AP Biology Unit 5 Practice Test: Heredity MCQs (Free PDF)

Here's a free AP Biology Unit 5 practice test: 20 AP-style MCQs on heredity, seven data-based stimulus sets across topics 5.1 through 5.5, and a full rationale for every answer, all aligned to the 2025 CED.

Below, three mistakes I see students make on Unit 5 and how to avoid them, plus one worked example so you can see the reasoning.

Unit 5 is roughly 8-11% of the exam (the full 8-unit breakdown lists weightings for the rest) and it's one of the units where a small handful of predictable stimulus types (pedigrees, dihybrid ratios, chi-square, blood groups) do most of the heavy lifting.

AP Biology is a stimulus-heavy exam. The Section I MCQs almost always sit under a short passage, a data table, a diagram or a graph, and you're being marked on how well you interpret the stimulus rather than how many facts you can recall cold (there's a full MCQ strategy walkthrough if you want the general playbook).

Unit 5 leans especially hard on this. Expect:

• Multi-generation pedigrees
• Punnett squares described in prose
• Chi-square-style tables where you decide whether observed counts fit a Mendelian ratio
• Non-Mendelian passages built around pea colours, snapdragon flowers or ABO blood types
• Linked-gene test-cross counts where the parental classes dominate the offspring

The practice pack mirrors that mix. 13 of the 20 questions sit on a data-table stimulus, including pedigree stimuli where you have to name the inheritance pattern and then reason about a specific individual's genotype, dihybrid test crosses that check whether you can spot independent assortment from four near-equal counts, chi-square data compared against the critical value at df = 1, and a temperature-sensitive Himalayan rabbit passage that tests genotype-environment interaction.

If you're prepared for the stimulus formats, the actual biology becomes much easier to see.

Every year I see students walk into Unit 5 MCQs making the same three moves. Recent Chief Reader Reports call them out too, so spot them before the exam does.

This one shows up in the 2024 Chief Reader Report on Q1. Students write that crossing over "changes or replaces genes" (it doesn't; it exchanges segments of chromatid, reshuffling existing alleles onto new chromosomes), or that crossing over IS a mutation (it isn't; mutation creates new alleles, crossing over rearranges old ones).

Then they place the event in metaphase I. Crossing over happens in prophase I, when homologous chromosomes synapse and form the synaptonemal complex. Metaphase I is only where the paired homologs line up at the equator with random orientation, and that random orientation is independent assortment, a separate source of variation.

The mental model that keeps this straight:

• Prophase I: pairing and exchange
• Metaphase I: orientation
• Anaphase I: homologs separating
• Anaphase II: sister chromatids separating

The 2024 Chief Reader Report Q1 flagged the crossing-over-as-mutation misconception directly (alongside 'crossing over changes or replaces genes' and 'independent assortment occurs between genes not chromosomes'), and the metaphase-I timing error is a related common student mistake.

Homologous chromosomes and sister chromatids look similar in textbook diagrams (two parallel structures), and students conflate them constantly. Homologs are two different chromosomes carrying the same genes at the same loci, one from each parent, and they can carry different alleles. Sister chromatids are two identical copies of one chromosome, produced by S-phase replication and joined at the centromere.

This distinction matters because homologs separate in meiosis I and sister chromatids separate in meiosis II. A quick way to remember it: pairs first, copies second. Mixing them up leads to wrong chromosome-number counts on any diagram-labelling MCQ.

Students see "affected" in a pedigree and default to recessive because it "makes sense". Many disease alleles are dominant, though: Huntington's disease and achondroplasia are the classic examples. Students also assume the phenotype seen in most of the population must be dominant, which conflates frequency with dominance. Frequency has nothing to do with dominance. Dominance is only about how alleles interact within a genotype: a dominant allele can be rare (Huntington's), and a recessive allele can be common (cystic fibrosis carrier frequency).

On pedigree MCQs, apply the logic instead of the intuition. Two unaffected parents with an affected child means the trait is recessive (rules out dominant). Two affected parents with an unaffected child means the trait is dominant (rules out recessive), because two recessive homozygotes couldn't produce an unaffected offspring.

Here's a question straight from the Unit 5 pack. The stimulus:

A pedigree tracks a rare disorder through three generations. Filled symbols are affected, open symbols are unaffected; squares are male, circles are female.

Generation I: Man I-1 (unaffected, open square) × Woman I-2 (unaffected, open circle). Generation II: They have three children, II-1 (affected male, filled square), II-2 (unaffected female, open circle), II-3 (unaffected female, open circle). II-2 marries II-4 (unaffected male, open square, unrelated). II-3 marries II-5 (unaffected male, open square, unrelated). Generation III: II-2 × II-4 have four children, III-1 (affected male), III-2 (unaffected female), III-3 (unaffected male), III-4 (affected male). II-3 × II-5 have three children, III-5 (unaffected female), III-6 (affected male), III-7 (unaffected female).

Summary counts across the whole pedigree:

The stem: 'Which inheritance pattern is most consistent with the pedigree data summarized in Table 6?'

• (A) Autosomal dominant, appearing in every generation and in males and females equally often.
• (B) Autosomal recessive, requiring both parents to be carriers before affected children appear.
• (C) Y-linked, transmitted only from father to son.
• (D) X-linked recessive, appearing mostly in males and passing through unaffected carrier mothers.

Answer: D. The pedigree shows affected males in generations II and III but no affected females, and the trait passes through unaffected mothers (II-2 and II-3, both daughters of the unaffected Gen I couple). Sons of a carrier mother have a 1/2 chance of being affected; III-1, III-4 (from II-2) and III-6 (from II-3) fit that pattern. That heavy male bias with transmission through unaffected females is the signature of X-linked recessive inheritance. Autosomal dominant (A) predicts affected in every generation with roughly equal sexes, which is not what we see. Autosomal recessive (B) would let affected females appear when both parents are carriers. Y-linked (C) requires an affected father passing the Y directly to sons, which the pedigree does not show.

Pedigree analysis follows a hierarchy.

1. If two unaffected parents produce an affected child, the trait is recessive.
2. If the trait skips generations and shows up almost only in males, it suggests X-linked recessive (males are hemizygous and can't hide a recessive allele; affected females need two copies).
3. If the trait appears in every generation and affected fathers pass it to all daughters but zero sons, it's X-linked dominant.

Running this three-step check in the first 20 seconds usually narrows a four-option MCQ to one answer before you touch any probabilities.

If the MCQ gives you a dihybrid cross (AaBb × AaBb) and offers ratios in the answer choices, the answer is 9:3:3:1 phenotypes unless something violates independent assortment.

• Linked genes on the same chromosome give ratios much closer to 3:1 because the parental phenotypes dominate.
• Epistasis produces modified ratios like 9:3:4 or 12:3:1.
• Incomplete dominance and codominance don't change the 1:2:1 genotypic ratio but change the phenotype count.

Recognising which "flavour" of ratio the question is testing lets you skip past three distractor answers straight away.

Unit 5 MCQs sometimes drop in a chi-square calculation. You don't need to compute the statistic from scratch every time; the question usually gives you the value and asks whether the null hypothesis holds.

The reflex: with 1 degree of freedom, the critical value at p = 0.05 is 3.84. If your test statistic is below 3.84, fail to reject the null (the data fit the expected Mendelian ratio). If it's above 3.84, reject the null (something else is going on: linkage, epistasis, selection).

Memorising 3.84 for 1 df and 5.99 for 2 df is faster than looking it up on the AP formula sheet.

A 3-pass method works well for a unit-length pack like this one.

• First pass, untimed: give yourself about 90 seconds per question, work through all 20, and use the rationales to understand every question you got wrong or guessed on. This is a learning pass, not a scoring pass.
• Second pass, timed: 30 minutes for 20 questions matches the AP pace of 1.5 minutes per MCQ. Circle any answer you're not confident on and move on. This is where you build exam pacing.
• Third pass, a week later: retry the questions you missed on the first two passes without looking at the answers, and see whether the reasoning has stuck.