Use a Punnett square to predict genotype and phenotype ratios from a monohybrid cross, and state Mendel's law of segregation.
Gregor Mendel, an Austrian monk working in the 1860s, discovered the basic rules of inheritance using pea plants -- a smart choice, since pea plants have several easily observed either-or traits (such as purple vs. white flowers), can self-pollinate or be cross-pollinated by hand, and can be bred into true-breeding lines that reliably pass down one version of a trait generation after generation. From thousands of carefully tracked crosses, Mendel proposed that each trait is controlled by two hereditary factors, now called alleles, one inherited from each parent. An organism's specific pair of alleles is its genotype, written with letters (such as BB, Bb, or bb), while the physical trait that genotype actually produces is its phenotype. An organism with two identical alleles (BB or bb) is homozygous; one with two different alleles (Bb) is heterozygous. When the alleles differ, one is often dominant, meaning it alone determines the phenotype (traditionally written as a capital letter), while the other is recessive, meaning it only shows up in the phenotype when no dominant allele is present (written lowercase) -- Mendel's law of segregation states that the two alleles for a gene separate from each other during gamete formation, so each gamete carries only one allele for that gene.
A Punnett square is simply a tool for predicting the possible offspring of a cross by laying out all the combinations of alleles each parent's gametes could contribute. In a monohybrid cross (tracking a single gene), crossing two heterozygous parents (Bb x Bb) produces a 1:2:1 genotype ratio among the offspring (1 BB : 2 Bb : 1 bb) -- and because both BB and Bb produce the dominant phenotype, that works out to a 3:1 phenotype ratio (3 showing the dominant trait for every 1 showing the recessive trait). It is worth being precise about what this ratio actually predicts: it describes the probability for any single offspring, and like a coin flip, small numbers of actual offspring can easily deviate from the exact predicted ratio by chance, even though the probability itself never changes.
Mastery looks like: They can build a Punnett square from scratch for a monohybrid cross and correctly state both the genotype and phenotype ratios it predicts.
Common stumbles: Assuming a predicted ratio like 3:1 must show up exactly in a small number of real offspring; confusing genotype (the allele pair) with phenotype (the resulting physical trait).