Progeny Testing and Combining Ability
Estimate parental breeding value from offspring performance across mates, families, environments, and target traits.
Estimate parental breeding value from offspring performance across mates, families, environments, and target traits.
Core science
A progeny test evaluates a parent by measuring its descendants. This is essential when the parent’s own phenotype is misleading, when traits are expressed only in one sex or late stage, or when selection aims at transmission rather than clone performance. Families can be half-sibs, full-sibs, selfs, testcrosses, or structured diallel and factorial sets.
General combining ability describes average performance across several mates and is often associated with additive effects. Specific combining ability describes an unusually favorable or unfavorable combination relative to parental averages and can include dominance and epistasis. Estimates depend on the mating design, tester choice, environments, family size, and statistical model. A parent tested with one mate does not have a reliable general combining-ability estimate.
Use common testers and checks, balanced family sizes where possible, and replicated trials. Preserve both mean and within-family variation. A cross with one exceptional offspring may be attractive for clone selection but poor for seed-family uniformity. Conversely, a parent producing consistently acceptable families can be valuable even if it never produces the single most extreme individual.
Why this matters in cultivation
- Decide whether the program wants elite individual clones, predictable seed families, or parental lines. Design progeny tests to match that objective.
Measure and record
Record 1
Record parent, tester, and cross IDs, mating design, family size, balance among crosses, environments, trait definitions, and checks. Keep the tester set explicit because combining ability is defined relative to that mating population.
Record 2
Measure family means, within-family variance, parental performance, fertility, off-types, and trait covariance across environments. Retain the raw progeny distribution rather than only top-family summaries.
Record 3
Document the statistical model used for general and specific combining ability, uncertainty of estimates, tester representation, selected crosses, and whether conclusions were confirmed in an independent generation or environment.
Common misconceptions
Evidence limits and uncertainty
General and specific combining-ability estimates are conditional on the tested parent and tester population and should not be treated as universal properties.
Unbalanced family sizes, selective missingness, and weak environmental replication can bias combining-ability estimates and inflate confidence.
Check your reasoning
- For "Progeny Testing and Combining Ability", which records are required to make the result traceable and decision-ready, and which missing field would most weaken the conclusion?
- A learner claims, "The best parent phenotype always gives the best progeny." Use the lesson’s science and evidence limits to explain why that claim is unreliable, then name one observation or measurement that could separate the competing explanations.
- Applied case — Decide whether the program wants elite individual clones, predictable seed families, or parental lines. Design progeny tests to match that objective. Build a verification plan using the lesson’s record set (Parent/tester/cross IDs; mating design; family size and balance; environments; trait definitions; family means and variances; parental and check values; GCA/SCA model and uncertainty; selected progeny; fertility and off-types.). What would you compare before and after the action, and what result would make you revise the original interpretation?
Require lesson-specific evidence, not memorized universal targets. Open the rationales after you have written or discussed your own answer.
Answer rationale 1: Mechanism / workflow rationale
- A strong answer should connect the response to the lesson objective: Estimate parental breeding value from offspring performance across mates, families, environments, and target traits.
- A progeny test evaluates a parent by measuring its descendants. This is essential when the parent’s own phenotype is misleading, when traits are expressed only in one sex or late stage, or when selection aims at transmission rather than clone performance. Families can be half-sibs, full-sibs, selfs, testcrosses, or structured diallel and factorial sets.
- General combining ability describes average performance across several mates and is often associated with additive effects. Specific combining ability describes an unusually favorable or unfavorable combination relative to parental averages and can include dominance and epistasis. Estimates depend on the mating design, tester choice, environments, family size, and statistical model. A parent tested with one mate does not have a reliable general combining-ability estimate.
- The most useful verification evidence includes Record parent, tester, and cross IDs, mating design, family size, balance among crosses, environments, trait definitions, and checks. Keep the tester set explicit because combining ability is defined relative to that mating population..
- Keep this limit explicit: General and specific combining-ability estimates are conditional on the tested parent and tester population and should not be treated as universal properties.
Answer rationale 2: Misconception rationale
- The shortcut is unreliable because the lesson explicitly teaches a more conditional explanation.
- Representative misconception: The best parent phenotype always gives the best progeny. Individual phenotype and breeding value are related imperfectly, especially for low-heritability or strongly environment-dependent traits.
- A progeny test evaluates a parent by measuring its descendants. This is essential when the parent’s own phenotype is misleading, when traits are expressed only in one sex or late stage, or when selection aims at transmission rather than clone performance. Families can be half-sibs, full-sibs, selfs, testcrosses, or structured diallel and factorial sets.
- A useful discriminator is Measure family means, within-family variance, parental performance, fertility, off-types, and trait covariance across environments. Retain the raw progeny distribution rather than only top-family summaries..
- Do not overextend the conclusion beyond this limit: General and specific combining-ability estimates are conditional on the tested parent and tester population and should not be treated as universal properties.
Answer rationale 3: Applied verification rationale
- In practice: Decide whether the program wants elite individual clones, predictable seed families, or parental lines. Design progeny tests to match that objective.
- Record before action: Record parent, tester, and cross IDs, mating design, family size, balance among crosses, environments, trait definitions, and checks. Keep the tester set explicit because combining ability is defined relative to that mating population..
- Also record: Measure family means, within-family variance, parental performance, fertility, off-types, and trait covariance across environments. Retain the raw progeny distribution rather than only top-family summaries..
- After the action, repeat the same measurement or observation so the comparison is valid.
- Revise the interpretation if the result conflicts with the lesson limit or the expected response: General and specific combining-ability estimates are conditional on the tested parent and tester population and should not be treated as universal properties.
Related lessons
Sources and evidence
- Allard — Principles of Plant BreedingV20-SRC-004
Foundational mating systems, selection, population improvement, backcrossing, and line development; general plant breeding, not Cannabis-specific.
- Falconer and Mackay — Introduction to Quantitative GeneticsV20-SRC-006
Inbreeding, variance, heritability, selection response, and quantitative-trait foundations; model assumptions must be stated.
- Bernardo — Breeding for Quantitative Traits in PlantsV20-SRC-007
Selection, prediction, multi-environment testing, genetic gain, and genomic selection; general crop-breeding source.
- Machine-learning multi-trait genomic prediction for cannabinoids, 2025V20-SRC-025
High-density genotyping and genomic-prediction research; prediction depends on training population and validation.
Downloads
No lesson-specific download is approved for this release. Use browser print/save-to-PDF when you need an offline reading copy.