Mendelian Genetics: Build Your Own Punnett Square

How to use this simulation

1. Start with the "Monohybrid cross" option above — Mendel's tall/dwarf pea plants — the simplest cross to learn on.
2. Pick a genotype for each parent (say, Tt and Tt); the Punnett square below fills in all four boxes instantly and prints the genotype and phenotype ratios.
3. Switch "Cross type" to try a dihybrid cross, a test cross, incomplete dominance, or human blood groups — each has its own genotype buttons.
4. Turn on "Simulate many offspring" with a small number first, then a large one, and watch the observed ratio drift toward the exact one the grid predicts.
5. Read the bar chart and the readings panel together — that is where the genotype ratio, phenotype ratio, and each phenotype's percentage are printed.

A monastery garden that rewrote biology

Look around your own family. Do you have your mother's eyes or your father's? Is height something that just happens, or does it follow a pattern? Between 1856 and 1863, an Augustinian monk named Gregor Mendel grew tens of thousands of pea plants in a monastery garden in what is now the Czech Republic, asking exactly that question with hard numbers instead of guesses. What came out of his careful counting is the foundation every biology course still teaches — and he did it without ever knowing what a gene, a chromosome, or DNA actually was.

This page puts a working Punnett square in your hands. Pick two parents' genotypes and watch their gametes drift into the grid; the offspring genotypes and phenotype ratios update the instant you change anything. Five crosses are built in — monohybrid, dihybrid, test cross, incomplete dominance, and human ABO blood groups — plus a "simulate offspring" mode where you can watch a small, noisy sample settle into the exact ratio a big one predicts.

Mendel and his pea plants

Mendel picked the garden pea (Pisum sativum) for good reasons. Left alone, a pea flower self-pollinates, so a line bred for generations stays pure for a trait — but a breeder can also pry the flower open and cross-pollinate two chosen plants by hand. One plant produces plenty of seeds, several generations fit inside a single growing season, and a small garden holds enough plants for real statistics.

He chose seven traits that came in exactly two clear-cut forms with nothing in between: plant height (tall or dwarf), seed shape (round or wrinkled), seed colour (yellow or green), flower colour, pod shape, pod colour, and flower position. No trait had a "medium" version — a plant was tall or it was dwarf, never something in between. That sharp either/or is what made his counting so clean.

He started by crossing a pure-bred tall plant with a pure-bred dwarf one (this starting generation is called P, for parental). Every single plant in the next generation (F1) was tall — the dwarf trait seemed to vanish. He then let those F1 plants self-pollinate, and in the following generation (F2) dwarf plants reappeared, with tall to dwarf landing close to a 3 : 1 ratio. That reappearance was the clue: the instruction for "dwarf" had not been destroyed in F1, only hidden.

Words that keep coming back in this chapter

These terms show up in almost every question on this topic — worth getting comfortable with them before anything else.

The law of segregation

Every organism carries two alleles for each trait — one inherited from each parent. When gametes (sperm or egg cells) form during meiosis, those two alleles separate, and each gamete receives only one of them. That separating-apart is the law of segregation.

A plant with genotype Tt, for instance, makes gametes that are half T and half t — never a gamete carrying both. Pick "Monohybrid cross" in the simulation and watch it happen: two distinct gamete letters drift out of a Tt parent's box into the grid's row or column, split exactly in half.

The law of independent assortment

When an organism carries genes for two or more different traits (and those genes sit on different chromosomes, or far enough apart on the same one), which allele of one trait ends up in a gamete has no bearing on which allele of the other trait ends up there — the two assort independently. This law is exactly what produces the famous 9:3:3:1 ratio in a dihybrid cross.

A plant with genotype RrYy, where seed shape (R/r) and seed colour (Y/y) sit on different chromosomes, makes four kinds of gametes — RY, Ry, rY, and ry — each at an equal 25%, with no pairing bias between shape and colour. Mendel had no concept of a chromosome, yet his counting had already found this independence on its own.

Monohybrid cross, worked out: tall × tall (Tt × Tt)

Mendel's F1 plants (all Tt, all tall in appearance) self-pollinating is exactly the simulation's default monohybrid cross: Tt × Tt. Line one parent's gametes (T, t) along the rows and the other's (T, t) along the columns, and the four-cell Punnett square below falls out.

• Genotypes: TT ×1, Tt ×2, tt ×1 — a 1 : 2 : 1 ratio
• Phenotypes: tall ×3 (75%), dwarf ×1 (25%) — a 3 : 1 ratio

Dihybrid cross, worked out: round-yellow × round-yellow (RrYy × RrYy)

Mendel's bigger experiment tracked two traits at once: seed shape (round R is dominant, wrinkled r is recessive) and seed colour (yellow Y is dominant, green y is recessive). Cross two plants that are heterozygous for both (RrYy), and each parent makes four kinds of gametes (RY, Ry, rY, ry), so the Punnett square grows to 4 × 4 = 16 cells.

• Round, yellow: ×9 (56.25%)
• Round, green: ×3 (18.75%)
• Wrinkled, yellow: ×3 (18.75%)
• Wrinkled, green: ×1 (6.25%)
• Ratio: 9 : 3 : 3 : 1 — the famous 9:3:3:1 is itself the best evidence for independent assortment

Test cross: how to catch a hidden allele

A tall pea plant could be TT or Tt — you cannot tell by looking, because both phenotypes are tall. Mendel's fix was the test cross: cross the unknown plant with a known homozygous recessive (tt) partner, because tt itself always contributes the same t gamete, so the offspring phenotypes reveal only what the unknown parent's gametes were.

If the unknown plant is homozygous dominant (TT), every offspring turns out Tt (tall) — all 4 of them, none dwarf (100%). But if the unknown plant is heterozygous (Tt), the offspring split evenly — 2 tall and 2 dwarf (1:1). So the moment even one dwarf offspring shows up, you know for certain the unknown parent was heterozygous.

Incomplete dominance: when neither allele wins outright

In Mendel's peas, the dominant allele fully masks the recessive one, but that is not universal. In the four o'clock flower (Mirabilis jalapa) or in snapdragons, the two alleles for flower colour blend into an intermediate shade instead — that's incomplete dominance. Cross a red (RR) allele with a white (rr) one, and the heterozygote (Rr) is pink: neither red nor white, a genuine blend.

• Red (RR): ×1 (25%)
• Pink (Rr): ×2 (50%)
• White (rr): ×1 (25%)
• Ratio: 1 : 2 : 1

Codominance and human blood groups

In codominance, both alleles express themselves fully and simultaneously — they don't blend the way incomplete dominance does. The human ABO blood group is the textbook example: three alleles — Iᴬ, Iᴮ, and i — where Iᴬ and Iᴮ are codominant and i is recessive.

A father with genotype Iᴮi (blood type B) and a mother with genotype Iᴬi (blood type A) can have a child of four different genotypes — resolving to AB, A, B, or O blood type, each at an equal 25%. That single couple can produce all four blood types, even though neither parent is type AB or type O.

• AB: ×1 (25%)
• A: ×1 (25%)
• B: ×1 (25%)
• O: ×1 (25%)
• Ratio: 1 : 1 : 1 : 1

Sex determination: XX, XY, and why the coin toss is fair

Two chromosomes decide human biological sex: X and Y. A mother carries two X chromosomes (XX), so every one of her eggs is the same — it carries an X. A father carries one X and one Y (XY), so exactly half his sperm carry an X and the other half carry a Y.

An egg (always X) fertilised by a sperm (X or Y) can therefore only make one of two combinations: X + X = XX (a daughter) or X + Y = XY (a son), each at exactly 50% — as fair as a coin toss. That single fact settles a very old argument: a baby's sex is decided by which chromosome the father's sperm happened to carry, and the mother's egg plays no part in deciding it at all.

Sex-linked traits, briefly: colour blindness and haemophilia

Some genes sit directly on the X chromosome itself. Red-green colour blindness and haemophilia (a blood-clotting disorder) are two well-known recessive sex-linked traits. A male has only one X chromosome, so a single recessive copy is enough to show the trait. A female has two X chromosomes, so she needs both copies to be recessive before the trait shows — which is why both conditions are far more common in males than in females.

If a mother is a carrier — meaning one of her two X chromosomes carries the recessive allele while she herself is unaffected — each of her sons has a fifty-percent chance of being affected, since a son's single X comes from whichever of her two X chromosomes he happens to inherit.

Why Mendel succeeded where others had failed

Plant breeders before Mendel had noticed inheritance patterns, but nobody had pulled out a clean, quantitative rule from them. The reasons show up in every mode of this simulation: he tracked only one or two sharply either/or traits at a time; he confirmed pure-breeding lines across several generations first; and — the part that mattered most — he counted hundreds or thousands of plants per cross and treated the results as statistics, which was an unusual thing to do at the time.

Over eight years Mendel grew roughly 28,000 pea plants. That sheer scale is what let him see past the noise of a small, unlucky batch and land on the true 3:1 and 9:3:3:1 ratios — exactly the same reason a small run of the simulation's "simulate" mode can look lopsided while a large one settles onto the ratio the grid predicts.

Try these experiments in the simulation

Press "Reset" before each experiment so the simulated offspring count starts back at zero.

Solved problems

Every number below comes straight out of the simulation's own model — nothing here is typed in by hand.

Common mistakes

Worth a quick check before writing anything down.

• Treating genotype and phenotype as the same thing — genotype is the hidden allele combination, phenotype is its visible result; Tt and TT differ in genotype but share the same phenotype (tall).
• Writing 1:1:1:1 for every dihybrid cross — the 9:3:3:1 ratio only appears when both parents are heterozygous for both traits; make one parent homozygous recessive and the ratio changes completely (see Problem 4).
• Mixing up incomplete dominance with codominance — incomplete dominance blends two alleles into a new intermediate look (pink); codominance expresses both alleles side by side without blending (an AB blood type shows both the A and B antigens, unmixed).
• Running a test cross against another heterozygote instead of a known homozygous recessive — a test cross only gives a clean answer when the known parent is tt; anything else muddies the result.
• Assuming the mother decides a baby's sex — biologically, it is the father's sperm (carrying X or Y) that decides it; the mother's egg always carries an X.
• Being surprised when a small sample doesn't land on the exact textbook ratio (like 3:1) — small samples wobble; that wobble shrinking as the sample grows is the law of large numbers at work, not a mistake in the maths.

Mendelian genetics in real life

Mendel's rules are still one of the most practically useful parts of biology.

• Genetic counselling: knowing both parents' genotypes lets a counsellor estimate the chance a recessive condition, such as sickle-cell disease, appears in a child before the child is born.
• Agriculture and animal breeding: plant and livestock breeders use test crosses and controlled pollination to build higher-yield or disease-resistant varieties.
• Blood transfusion and surgery: understanding how ABO and Rh blood types are inherited makes it far easier to locate a safe donor match.
• Forensic science: paternity and identity testing rely on the same inheritance logic, now read off DNA markers instead of pea-plant traits.
• Understanding inherited-disease risk: pedigree charts for sex-linked conditions like haemophilia or colour blindness are read using exactly Mendel's rules to predict which generation a trait may resurface in.

Exam corner

Genetics problems reward working the Punnett square methodically rather than memorising a final ratio — examiners routinely change one parent's genotype specifically to break a memorised answer.

Watch for: a question that gives you an F1 and F2 phenotype ratio and asks you to work backward to the parents' genotypes; a dihybrid problem that quietly uses a test cross instead of two heterozygotes; and any blood-type question that expects you to realise a child's type can differ from both parents'.

• Always define your allele letters explicitly before starting a cross — a grader cannot follow T and t if you never stated which one is dominant.
• Draw the full Punnett square, even for a "simple" monohybrid cross — a rushed mental shortcut is where most marks are lost.
• State both the genotype ratio and the phenotype ratio separately; they are frequently different numbers for the same cross, and graders check both.
• For a test cross, explicitly state which observed ratio (uniform vs 1:1) corresponds to which unknown genotype — that reasoning step is usually worth its own marks.

Revision: everything on one screen

A last read-through before the exam.

• Law of segregation: the two alleles for a trait separate during gamete formation, so each gamete carries only one.
• Law of independent assortment: genes on different chromosomes assort into gametes independently of one another.
• Monohybrid cross (Tt × Tt): genotype 1 : 2 : 1, phenotype 3 : 1.
• Dihybrid cross (RrYy × RrYy): phenotype 9 : 3 : 3 : 1.
• Test cross: an unknown dominant parent crossed with tt — a uniform result means homozygous, a 1:1 split means heterozygous.
• Incomplete dominance (Rr × Rr): genotype and phenotype ratios match, 1 : 2 : 1.
• Codominance (blood type, Iᴬi × Iᴮi): AB, A, B, O each at 25%.
• Sex determination: XX and XY are each 50% likely, decided by the father's sperm, never the mother's egg.
• The bigger the sample, the closer the observed ratio sits to the predicted one — the law of large numbers.

Frequently asked questions