Talk:Response to selection
Discussion page of Response to selection
More actions
Talk
Discussion for: Response to selection • View • Links • Related changes
Use this page for verification notes, disputes, and action items. Keep narrative content on the article.
Page scope
Response to selection covers the breeder's equation and what follows from it: selection intensity, selection of one sex, the source of the gain, realised heritability, gain per year, responses in other traits and environments, departures from the prediction, use of the equation in the wild, and what selection has done in cannabis. Variance components and the estimation of heritability belong to Heritability, the method of choosing plants and the size of the breeding population to Mass selection, the ways of choosing parents and the Illinois maize experiment to Artificial selection, and the inheritance of particular traits to Genetic traits.
Data verification
Record source conflicts, figures that differ between studies and sources not used here. Keep narrative content on the article.
Discussion
Sources
- Beavis and colleagues (2023), Quantitative Genetics for Plant Breeding, chapter 9: the differential, the regression of offspring on the mid-parent, the equation, Hanson's definition, realised heritability and the bias from epistasis.
- Felsenstein (2019), Theoretical Evolutionary Genetics, pp. 443–450: selection of one sex, the table of intensities, the mouse example, the source of the gain, Bulmer's approximation, complications and the history.
- Hill (2014), Genetics: Lush and his school, Lande, Robertson and Price, Falconer's realised heritability and correlation across environments, the Edinburgh and pig experiments, selection limits.
- Cobb and colleagues (2019), Theoretical and Applied Genetics, and Cooper and colleagues (2023), Frontiers in Plant Science: the plant breeders' forms of the equation.
- Morrissey and colleagues (2010), Journal of Evolutionary Biology and (2012), Evolution; Pujol and colleagues (2018), Trends in Ecology & Evolution: the equation in natural populations.
- Hennink (1994), Euphytica; de Meijer (1994), Diversity in Cannabis; Bredemann and colleagues (1956), Bulletin on Narcotics; Baluch (2026); Mishchenko (2020), Plant Breeding and Seed Production, and Mishchenko and colleagues (2021), Plant Varieties Studying and Protection: selection in cannabis.
- Schilling and colleagues (2022), The Plant Journal: generation time in hemp. Yoosefzadeh Najafabadi and Torkamaneh (2024), The Plant Journal: genomic prediction of cannabinoids. Petit and colleagues (2020), Naim-Feil and colleagues (2023), Clarke and Merlin (2016).
Where the sources disagree
- Who gave the equation, and when. Beavis and colleagues write that Lush defined the response in 1940. Hill calls it Lush's breeder's equation without a date, and Cooper and colleagues trace the structure of the equation to Lush's book of 1937 among other works.
- The symbol for the differential. Felsenstein writes I and Beavis and colleagues S.
- The gain equation. Cobb and colleagues write ΔG = σair/L with additive genetic variation described as σa. Beavis and colleagues write R = ih2σP per cycle.
- The Illinois experiment. Hill counts more than 100 generations. Laurie and colleagues (2004) report generation 89 for the oil lines they mapped.
- Hemp fibre figures. Bredemann and colleagues give about 12 rising to about 24 per cent. Hennink gives about 12–15 rising to 25–35 per cent, over a longer period and across breeding in general. The figures are the breeders' own, or summaries of them.
How far the evidence goes
- Hennink's summary reports that bast fibre content had a high heritability and that mass selection for it was efficient. Naim-Feil and colleagues (2021) quote his narrow-sense estimate of 0.22 for stem diameter.
- The cannabis results differ in kind: selection of one sex over breeding programmes, single-plant selection with selfing, a cultivar compared with a standard, and years of mass selection. The Vik 2020 figures compare the cultivar with the standard Hliana. Its starting lines are given only as above 0.5 per cent CBG, so the difference between 0.002 and 1.034 per cent is not a response.
- Mishchenko and colleagues (2021) cite work attributing high CBG to a single recessive allele. The breeder's equation is framed for traits under many loci, though Felsenstein notes that Latter found a few loci of large effect rarely cause serious difficulties.
- Schilling and colleagues report generation time. Cobb and colleagues' argument about cycle time concerns other crops.
- The broad-sense estimates of Naim-Feil and colleagues (2023) include non-additive variance. Their forecast about cannabinoid composition is their own judgement, and it is not a measured response.
- Figure 3 of Mishchenko and colleagues (2021) prints the THC class limits on its CBG axis. Their Table 1 and text give CBG from 0.8538 to 1.2242 per cent.
- Morrissey and colleagues' Soay sheep results concern four morphological traits in one population.
Terms
- Response to selection, selection response and genetic gain are used interchangeably by Beavis and colleagues.
- Bark fibre (de Meijer) and bast fibre (Hennink) refer to the same tissue here.
- Selection intensity and standardised selection differential are the same quantity, i.
Images reserved in the article
One image has a place in the article but is not yet uploaded:
File:Selection-intensity-by-share-kept.jpg(Selection intensity), a chart to be drawn from the twelve values of Felsenstein's table (p. 446): selection intensity against the share kept, on a logarithmic scale.