Plant Breeding 25 Jun 2026 · 14 views · By Shrikant Sir

PB Ch 27. Recurrent Selection

If standard mass selection is like picking the tallest students in a class, Recurrent Selection is like running an elite training academy over multiple generations.
The core philosophy of recurrent selection began taking shape between 1919 and 1920 with researchers like Hayes, Garber, East, and Jones. In 1940, Jenkins published the first detailed methodology based on his maize experiments. However, it was Hull in 1945 who officially named it "recurrent selection" and defined its modern purpose in 1952.
The first detailed description was published by Jenkins in 1940 based on experiments with early testing for GCA in maize.
Hull (1945) named the method 'recurrent selection' and in 1952 defined it as: 'A method which involves reselection generation after generation with interbreeding of selects to provide for genetic recombination.'

Recurrent Selection: A cyclical selection programme in which (1) plants are selected from a population, (2) tested for their breeding value (with or without progeny testing), and (3) intermated among themselves to form the improved base population for the next selection cycle.
The objective is to gradually increase the frequency of favourable alleles in the population without reducing genetic variability appreciably.

Advantages of Recurrent Selection

Rate of inbreeding can be kept at a low level — prevents excessive homozygosity and inbreeding depression
Frequency of favourable genes in the population gradually increases with each selection cycle
Greater opportunity for genetic recombination in each cycle — novel favourable gene combinations can emerge

Type 1 — Simple Recurrent Selection (Phenotypic Recurrent Selection)

Year	Generation	Action
Year 1	Original selection cycle — Year 1	Select several phenotypically superior plants from source population. Self each selected plant. Harvest selfed seed separately. Save for Year 3 planting.
Year 2	Original selection cycle — Year 2	Grow individual plant progeny rows from the selfed seed (from Year 1). Intercross all progeny rows in all possible combinations. Take equal amounts of seed from each cross and mix. This mixed seed = source population for the next selection cycle.
Year 3	First selection cycle — Year 1	Plant the mixed seed. Select several phenotypically superior plants. Self each selected plant. Harvest selfed seeds separately. Save for Year 5.
Year 4	First selection cycle — Year 2	Grow progeny rows. Intercross in all possible combinations. Composite equal amounts of seed from each cross. This forms the source for the second cycle.
Year 5 onward	Second/third selection cycle	Repeat as above for one or two more cycles. After 3-5 cycles of selection, the improved population has higher frequency of desirable alleles.

Application: Useful for characters that can be measured on individual plants with high heritability. Simple to implement. However, no progeny test — selection efficiency depends on heritability of the trait.

Type 2 — Recurrent Selection for General Combining Ability (RSGCA)

Proposed by Jenkins in 1935.

Tester: A BROAD-BASED TESTER — open-pollinated variety, synthetic variety, or segregating generation. This evaluates GCA of each selected plant.

Year	Step	Action
Year 1	Selection + Crossing	Select several superior plants from source population. Each plant is SIMULTANEOUSLY: (a) selfed → selfed seed saved for Year 3; (b) crossed with broad-based tester → test-cross seed harvested separately.
Year 2	Test cross evaluation	Replicated yield trial conducted using test-cross seeds. Superior progenies identified based on yield trial performance.
Year 3	Recombination	Selfed seed (from Year 1) of plants that produced SUPERIOR test-cross progenies in Year 2 is planted in progeny rows. These selfed progenies intercrossed in all possible combinations. Equal seed from each intercross composited → source population for next selection cycle.
Years 4-6	Second cycle	Repeat Year 1-3 procedure for a second selection cycle. Process may be repeated for one or two more cycles as needed.

End use

Improved population may be released as a synthetic variety; OR
Population with increased frequency of desirable genes used for isolating superior inbreds for hybrid development.

Type 3 — Recurrent Selection for Specific Combining Ability (RSSCA)

Proposed by Hull in 1945.
Tester: A NARROW-BASED INBRED TESTER — a specific inbred line with which the selected plants need to combine well.
Procedure: IDENTICAL to RSGCA except the tester used is a specific inbred line rather than a broad-based tester. Plants selected for their SCA with the specific inbred tester — these would become good parents in hybrids with that inbred.

Application: Used to develop populations whose plants will perform well as one parent in hybrids with a specific elite inbred. Useful for hybrid breeding — identifies lines with favourable SCA for exploitation of non-additive (dominance/epistasis) genetic variance.

Type 4 — Reciprocal Recurrent Selection (RRS)

Proposed by Comstock, Robinson, and Harvey in 1949.

Reciprocal Recurrent Selection: Two source populations (A and B — typically representing two heterotic groups) are improved SIMULTANEOUSLY.
Plants from Population A are tested by crossing with random plants from Population B, and vice versa. The two populations serve as reciprocal testers for each other.

Year	Step	Action
Year 1	Selection + Crossing in both populations	From Population A: select superior plants; self each plant (save selfed seed for Year 3); cross each selected plant from A with RANDOM plants from B (test-cross seeds from A harvested). From Population B: select superior plants; self each (save for Year 3); cross each selected plant from B with random plants from A (test-cross seeds from B harvested).
Year 2	Test cross evaluation	Two separate replicated yield trials: (1) test-cross progeny rows from Population A (tested against B); (2) test-cross progeny rows from Population B (tested against A). Plants from Year 1 that produced superior test-cross progenies identified.
Year 3	Recombination in each population separately	Selfed seed from Year 1 of superior plants of Population A planted in separate crossing plot; all possible intercrosses among these made; equal seed from each intercross composited → new source population A. Same process for Population B separately → new source population B. These are the improved A and B source populations for the next selection cycle.
Years 4-6	Second cycle	Repeat Years 1-3. Population A becomes more suited to combining with Population B, and vice versa. After several cycles, the heterotic pattern between A and B becomes stronger.

Comparison and Efficiency of Recurrent Selection Type

Condition	Relative Efficiency
If dominance is INCOMPLETE (partial dominance)	RSGCA and RRS are equal in efficiency but BOTH are superior to RSSCA
If dominance is COMPLETE	All three methods (RSGCA, RSSCA, RRS) are EQUAL in efficiency
If OVERDOMINANCE is present	RRS and RSSCA are equally effective; BOTH are superior to RSGCA

Practical implication: Since the genetic basis of heterosis in most crops appears to involve a combination of dominance and overdominance, Reciprocal Recurrent Selection is generally the most versatile and powerful choice — it exploits BOTH additive and non-additive genetic variance simultaneously. This is why RRS has become the foundation for developing heterotic groups in modern maize breeding programmes.

6.7 RSGCA vs RSSCA — Key Distinction

Feature	RSGCA	RSSCA
Proposed by	Jenkins, 1935	Hull, 1945
Tester type	BROAD-BASED — OPV, synthetic variety, or segregating generation	NARROW-BASED — specific inbred line
What is estimated	GCA — general combining ability across many crosses	SCA — specific combining ability with the tester inbred
Genetic variance exploited	Additive genetic variance (VA)	Non-additive (dominance + epistasis) genetic variance (VD + VI)
Best when	Dominance is incomplete; additive variance predominates	Overdominance is present; non-additive variance predominates
End product	Improved population; used to develop synthetic varieties or isolate high-GCA inbreds	Population whose plants combine specifically well with the tester inbred; used in hybrid programmes
Indian application	Maize populations (e.g., DMR, CIMMYT population improvement programmes)	Developing parental lines for specific hybrid combinations in sorghum, bajra

IFoS 2022 (Q2b, 15M) — Explain the method of recurrent selection. Write the importance of combining ability in crop improvement.
IFoS 2023 (Q2biii, 5M) — Differentiate between RSGCA and RSSCA.
CSE 2018 (Q2c, 10M) — Discuss recurrent selection schemes for population improvement.

← PB Ch 26. GCA and SCA

PB Ch 28. Synthetic and Composite Varieties →