🧭 Overview
🧠 One-sentence thesis
Recurrent selection is a cyclical breeding method that progressively increases the frequency of favorable alleles in a population through repeated cycles of selection and intermating, with different approaches suited to synthetic, clonal, pure-line, and hybrid cultivar development.
📌 Key points (3–5)
- What recurrent selection does: cycles of selecting superior individuals, intermating them, and using the resulting population as the starting point for the next cycle, progressively enriching favorable alleles.
- Synthetic cultivars: the improved population itself becomes the cultivar; most extensive use of recurrent selection, especially in forage and turf species.
- Selection timing matters: eliminating undesirable plants before flowering (so they don't contribute pollen) produces greater frequency shifts than eliminating them after flowering.
- Common confusion: clonal vs. pure-line vs. hybrid cultivars require different recurrent selection strategies because they differ in how easily self-pollinated or hybrid seed can be obtained, the impact of inbreeding depression, and the role of combining ability.
- Phenotypic vs. genotypic selection: phenotypic selection evaluates individual plants directly; genotypic selection evaluates families (half-sib, full-sib, or selfed progeny) to assess combining ability or breeding value.
🔄 Core mechanism: cycles and population improvement
🔄 What a cycle is
The initial population used for recurrent selection is referred to as cycle 0. Each subsequent cycle of selection is identified with a consecutive number.
- Each cycle involves: evaluate individuals → select the best → intermate them → harvest seed in bulk → that seed becomes the next cycle population.
- Example: Cycle 0 → select resistant plants → intermate → bulk seed = Cycle 1 → repeat for Cycle 2, 3, etc.
- The goal is to increase the frequency of favorable alleles step by step.
🧬 Why intermating selected individuals matters
- Key principle: By allowing only selected (e.g., resistant) plants to intermate, the frequency of favorable alleles in the next cycle will be greater than if selected and unselected plants had intermated and seed was harvested only from the selected ones.
- This principle is detailed on pages 174–178 of chapter 15.
- Example: If you let resistant and susceptible plants intermate, susceptible pollen fertilizes resistant plants, diluting the favorable allele frequency; if you remove susceptible plants before flowering, only resistant pollen is available.
🌱 Clones and genotypes
- In recurrent selection, each plant is a different genotype.
- Individual plants (genotypes) are commonly referred to as clones.
- Breeders often evaluate large numbers of individuals each cycle (e.g., 10,000 plants in the FreedomMR example).
🌾 Synthetic cultivars: the population is the product
🌾 What synthetic cultivars are
- The improved population itself is used as the cultivar, not individual selected plants.
- Most extensive use of recurrent selection is for synthetic cultivars, particularly for forage and turf species.
- Phenotypic selection alone or in combination with genotypic selection is common practice.
🦠 Example: FreedomMR (phenotypic selection only)
Cycle 0: The synthetic cultivar Freedom was the starting population.
Cycle 1:
- ~10,000 plants germinated in greenhouse, exposed to natural powdery mildew infections.
- ~40% susceptible plants discarded.
- Remaining 60% transplanted to field; ~8.7% more discarded as susceptible.
- Remaining resistant plants allowed to intermate by insect pollination.
- Seed harvested in bulk with a combine = Cycle 1 population.
Cycles 2–5: Same procedure repeated; selection intensity (percentage of resistant plants selected) varied among cycles.
Final cultivar: Cycle 5 resistant plants further selected for absence of pubescence; resistant, non-pubescent plants intermated in a polycross to produce breeder (Syn 1) seed of FreedomMR.
🌾 Example: Warrior indiangrass (phenotypic + genotypic selection)
Cycle 0: The cultivar Oto (from native prairies in southern Nebraska and eastern Kansas).
Cycle 1 (genotypic selection via half-sib families):
- 146 individual clones from Cycle 0 planted in a polycross.
- Clones intercrossed naturally by wind pollination.
- Seed harvested from each of the 146 clones (each clone served as the tester; the 146 clones' pollen fertilized each other).
- Progeny from each clone planted in two field replications; data collected for quality and yield.
- 29 of 146 clones selected based on performance.
- Two cuttings from each of the 29 clones planted in a polycross with two replications.
- 29 clones intercrossed by wind; seed from each clone harvested separately; equal quantity from each bulked = Cycle 1 population.
Cycle 2 (restricted recurrent phenotypic selection, RRPS):
- Gridding: 53 rows, each with 14 plants; ~3 best plants per row harvested; plants in one row not compared to plants in other rows.
- 39 plants selected and grown in a polycross; seed from polycross = Cycle 2 population.
Cycle 3 (RRPS again):
- 38 selected plants grown in a polycross; seed from polycross = cultivar Warrior.
🔬 Polycross details
A polycross is a nursery in which selected clones are intermated.
- Breeder options:
- Grow a single plot of each plant OR clonally propagate each plant into multiple replications.
- Harvest seed from all clones in bulk OR harvest clones individually and bulk equal quantities of seed from each.
- Details on alternative procedures: pages 181–188 of chapter 15.
🧪 Half-sib families and general combining ability
- For species with open pollination (wind or insects), half-sib families are formed in a polycross.
- Seed harvested from each clone in the polycross has a common parent (the tester).
- The clones themselves serve as the tester because their pollen fertilizes each other.
- Evaluation of half-sib seed determines the general combining ability of the clones for quality and yield.
- Note: Starting with half-sib family evaluation (as in Warrior Cycle 1) is less common than conducting recurrent phenotypic selection first for highly heritable traits, then evaluating half-sib families in later cycles.
🌿 Maintenance nursery
- During field testing of progeny, each clone is grown in a maintenance nursery.
- If a clone is selected for future use, vegetative tissue for propagation is taken from the plant in the maintenance nursery.
🧬 Clonal propagation principle
- By clonally propagating selected individuals (e.g., the 29 clones in Warrior Cycle 1), all female and male gametes involved in producing the next cycle seed come from selected individuals.
- This principle is described on pages 106–110 of chapter 8.
🌿 Clonal, pure-line, and hybrid cultivars: improving source populations
🌿 Key difference from synthetics
- For clonal, pure-line, and hybrid cultivars, recurrent selection improves populations from which superior individuals are selected.
- Unlike synthetic cultivars, the improved population is not used as a cultivar per se.
🌿 Clonal cultivars
- Most common method: recurrent phenotypic selection.
- Why: Combining ability is not a factor in identifying individuals that will perform well as a clonal cultivar, which minimizes the value of testing half-sib or full-sib families.
- Inbreeding depression limits the value of testing selfed progeny.
🌾 Pure-line cultivars
- Most common method: using self-pollinated individuals and their progeny, because self-pollinated seed is readily obtained.
- Recurrent phenotypic selection: can be used for quantitative traits with high heritability, if enough hybrid seed can be obtained for the next cycle when intermating selected individuals.
- Half-sib selection: possible if enough seed can be obtained from an individual when crossed to a tester.
- Genetic male sterility: has been used to facilitate production of hybrid seed by open pollination for recurrent phenotypic and half-sib selection (chapter 16).
- Full-sib selection: limited by difficulty of producing hybrid seed when crossing two individuals.
🌽 Hybrid cultivars
- All methods of recurrent selection are technically possible for improving populations from which inbred lines are obtained for use in hybrids.
- Comparison of methods for genetic gain will be discussed in Principles of Cultivar Development under maximizing genetic gain.
🔍 Factors influencing method choice
The type of recurrent selection is influenced by:
- Feasibility of obtaining self-pollinated seed.
- Impact of inbreeding depression.
- Feasibility of obtaining hybrid seed.
- Role of combining ability in assessing genetic potential of an individual.
🧮 Selection timing: before vs. after flowering
🧮 Why timing matters
- Key principle: Eliminating undesirable plants before flowering produces a greater shift in allele frequency than eliminating them after flowering.
- Reason: If undesirable plants flower, their pollen can fertilize desirable plants, diluting the favorable allele frequency in the next generation.
🧮 Worked example scenario (from review questions)
Setup: Random-mated maize population; 64% susceptible, 36% resistant; resistance controlled by dominant allele P.
Scenario 1: Eliminate susceptible plants after flowering
- Susceptible plants have already contributed pollen.
- Seed is harvested only from resistant plants, but some of that seed was fertilized by susceptible pollen.
- Result: Cycle 1 population will have a certain frequency of P and p alleles.
Scenario 2: Eliminate susceptible plants before flowering
- Susceptible plants do not contribute pollen.
- Only resistant plants contribute both female and male gametes.
- Result: Cycle 1 population will have a higher frequency of the P allele than in Scenario 1.
🧮 Genotypic and phenotypic frequencies
- After selection and intermating, the next cycle population will have different genotypic frequencies (PP, Pp, pp) and phenotypic frequencies (resistant vs. susceptible).
- If resistant S₀ plants are self-pollinated, some S₀:₁ lines will be heterogeneous (segregating) for resistance if the S₀ parent was Pp.
- Within a heterogeneous line, genotypic and phenotypic frequencies follow Mendelian ratios.
🔬 Phenotypic vs. genotypic selection
🔬 Phenotypic selection
- Evaluates individual plants directly based on their observable traits.
- Example: FreedomMR used only phenotypic selection—plants were visually screened for disease resistance and pubescence.
- Common for highly heritable traits.
🔬 Genotypic selection
- Evaluates families (progeny) to assess the breeding value or combining ability of parents.
- Types of families:
- Half-sib families: progeny from one known parent and multiple unknown pollen donors (common in polycross).
- Full-sib families: progeny from two known parents.
- Selfed progeny: progeny from self-pollination.
- Example: Warrior Cycle 1 evaluated 146 half-sib families to determine general combining ability.
🔬 Combining phenotypic and genotypic selection
- Some breeding programs use phenotypic selection in early cycles (to quickly improve highly heritable traits) and genotypic selection in later cycles (to assess combining ability or breeding value for complex traits).
- Example: Warrior used genotypic selection (half-sib families) in Cycle 1, then phenotypic selection (RRPS) in Cycles 2 and 3.
🧩 Restricted recurrent phenotypic selection (RRPS)
🧩 What RRPS is
- A variant of recurrent phenotypic selection described on page 181 of chapter 15.
- Features include gridding and intercrossing only selected individual clones.
🧩 Gridding
- The field is divided into a grid (e.g., 53 rows, each with 14 plants).
- A fixed number of best plants are selected from each row (e.g., ~3 plants per row).
- Key principle: Plants in one row are not compared to plants in other rows.
- This ensures selection is distributed across the entire population, not just concentrated in a few high-performing areas.
🧩 Intercrossing selected clones
- After evaluation, selected plants (e.g., 39 plants in Warrior Cycle 2) are grown in a polycross.
- Seed harvested from the polycross constitutes the next cycle population.
📚 Key definitions and terms
Recurrent selection: A cyclical breeding method involving repeated cycles of selection and intermating to progressively increase the frequency of favorable alleles in a population.
Cycle 0: The initial population used for recurrent selection; each subsequent cycle is identified with a consecutive number.
Clone: An individual plant (genotype); in recurrent selection, each plant is a different genotype.
Polycross: A nursery in which selected clones are intermated by open pollination (wind or insects).
Half-sib family: Progeny from one known parent (the tester) and multiple unknown pollen donors; used to assess general combining ability.
General combining ability: The average performance of a clone's progeny when crossed with multiple other clones.
Maintenance nursery: A field nursery where clones are grown during progeny testing; vegetative tissue for propagation is taken from plants in this nursery if the clone is selected.
Restricted recurrent phenotypic selection (RRPS): A method using gridding and intercrossing only selected clones.
Gridding: Dividing the field into a grid and selecting a fixed number of best plants from each grid unit, without comparing across units.
Synthetic cultivar: A cultivar in which the improved population itself is the product, not individual selected plants.
🔄 Don't confuse
| Concept | What it is | What it is NOT |
|---|
| Synthetic cultivar | The improved population itself is the cultivar | Individual selected plants are the cultivar |
| Phenotypic selection | Evaluates individual plants directly | Evaluates families (progeny) |
| Genotypic selection | Evaluates families to assess breeding value | Evaluates individual plants directly |
| Half-sib family | One known parent, multiple unknown pollen donors | Two known parents (that's full-sib) |
| Eliminating plants before flowering | Only selected plants contribute pollen | Selected and unselected plants both contribute pollen (after flowering) |
| Clonal cultivar breeding | Combining ability not a factor; phenotypic selection most common | Combining ability is key (that's hybrid breeding) |
| Gridding in RRPS | Plants in one row not compared to other rows | All plants compared across the entire field |