Part 3 — GED ® Science · Unit 3 — Inheritance and Change Across Generations
Chapter 11 — Inheritance Without Fortune-Telling
GED® test targets: L.e.2, SP.8.c
An attendance sheet tells you who showed up today; it does not tell you who will come tomorrow. A Punnett square is also limited: it organizes possible allele combinations under a stated model. It is not a damn prophecy about a child’s appearance, intelligence, character, or future. We use the model precisely because we know what it includes and what it leaves out.
A Punnett square is a probability model. It organizes possible allele combinations from two parents under stated assumptions. It is not a four-box crystal ball. If the model shows a one-in-four chance, nature does not owe a family exactly one outcome in every four offspring.
Genotype and phenotype
An allele is a version of a gene. A genotype is the allele combination being considered. A phenotype is an observable characteristic, influenced by genotype and often by environment. In a simple model, A might represent a dominant allele and a a recessive allele. AA and Aa express the dominant phenotype, while aa expresses the recessive phenotype.
Dominant means an allele’s effect appears in the heterozygous combination under the model. It does not mean stronger, better, more common, or morally superior. Recessive does not mean weak or defective. Homozygous means two matching alleles; heterozygous means two different alleles at the gene being considered.
Many real traits do not follow simple complete dominance. In incomplete dominance, the heterozygote has an intermediate phenotype. In codominance, both allele effects are expressed. Many traits are polygenic, influenced by multiple genes. Use the inheritance pattern supplied in the question rather than forcing every trait into one convenient square.
Build the model one parent at a time
For an Aa × Aa cross, each parent can contribute A or a. Put one parent’s possibilities across the top and the other’s down the side. Combine one allele from each parent in each box.
| Parent 1: A | Parent 1: a | |
|---|---|---|
| Parent 2: A | AA | Aa |
| Parent 2: a | Aa | aa |
The four equally likely combinations are AA, Aa, Aa, and aa. Genotype probabilities are 1/4 AA, 1/2 Aa, and 1/4 aa. With complete dominance, the dominant phenotype probability is 3/4 and the recessive phenotype probability is 1/4. The two Aa boxes count separately because they arise from different parental contributions, even though the resulting genotype is the same.
For Aa × aa, the first parent contributes A or a, while the second contributes only a. Half the expected combinations are Aa and half aa. If a question changes one parent’s genotype, rebuild the possibilities. Do not drag the famous 3:1 ratio into every problem like it owns the place.
Probability across multiple offspring
Under the model’s independence assumption, the probability that two offspring from Aa × Aa both have aa is 1/4 × 1/4 = 1/16. The probability that neither has aa is 3/4 × 3/4 = 9/16. Therefore the probability that at least one has aa is 1 − 9/16 = 7/16.
“At least one” includes one or two. “Exactly one” excludes two. Exactly one aa outcome has two orders: aa then not-aa, or not-aa then aa. Add their probabilities: (1/4 × 3/4) + (3/4 × 1/4) = 6/16 = 3/8. Wording changes the event you are counting.
For 80 offspring under the simple Aa × Aa model, the expected number with aa is 80 × 1/4 = 20. Actual counts may differ through chance. An expected number is the long-run model average, not a guaranteed head count. A modest mismatch does not automatically disprove the inheritance model.
Read a pedigree cautiously
A pedigree is a family diagram showing a trait across generations. Its legend explains symbols, shading, and relationships. Shading usually marks the studied phenotype, not whether a person is healthy in every respect. Symbols may conventionally indicate sex, but use the supplied key rather than guessing.
In a simplified autosomal recessive model, two parents without the recessive phenotype can have an offspring with it if both are carriers, Aa. A carrier has the recessive allele without expressing that recessive phenotype in this model. If an affected aa offspring is shown, each biological parent must have supplied an a allele under the assumptions.
A pedigree alone may not distinguish every possible inheritance pattern. Small families, incomplete information, and environmental influences can leave multiple explanations. If the question specifies autosomal recessive inheritance, use that information. Do not pretend the diagram established it independently.

A simplified pedigree with a stated recessive model. Both unshaded parents are carriers; the unshaded offspring’s exact genotype is not determined by phenotype alone.
In this supplied recessive model, the shaded offspring has genotype aa. Each unshaded parent supplied an a allele, so each must be Aa under the stated assumptions. The unshaded sibling could be AA or Aa; appearance alone does not distinguish those possibilities. An unshaded symbol does not automatically mean no recessive allele is present.
Worked inheritance question
Question: In the stated complete-dominance model, a parent with genotype AA and a parent with genotype aa produce offspring. Which prediction is supported? A. All show the recessive phenotype. B. Half are AA and half aa. C. All offspring are expected to be Aa. D. Each offspring receives both alleles from one parent.
Answer: C. The AA parent contributes A in every gamete at this gene, and the aa parent contributes a. The combination is Aa. B incorrectly keeps parental pairs intact. A ignores complete dominance. D contradicts the one-allele-from-each-parent model.
The most reliable method is to identify the inheritance rule, list parental contributions, combine them, count the requested outcome, and translate the count into a probability. That sequence is reasoning, not fortune-telling.
Inheritance Is Probability, Not a Family Sentence
Begin with the parents’ genotypes, not their appearances alone. In a complete-dominance model, AA and Aa can show the same dominant phenotype even though their possible gametes differ. An AA parent contributes A at that gene; an Aa parent can contribute A or a. Each offspring receives one allele from each parent. The square combines contributions, not entire parental genotypes.
Consider Aa crossed with aa as a worked demonstration. The first parent’s possibilities are A and a; the second contributes a. Pairing them yields Aa and aa with equal model probabilities. Expected dominant phenotype = 1/2 = 50%, and expected recessive phenotype = 1/2 = 50%. Among 80 offspring, the expected recessive count is 80 × 0.5 = 40. Expected does not mean guaranteed; actual outcomes vary.
If successive offspring are independent under the model, an earlier result does not use up an allele combination. Two recessive outcomes in a row do not force a dominant result next. That “it is due now” argument is the same bullshit that confuses chance in other settings. The probability for each offspring remains whatever the stated genetic model gives.
Dominant does not mean common, healthier, stronger, or morally better. It describes how alleles relate to phenotype in a heterozygote. In incomplete dominance, the heterozygous phenotype can be intermediate; in codominance, both allele effects are expressed. Many traits depend on multiple genes and environmental conditions. Do not stretch a one-gene pea-plant model into a theory of everything human.
A pedigree shows observed family relationships and traits, while a Punnett square models possible combinations. When a GED-style passage supplies an inheritance rule for an unfamiliar organism, use that rule rather than guessing from a real trait you remember. State the genotype possibilities first, translate them into phenotype using the rule, and only then calculate the requested probability. That order keeps the fancy terminology attached to actual reasoning.
Your Turn — Guided and Independent Practice — Practice Check S11
Guided: In a simple model, one parent is heterozygous (Aa) and the other is homozygous recessive (aa). Make the four equally likely genotype combinations for Aa × aa and find the probability of aa.
Guided answer: The combinations are Aa, Aa, aa, and aa. Two of four are aa, so the probability is 2/4 = 1/2, or 50%. This predicts repeated possibilities, not the guaranteed outcome of four actual children.
Independent: A trait appears more often in one family but is also affected by diet. Which statement is best? A. A pedigree proves causation by itself. B. The trait must be controlled by one dominant allele. C. Environment never matters. D. Genes and environment can both influence phenotype.
Independent answer: D. Phenotype can reflect genotype, environment, and their interaction. B assumes a simple pattern not supplied, C ignores the prompt, and A overstates what a family pattern can establish. Common mistake: turning probability into destiny.