Gregor Mendel and the laws of inheritance: the pea experiments and what came after

Gregor Mendel was a friar and part-time science teacher in what is now the Czech Republic. For about eight years he crossed pea plants in a monastery garden, counted the offspring with unusual care, and found that heredity follows simple numerical rules. He published in 1866. Almost nobody understood what he had done until 1900, sixteen years after his death.
This piece follows the story as a timeline, then explains the experiments, the two laws in modern words, the patterns that do not fit them, and how his ratios grew into population genetics.
Mendel and his legacy, year by year
| Year | What happened |
|---|---|
| 1822 | Johann Mendel is born on July 20 into a farming family in Heinzendorf, Silesia, then part of the Austrian Empire |
| 1843 | He enters the Augustinian Abbey of St Thomas in Brno (then Brünn) and takes the name Gregor |
| 1850 | He fails part of the exam to become a certified teacher |
| 1851-1853 | The abbey sends him to the University of Vienna, where his teachers include the physicist Christian Doppler |
| 1856 | He fails the teaching exam a second time, and around now begins systematic pea crosses |
| 1856-1863 | He grows and tests roughly 28,000 pea plants |
| 1865 | He presents his results at two meetings of the Natural History Society of Brno, on February 8 and March 8 |
| 1866 | His paper, "Experiments on Plant Hybridization", appears in the society's proceedings |
| 1866-1873 | He corresponds with the botanist Carl Nägeli, who encourages work on hawkweed; the results seem to contradict the peas |
| 1868 | He is elected abbot, and administration crowds out research |
| 1884 | He dies in Brno on January 6 |
| 1900 | Hugo de Vries, Carl Correns and Erich von Tschermak each report similar results and point back to Mendel |
| 1902-1903 | Walter Sutton and Theodor Boveri connect Mendel's "factors" to chromosomes |
| 1906 | William Bateson, Mendel's loudest champion in English, proposes the name "genetics" for the new field |
| 1908 | G. H. Hardy and Wilhelm Weinberg independently show how gene frequencies behave in a population |
| 1909 | Wilhelm Johannsen coins the word "gene" |
| 1910 | Thomas Hunt Morgan finds a white-eyed fruit fly, leading to sex-linked inheritance and gene linkage |
| 1918 | Ronald Fisher shows that many Mendelian genes acting together can explain smoothly varying traits such as height |
| 1936 | Fisher argues that Mendel's numbers fit his predictions too well |
| 1990 | Researchers identify the gene behind Mendel's wrinkled peas: a starch-branching enzyme gene disrupted by an inserted stretch of DNA |
| 2021 | Ahead of his 200th birthday, his remains are exhumed and his DNA analyzed |
The hawkweed episode explains part of the long silence. Hawkweed often sets seed without fertilization, so its offspring copy the mother plant, and Mendel's rules seemed not to apply. He had no way of knowing that. Some historians think it shook his own confidence in how general his pea results were.
Why peas were a smart choice
Garden peas suited the work for practical reasons. They normally pollinate themselves, so a variety grown for generations breeds true. They can also be crossed by hand: open a flower before it matures, remove the pollen-bearing parts, and dust it with pollen from another plant. A generation takes one season, and a monastery garden holds thousands of plants.
Most important, Mendel chose traits that came in two clear forms with nothing in between. He studied seven:
- Seed shape: round or wrinkled
- Seed color: yellow or green
- Flower color: purple or white
- Pod shape: inflated or constricted
- Pod color: green or yellow
- Flower position: along the stem or at the tip
- Stem length: tall or dwarf
The numbers that gave it away
Cross a true-breeding round-seeded plant with a true-breeding wrinkled one, and every seed in the first generation is round. Wrinkled seems to vanish. Let those plants self-pollinate, and wrinkled comes back.
Mendel counted. In the second generation he reported 5,474 round seeds and 1,850 wrinkled ones, a ratio of 2.96 to 1. For seed color he counted 6,022 yellow and 2,001 green, 3.01 to 1. The other traits gave similar figures, all close to 3 to 1.
When he followed two traits at once, crossing round yellow plants with wrinkled green ones, the second generation came out as 315 round yellow, 108 round green, 101 wrinkled yellow and 32 wrinkled green. That is very close to 9 to 3 to 3 to 1, exactly what you get if the two traits are inherited separately and each follows its own 3 to 1 rule.
The laws in modern words
Mendel described "factors" and did not call his conclusions laws; later textbooks did. In today's vocabulary:
The law of segregation
Each individual carries two copies of each gene, one from each parent. The copies (alleles) separate when eggs and pollen or sperm are made, so each reproductive cell carries just one. Fertilization pairs them up again at random.
If R is the dominant round allele and r the recessive wrinkled one, two Rr parents give this:
| R (from parent 2) | r (from parent 2) | |
|---|---|---|
| R (from parent 1) | RR, round | Rr, round |
| r (from parent 1) | Rr, round | rr, wrinkled |
Three round to one wrinkled, the ratio Mendel counted. This grid is called a Punnett square, after the geneticist Reginald Punnett, who popularized it in the early 1900s.
The law of independent assortment
Alleles of different genes sort into reproductive cells independently of each other. Whether a pea gets R or r tells you nothing about whether it gets Y or y. Multiply two 3 to 1 ratios and you get 9 to 3 to 3 to 1.
Dominance
Some textbooks list dominance as a third principle: in a pair of different alleles, one can mask the other. It is real, but it is a description of particular genes rather than a rule of inheritance, as the next section shows.
Where the ratios break: nonmendelian genetics
"Nonmendelian" does not mean Mendel was wrong. It means inheritance that does not give his simple ratios, usually because one of his tidy conditions does not hold.
| Pattern | What happens | Classic example |
|---|---|---|
| Incomplete dominance | Neither allele fully masks the other, so the mix is in between | Red and white snapdragons give pink offspring, then 1 red to 2 pink to 1 white |
| Codominance | Both alleles show fully | Blood type AB, with both A and B markers on red cells |
| Multiple alleles | A gene has more than two common versions | The ABO gene, with A, B and O alleles |
| Linkage | Genes close together on a chromosome travel together | Morgan's fruit fly crosses |
| Sex linkage | Genes on the X chromosome show different patterns in males and females | Red-green color blindness, more common in males |
| Polygenic traits | Many genes, plus environment, add small effects | Human height |
| Epistasis | One gene changes or hides the effect of another | Coat color in Labrador retrievers |
| Mitochondrial inheritance | Genes in mitochondria pass mainly through the mother | Some inherited optic nerve diseases |
| Genomic imprinting | A gene's effect depends on which parent it came from | Prader-Willi and Angelman syndromes |
| Incomplete penetrance | Not everyone with a genotype shows the trait | Many inherited disease risk variants |
Mendel was also lucky, or shrewd. Several of his seven genes sit on the same chromosomes, but the linked pairs are either far enough apart to behave almost independently or were not among the combinations he reported, so linkage did not spoil his two-trait ratios.
From one garden to whole populations
Population genetics asks Mendel's questions about a whole species rather than one family. Its starting point is the Hardy-Weinberg principle from 1908. If an allele has frequency q and its partner frequency p (so p + q = 1), then in a large, randomly mating population with no selection, migration or new mutation, genotype frequencies settle at p², 2pq and q², and stay there generation after generation.
A worked example with round numbers: suppose 1 child in 2,500 is born with a recessive condition. Then q² = 1/2,500, so q = 0.02 and p = 0.98. Carriers, who have one copy and are usually unaffected, make up 2pq = 2 × 0.98 × 0.02 = 0.039, or about 1 person in 25. A rare condition can come with a surprisingly common carrier state, which is the basic logic of carrier screening.
Real populations break the assumptions all the time, and measuring how far they stray is how geneticists detect selection, migration and inbreeding. In the 1920s and 1930s Fisher, J. B. S. Haldane and Sewall Wright built this into the mathematical core of evolutionary biology, joining Mendel's genetics with Darwin's natural selection.
Were his numbers too good?
In 1936 Fisher calculated that Mendel's results matched the expected ratios more closely than chance usually allows. The argument has never fully closed. Suggested explanations range from unconscious bias in sorting borderline seeds to Mendel reporting his clearest experiments and leaving others out. Few scholars now accuse him of fraud, and his conclusions have been confirmed in countless crosses since.
Taking Mendel into the classroom
For students, the fastest way into Mendel is to count something. Ears of genetics corn with purple and yellow kernels show a 3 to 1 or 9 to 3 to 3 to 1 pattern that a class can tally in a lesson, and the gap between their counts and the ideal ratio opens a natural discussion of chance and of Fisher's complaint. Human examples come later and need care; our piece on talking about stem cells in the classroom has suggestions for handling sensitive biology with teenagers.
For a longer read, Siddhartha Mukherjee's The Gene: An Intimate History opens its account of genetics in Mendel's garden and carries the story forward to modern genome science. Book suggestions on this site come from the books' contents and published reviews. Our list of science books about cells and genes has more, and our explainer on mRNA picks up where Mendel's "factors" lead: genes being read into proteins.
Mendel's 1 in 4 still turns up in hospitals. Each full sibling of a patient needing a blood stem cell transplant has about a 1 in 4 chance of being a tissue-type match, because each parent passes on one of two sets of HLA genes, a direct echo of the pea counts. Our piece on finding a transplant match explains what happens when no sibling matches.
This explainer is general information, not medical advice. Decisions about treatment, donation or trials belong with a clinician who knows your history.



