Heterosis in crop mutant crosses and production of high yielding lines using doubled haploid systems

Heterosis appearing in crosses between mutants derived from the same parent variety and crosses of mutants with parent varieties has been observed by many authors for many plant species including such important crops as barle y, faba bean, maize, pea, pearl millet, rice, sunflower, sesame, sweet clov er, triticale and wheat. Mutant heterosis was reported for crosses of spont aneous mutants, mutants obtained after treatment with various mutagens and recently also for somaclonal variants. The heterotic effects are usually re lated to an increase in some yield components. There is a lack of correlati on between mutation of a particular plant character and appearance of heter otic effect. The yielding performance of a mutant was not correlated with i ts potential to give yield heterosis in crosses with another mutant, often also a very poor one, or with the parent variety. Poor yielding barley muta nts from the collection of semi-dwarf forms of the Department of Genetics, Silesian University gave heterosis in crosses with other mutants or parent varieties for such characters as tillering, grain number and weight per pla nt. In addition to mutants with deleterious mutations of such characters as chlorophyll synthesis or fasciated stem there were also lines mutated for earliness, semi-dwarfness, low glucoside or high protein and lysine content which gave significant heterosis in crosses. One possible explanation of t he phenomenon of mutant heterosis is related to the frequency of mutations induced by chemical and physical mutagens. The appearance and the level of heterotic effect of mutated genes will depend on their interaction with oth er mutated genes or with genes from the parental genotype. High specific co mbining ability of mutants giving heterotic effect makes hybrid seed produc tion, based on crosses with defined sources of cytoplasmic or genetic male sterility, unfeasible or even impossible. Doubled haploids provide a unique system to attempt the `fixing' of hybrid performance in homozygous lines a nd to avoid the step of hybrid seed production. The assumption on the `fixa bility' of hybrid yield in homozygous lines based on reports that large add itive genetic variance is responsible for yield heterosis in wheat or barle y was proven also for mutant crosses.