Mass selection
More actions
Mass selection is a method of plant breeding in which plants are chosen from a population on their own phenotype and the seed of the chosen plants is bulked to sow the next generation.[1][2] It was the earliest method used to improve cross-pollinated crops. Farmers who each year sowed bulked seed from the best ears in their maize fields produced an array of genetically different open-pollinated cultivars.[1] Drug-type and fibre-type cultivars of Cannabis were both developed by mass selection for many years. Cultivated landraces arose through it.[3]
The change mass selection makes in a population depends on the heritability of the trait selected, the share of its variation that parents pass on to offspring.[4] Selection on characters of high heritability is effective.[2] In an outcrossing crop the method has a second limit. When seed is taken from chosen plants after they have been pollinated, only the seed parents have been selected. The expected gain is then half of what is gained when the pollen parents are also chosen.[5] In dioecious cannabis pollen and seed are borne on different plants,[3] so the choice of which plants are left to shed pollen decides which of the two cases applies.
One cycle of selection
Mass selection is one of the methods of recurrent selection. A grower starts from a population, picks the plants that best show the wanted character, harvests their seed and sows it in bulk. Each time a new population is formed in this way a cycle of selection is complete. The starting population is cycle 0 and the next is cycle 1.[1] The plants are judged on their own appearance or measurements. Some other forms of recurrent selection judge plants by the performance of their progeny.[1]
The name has been used in a broad and a narrow sense. Breeders have called cyclic selection of individual plants in a cross-pollinated population mass selection, recurrent phenotypic selection, phenotypic recurrent selection, simple recurrent selection or directed mass selection. The terms are largely used interchangeably. Some keep "mass selection" for the case in which the chosen plants are seed parents already pollinated by chosen and unchosen plants alike. They use "recurrent phenotypic selection" when only chosen plants are crossed with one another.[1]
Plants in one field differ in their genes and also in the soil, fertility and moisture where they grow. A breeder who picks the highest-yielding plants may take them all from the most fertile corner of the field.[5] In 1961 C.O. Gardner divided a maize selection field into a grid of 100 blocks of 40 plants each and kept the four best plants in every block, a tenth of the population, so that each was compared only with its neighbours.[5] A study of maize in Indonesia published in 2022 used the same layout: 100 subplots of 40 plants, with the best 5 per cent kept in each.[2]
Response to selection
The selected plants differ from the population they came from by the selection differential, the difference between their mean and the population mean.[4] Their offspring differ from it by less. The response to selection, the change in the mean from one generation to the next, is the selection differential multiplied by the narrow-sense heritability, R = h²S. Jay Lush defined the response in 1940. The formula is known as the breeder's equation.[4] The narrow-sense heritability is the additive part of the genetic variance divided by the phenotypic variance. It equals the slope of the regression of offspring on the mean of their two parents.[4] With a heritability of 0.3, parents chosen 10 kg above the population mean give offspring that average 3 kg above it.[6]
The response comes from changes in allele frequencies at the loci that affect the trait, so the gain is kept in later generations.[7] Plants that stand out only because they grew on a better spot are not superior genetically.[5] Mass selection works best on characters of high heritability.[2]
In a study of 123 hemp accessions grown at three sites in northern, central and southern Europe, the time of flowering, the sex of the plants and the content of bast fibre all had broad-sense heritabilities between 0.93 and 0.96. Most agronomic traits, the fineness of the fibre and the pectin-related sugars had low genetic components. Their rankings changed from site to site.[8] In hemp cultivars from European and Asian breeding programmes grown in Colorado in 2016, genotype accounted for 97 per cent of the variance in days to maturity at the one site where it was measured, over 80 per cent in THC content and 83 per cent in CBD content. Genotype accounted for 28 per cent of the variance in grain yield. Environment accounted for 25 per cent and the interaction of genotype with environment for 35 per cent.[9] In the review by Barcaccia and colleagues, the cannabinoid profile of a plant, its chemotype, depends strongly on genotype, while the quantity of cannabinoids is affected by growing practice, environment and genotype-by-environment interaction.[3] Both trials measured differences between accessions or cultivars, which include non-additive genetic effects. The response to mass selection within one population depends on the additive part alone.[4]
Choosing the pollen parent
Every seed has a seed parent and a pollen parent. When the seed parents are chosen after they have been pollinated by chosen and unchosen plants alike, only the half of the alleles carried in the egg cells is selected. The other half arrives in pollen that nobody chose and adds nothing to the gain.[10] In the breeder's equation the response to selection of one sex, with the other mated at random, is half the response to selection of both.[6] Plant breeders call this parental control. It rises from one-half to one when plants are chosen before the seed parents are pollinated.[10] J.R. Harlan chose both parents in his selection of side-oats grama from 1943, having found in earlier work that selection was not effective when only the seed parent was chosen.[1] In the Indonesian maize study, plants were chosen for height and leaf number before they flowered and only the chosen plants were crossed. After seven cycles the yield of the population was 43 per cent higher than at the start.[2]
Cannabis is wind-pollinated and in nature reproduces only by outcrossing. Its male and female flowers grow on separate plants.[3] A grower who removes male plants before they shed pollen, leaving only chosen pollen sources, selects both parents of the next generation. A grower who keeps seed from the best female plants of a field in which every male flowered selects the seed parent only.[10]
How many parents
Choosing few plants raises the immediate response but shrinks the population that breeds.[7] In a population with separate sexes the effective population size is about 4NfNm / (Nf + Nm), where Nf and Nm are the numbers of females and males that breed. When one sex is much scarcer than the other, the effective size approaches four times the number of the scarcer sex. A population of 5 females and 95 males has an effective size of about 20. A field in which most male plants have been culled breeds from few fathers. Its effective size is then set mainly by their number.[11] Genetic drift in a small breeding population lowers heritability on average. In any population of finite size the additive genetic variance is eventually lost and selection stops making progress. For the final limit of selection, as distinct from the gain in one generation, Robertson showed that the best course is to keep the upper half of the population.[7] Effective recurrent selection needs a high level of genetic variation for the characters chosen. The chance that different alleles are present rises with the number of parents.[12]
Limits of the method
Mass selection is a form of artificial selection. Natural selection acts on the same population at the same time. Plants that best meet a breeder's needs may be among the least fertile. Their offspring may be among the least likely to survive. The two kinds of selection then work against each other. The response can reach a plateau before the additive variance is used up.[7] Changes in growing conditions over the years can hide a real response or be mistaken for one. Improved husbandry may raise yields that selection is credited with, while a deteriorating environment can make an effective selection programme appear to fail.[7]
In cannabis
Populations of cannabis bred by mass selection are made up of highly heterozygous genotypes that share a common gene pool, as natural populations are.[3] Landraces formed in this way show wide variation in form and chemistry. Clones propagated by cuttings from single superior plants are more uniform and more stable in appearance.[3]
See also
References
- ↑ 1.0 1.1 1.2 1.3 1.4 1.5 Fehr, W.R. (1991) [1987]. "Recurrent selection". Principles of cultivar development: Theory and technique. Vol. 1. Ames, Iowa: Walter R. Fehr. pp. 172–175. ISBN 0-9635989-0-2.
- ↑ 2.0 2.1 2.2 2.3 2.4 Sudika, I.W.; Soemeinaboedhy, I.N. (2022). "The improvement of corn yield and fresh stover weight through two mass selection techniques in dry land". Journal für Kulturpflanzen. 74 (5–6): 134–141. doi:10.5073/JfK.2022.05-06.04.
- ↑ 3.0 3.1 3.2 3.3 3.4 3.5 Barcaccia, G.; Palumbo, F.; Scariolo, F.; Vannozzi, A.; Borin, M.; Bona, S. (2020). "Potentials and challenges of genomics for breeding cannabis cultivars". Frontiers in Plant Science. 11 573299. doi:10.3389/fpls.2020.573299.
- ↑ 4.0 4.1 4.2 4.3 4.4 Beavis, W.; Lamkey, K.; Mahama, A.A.; Suza, W. (2023). "Selection response". In Suza, W.P.; Lamkey, K.R. (eds.). Quantitative genetics for plant breeding. Iowa State University Digital Press. doi:10.31274/isdup.2023.146.
- ↑ 5.0 5.1 5.2 5.3 Fehr, W.R. (1991) [1987]. "Recurrent selection". Principles of cultivar development: Theory and technique. Vol. 1. Ames, Iowa: Walter R. Fehr. pp. 178–181. ISBN 0-9635989-0-2.
- ↑ 6.0 6.1 Felsenstein, J. (December 2019). Theoretical evolutionary genetics. Seattle: Department of Genome Sciences, University of Washington. pp. 443–444.
- ↑ 7.0 7.1 7.2 7.3 7.4 Felsenstein, J. (December 2019). Theoretical evolutionary genetics. Seattle: Department of Genome Sciences, University of Washington. pp. 447–450.
- ↑ Petit, J.; Salentijn, E.M.J.; Paulo, M.-J.; Thouminot, C.; van Dinter, B.J.; et al. (2020). "Genetic variability of morphological, flowering, and biomass quality traits in hemp (Cannabis sativa L.)". Frontiers in Plant Science. 11 102. doi:10.3389/fpls.2020.00102.
- ↑ Campbell, B.J.; Berrada, A.F.; Hudalla, C.; Amaducci, S.; McKay, J.K. (2019). "Genotype × environment interactions of industrial hemp cultivars highlight diverse responses to environmental factors". Agrosystems, Geosciences & Environment. 2 (1): 1–11. doi:10.2134/age2018.11.0057.
- ↑ 10.0 10.1 10.2 Fehr, W.R. (1991) [1987]. "Maximizing genetic improvement". Principles of cultivar development: Theory and technique. Vol. 1. Ames, Iowa: Walter R. Fehr. p. 239. ISBN 0-9635989-0-2.
- ↑ Felsenstein, J. (December 2019). Theoretical evolutionary genetics. Seattle: Department of Genome Sciences, University of Washington. pp. 266–268.
- ↑ Fehr, W.R. (1991) [1987]. "Recurrent selection". Principles of cultivar development: Theory and technique. Vol. 1. Ames, Iowa: Walter R. Fehr. p. 173. ISBN 0-9635989-0-2.