics
ConceptClassical Non-Mendelian ViewModern Genomics Extension
Polygenic inheritance2–5 loci with additive alleles; bell-curve phenotypic distributionPolygenic risk scores from thousands of SNPs; machine learning models for phenotype prediction
EpistasisTwo-locus interactions producing modified dihybrid ratiosHigher-order epistasis networks mapped via computational biology; G×G interaction terms in statistical models
EpigeneticsNot originally in scope of non-Mendelian geneticsDNA methylation, histone modification, and transgenerational epigenetic inheritance as additional layers of phenotypic regulation
Penetrance & expressivityQualitative observation that not all carriers show the phenotypeQuantified as probabilistic functions of genetic background, modifier loci, and environmental exposure

As you advance through genetics coursework, you will encounter the infinitesimal model proposed by R.A. Fisher, which treats phenotypic variation as the sum of infinitely many loci each with infinitely small effect — a mathematical idealization of polygenic inheritance that underpins modern animal and plant breeding programs. You will also explore heritability (both broad-sense and narrow-sense) as a statistical measure partitioning phenotypic variance into genetic and environmental components. These tools represent the natural extension of the non-Mendelian concepts introduced here, scaled up to the genome-wide level.

Practice Problems

PROBLEM 1CONCEPTUAL
Explain why incomplete dominance and codominance both produce a 1:2:1 phenotypic ratio in the F₂ generation, whereas complete dominance produces a 3:1 phenotypic ratio, despite all three inheritance patterns producing a 1:2:1 genotypic ratio. What molecular distinction accounts for the difference in how the heterozygous genotype maps to phenotype?
PROBLEM 2BASIC CALCULATION
In four-o'clock flowers, red (CRCR) is incompletely dominant over white (CWCW), producing pink heterozygotes. If two pink flowers are crossed, what fraction of the F₂ offspring will be pink?
PROBLEM 3INTERMEDIATE
In Labrador retrievers, coat color is determined by two genes: the E locus (E = pigment deposited, ee = yellow regardless of B genotype) and the B locus (B_ = black, bb = brown). This is an example of recessive epistasis. If two dogs with genotype BbEe are crossed, what fraction of the puppies will be brown? What fraction will be yellow?
PROBLEM 4APPLIED
Skin pigmentation in humans is modeled as a polygenic trait controlled by at least 3 independently assorting loci, each with two alleles (contributing allele = uppercase, non-contributing = lowercase). Assume additive allelic effects and no environmental variation. If two individuals who are heterozygous at all three loci (AaBbCc × AaBbCc) have children, what is the probability that a child will have the darkest possible skin phenotype (6 contributing alleles)? How many distinct phenotypic classes would be expected?
PROBLEM 5CRITICAL THINKING
A researcher crosses two pure-breeding white-flowered plants and obtains an F₁ generation that is entirely purple. When the F₁ plants are self-crossed, the F₂ generation displays a 9:7 ratio of purple to white. Propose a genetic model explaining these results. Include the genotypes of the parental lines, the F₁ generation, and the molecular logic behind why two white parents produce purple offspring.

Lesson Summary

Non-Mendelian genetics extends classical inheritance beyond the limitations of the simple dominant–recessive model. Incomplete dominance produces blended intermediate phenotypes in heterozygotes, while codominance results in simultaneous expression of both allelic products, as exemplified by the ABO blood group system with its three alleles (IA, IB, i). Epistasis describes interactions between genes at different loci that modify the standard 9:3:3:1 dihybrid ratio into characteristic patterns such as 9:3:4 (recessive epistasis), 12:3:1 (dominant epistasis), 9:7 (complementary interaction), and 15:1 (duplicate gene interaction). Polygenic inheritance underlies continuous phenotypic variation by distributing trait control across multiple additive loci, producing 2n + 1 phenotypic classes that approximate a normal distribution.

Critically, these non-Mendelian patterns do not contradict Mendel's laws of segregation and independent assortment — alleles still separate during meiosis and assort independently when on different chromosomes. What changes is the genotype-to-phenotype mapping, which is shaped by protein biochemistry, gene regulatory networks, chromosomal location (sex-linked inheritance), and organellar DNA (maternal inheritance). Understanding these extensions is essential for interpreting genetic data in clinical diagnostics, agricultural breeding, forensic science, and evolutionary biology — and provides the conceptual foundation for modern quantitative genetics and genomics.

Varsity Tutors • College Biology • Non-Mendelian Genetics