Genotype-environment interaction in winter wheat F₁ progeny

Graduation date: 1970Seven winter wheats were evaluated for plant height, the components\ud of yield and yield in a diallel cross at Pendleton, Oregon and\ud Lind, Washington. An additional parent was added to the diallel cross\ud at Pullman, Washington. Two levels of nitrogen and five replications\ud were utilized at each of the locations. Ten seeds of each F₁ or parent\ud were blended with 200 seeds of WA 4303 and seeded in a ten foot row.\ud This was done in order to simulate a solid seeding with the limited\ud amount of F₁ seed. The data were analyzed by using Griffing's diallel\ud analysis, Method 4, Model 1.\ud The morphological traits measured were: heading date, kernels\ud per spike, spikelets per spike, kernels per spikelet, weight per 1000\ud kernels, kernels per plant, spikes per plant and yield per plant.\ud In an effort to obtain an unbiased combining ability estimate,\ud a blend method of seeding (F₁'s mixed with a short semidwarf\ud WA 4303) was used in order to simulate solid seedings. A comparison\ud between the solid and blend seedings indicated, however, that\ud WA 4303 did not exert the same influence on all hybrids. The\ud estimate obtained should be less bias than those results obtained\ud from other investigations where space or hill plantings were utilized.\ud Significant general combining ability estimates were associated\ud with all measured traits within the six environments, except for kernels\ud per plant (high fertility) at one location. Specific combining\ud ability estimates were significant for all traits at one location. Only\ud part of the measured traits exhibited significant specific combining\ud ability effects at the other two locations. Under the environments\ud where specific combining ability estimates were significant they were\ud larger than those for general combining ability for most traits.\ud The lines utilized in this study were previously selected on the\ud basis of their yielding ability. Therefore, the additive portion of the\ud total genetic variance had already been maximized. The high specific\ud combining ability estimates were attributed to this previous selection\ud for additive gene action in addition to the nonadditive gene action\ud which resulted from crossing of the selected lines.\ud Heterosis and heterobeltiosis were exhibited by some hybrids\ud for all measured traits and within all six environments. Heterosis and specific combining ability estimates were not always in close\ud agreement which could result in some confusion in determining\ud which crosses would perform best in a specific environment. These\ud data would indicate that specific combining ability is a valid estimate\ud of the total amount of nonadditive genetic variance present but in\ud addition the additive genetic variance and the multiplicative action\ud of the components are important and contribute to heterosis.\ud The influence of the environmental-genotypic interaction as\ud measured across locations was greater for specific combining\ud ability estimates than that for the estimates of general combining\ud ability for all traits measured. This suggests that the nonadditive\ud genetic variance is more susceptible to changes in environmental\ud conditions than is the additive portion of the total genetic variance