Multivariate analysis of interaction between hybrids and planting dates for oil content, yield and yield components in sunflower. [Doctoral dissertation] | Multivarijaciona analiza uticaja interakcije hibrida i rokova setve na sadrzaj ulja, prinos i komponente prinosa suncokreta. Doktorska disertacija
2009
Balalic, I.
Three sunflower hybrids (Miro, Rimi, and Pobednik) were grown in eight sowing dates during three vegetation periods (2005, 2006, and 2007). The following characters were analyzed: total dry matter accumulation and distribution among various plant organs, leaf area per plant, plant height, head diameter, seed yield, 1000 seed weight, oil content and oil yield. Interaction hybrid x sowing date was analysed by AMMI (Additive Main Effects and Multiplicative Interaction) model. In order to evaluate the biological basis of interaction hybrid x sowing date, climatic covariables were included and PLSR (Partial Least Squares Regression) and FR (Factorial Regression) models were used. The results of three-factorial ANOVA showed highly significant differences for main effects (year, hybrid, sowing date) and their interactions for most of the investigated characters. Most of the characters were predominantly influenced by the year of growing. In AMMI ANOVA for all characters significant differences were found for IPC1. Results of biological explanation, using PLSR and FR models, indicated that climatic covariables in the flowering stage, such as temperatures, precipitations, sunshine hours and relative air humidity had major influence on hybrid x sowing date interaction. Both models showed similar results in the ranking of covariables in the explanation of hybrid x sowing date interaction in sunflower. Very similar results were obtained using correlation analysis in AMMI model. The results of these investigations could be of importance in the recommendation of high yielding, stable sunflower hybrids and most convenient sowing dates.
显示更多 [+] 显示较少 [-]