How to Make a Punnett Square (Monohybrid and Dihybrid Crosses)

A Punnett square is one of the first tools every genetics student meets, and also one of the easiest to fill in wrong. The grid itself is simple: parental gametes go on the edges, offspring combinations fill the inside. What trips people up is everything around it, picking the right gametes, keeping a monohybrid 2x2 from quietly turning into a dihybrid 4x4, and then reading the genotype ratio when the question actually asked for the phenotype ratio.

This guide walks through how a Punnett square works for both monohybrid and dihybrid crosses, covers the special cases like sex linkage and codominance, and shows how to lay out a clean, labeled grid with the SciDraw AI Punnett Square Maker.

A monohybrid Punnett square puts one parent's gametes across the top and the other's down the side, then fills each cell with the combined offspring genotype.

Quick Answer

A Punnett square predicts the possible genotypes of offspring from a cross. You write each parent's gametes along the two axes, then combine the row allele and column allele inside each cell. A single-gene (monohybrid) cross uses a 2x2 grid; a two-gene (dihybrid) cross uses a 4x4 grid. From the filled cells you count two different things: the genotype ratio (the exact allele combinations) and the phenotype ratio (the visible traits, after dominance is applied).

	Monohybrid cross	Dihybrid cross
Genes tracked	one	two
Gametes per parent	2 (e.g. A, a)	4 (e.g. AB, Ab, aB, ab)
Grid size	2x2	4x4
Classic parents	Aa x Aa	AaBb x AaBb
Genotype ratio	1 : 2 : 1	1:2:1:2:4:2:1:2:1
Phenotype ratio	3 : 1	9 : 3 : 3 : 1

The famous 3:1 and 9:3:3:1 ratios only appear for those specific heterozygous crosses. Change the parents and the ratios change, so always derive the gametes from the actual genotypes rather than reaching for a memorized number.

How a Punnett Square Is Built

Every Punnett square is built from the same three pieces, regardless of size.

First come the gametes. Each parent contributes one allele per gene to each gamete, so you split the genotype into its possible halves. For a single gene, Aa produces two gamete types, A and a. For two genes, you take one allele from each gene and list every combination, so AaBb produces four gametes: AB, Ab, aB, and ab. Getting this list right is the whole game, because a wrong gamete list guarantees a wrong grid.

Next come the axes. One parent's gametes label the columns across the top, the other parent's gametes label the rows down the side. A monohybrid cross needs two columns and two rows; a dihybrid cross needs four of each, which is exactly why the grid jumps from 2x2 to 4x4.

Finally come the cells. Each inner cell combines the allele from its column with the allele from its row, giving one possible offspring genotype. Convention is to write the dominant allele first and keep paired alleles together, so a cell reads Aa rather than aA, and AaBb rather than aBAb. Once every cell is filled, the grid contains every equally likely offspring genotype, and the ratios fall straight out of counting them.

Common Mistakes

Mistake 1: Writing the wrong gametes for a dihybrid cross

The most common dihybrid error is listing only two gametes instead of four. AaBb does not give AB and ab; it gives all four combinations, AB, Ab, aB, and ab, because the two genes assort independently. Miss two gametes and the 4x4 grid collapses, and the 9:3:3:1 ratio disappears with it.

Mistake 2: Reporting the genotype ratio when the question wants the phenotype ratio

These are not the same answer. A monohybrid Aa x Aa cross gives a genotype ratio of 1 AA : 2 Aa : 1 aa, but a phenotype ratio of 3 dominant : 1 recessive, because AA and Aa look identical. Read the question carefully and apply dominance only when it asks for phenotypes.

Mistake 3: Treating codominance and blood types like simple dominance

In codominance and multiple-allele systems, heterozygotes show their own phenotype. ABO blood type uses three alleles, where I^A and I^B are codominant and i is recessive. An I^A i x I^B i cross produces four phenotypes, A, B, AB, and O, not a clean 3:1, so you cannot collapse the heterozygote into a dominant class.

Mistake 4: Drawing sex-linked genes on the wrong chromosome

For X-linked traits, the allele rides on the X chromosome, so the male parent contributes either his single X allele or a blank Y. Writing genotypes as X^B X^b and X^B Y keeps this visible. Forgetting the Y, or treating the trait as autosomal, is what produces the wrong ratios for color blindness and similar crosses.

How to Make a Punnett Square with SciDraw AI

SciDraw AI is a drawing tool, not a probability calculator. You work out the genetics, the gametes, the cross, the alleles, and SciDraw AI draws and labels a clean grid from your description. It lays out the axes, fills the offspring genotypes you specify, and produces a figure you can drop into a worksheet, a slide, or a lab report.

Open https://sci-draw.com/punnett-square-maker and describe the cross in plain language. You will get the best results when you include:

• the cross type (monohybrid or dihybrid) and grid size,
• each parent's genotype and the gametes on each axis,
• the offspring genotype for every cell,
• what the alleles mean (the trait and dominance),
• any genotype or phenotype ratio you want labeled.

For a monohybrid cross, a prompt that works well:

Create a 2x2 monohybrid Punnett square for Aa x Aa. Put gametes A and a across the top and A and a down the side. Fill the four cells AA, Aa, Aa, aa. Label it as a heterozygous cross and note the genotype ratio 1:2:1 and phenotype ratio 3:1.

For a dihybrid cross:

Create a 4x4 dihybrid Punnett square for AaBb x AaBb. Use gametes AB, Ab, aB, ab on both axes and fill all 16 offspring genotypes. Label A as the dominant allele for seed shape and B for seed color, and note the 9:3:3:1 phenotype ratio.

And for a sex-linked cross:

Create a 2x2 Punnett square for an X-linked trait: carrier mother X^B X^b crossed with unaffected father X^B Y. Put the mother's gametes on one axis and the father's X and Y on the other, fill the four offspring genotypes, and label which offspring are carriers or affected.

A vague prompt like "make a Punnett square for eye color" forces the tool to guess the alleles, the parents, and the grid size. Spelling out the genotypes and every cell gets you a figure that is correct as well as clean.

For a classroom worksheet, the labeled monohybrid grid is usually enough. For a report or a slide, ask for clear allele labels, the parental genotypes shown on the axes, and the genotype and phenotype ratios annotated beside the grid.

Work out the genetics once, then get a clean, labeled grid with the SciDraw AI Punnett Square Maker.