Correlation-Regression Analysis of Economic Traits in Breeding of Oil Flax

Abstract

The main task of plant breeding is to increase yields. The crop is the result of the life cycle that the plant has passed, and the maximum yield is the realization of the most favorable equilibrium between all its components. Breeding for high yields of linseed oil is one of the priority areas, along with breeding for oilseed and altered fatty acid composition of oil. Zaporizhzhya breeding school is successfully working in this direction and the result of this work - created in different years 13 varieties, developed a model variety for the Steppe zone, a variety of source material and methodological basis for the creation of varieties. Statistical correlation-regression analysis is increasingly used to establish patterns of correlation between quantitative and qualitative traits, to clarify the complex relationships between many different causes and their mutual effect on one another. The purpose of our research was to establish correlation relationships of yield with the main economically valuable features of linseed oil and to obtain regression equations of productivity elements to improve the efficiency of further breeding work. We found correlation between the main economically valuable traits of the nine flax varieties of Zaporizhzhya breeding. Five-year estimates show significant correlations of yields with basic economic values that are high and stable enough in different weather conditions. The absence of very weak ones and the small number of weak links between the variables are probably due to the presence of similar complexes of selected features in the studied varieties. A very close direct correlation between yield and seed weight per plant (r = 0.99), yield and number of lateral stems per plant (r = 0.92) was established. Evaluation of the link strength showed the existence of a direct close correlation between the mass of seeds per plant and the number of lateral stems (r = 0.93), as well as a close correlation between the duration of the growing season and height (r = 0.87 ) and between height and number of lateral stems (r = 0.83). Seed yield is the most important trait that characterizes the genetic potential of modern oilseed flax varieties. Therefore, we performed a paired correlation-regression analysis of the dependence of yield on the studied economic and valuable features. As a result of our analysis, it was found that the most closely correlated yield was correlated with the seed mass of 1 plant (r = 0.99 ± 0.01). In our opinion, as a result of purposeful breeding work, genotypes have been selected that are capable of forming a large number of seeds on each plant. In the course of the researches the regression equation was calculated and the degree and orientation of the relationships between a number of investigated breeding traits were revealed. In the course of the researches the regression equation was calculated and the degree and orientation of the relationships between a number of investigated breeding traits were revealed. The diagrams we have constructed and the calculated regression equations will allow us to influence the formation, control the quantitative traits and control the systemic phenomena in the selection process of linseed oil in further studies. The use of correlation regression analysis allowed us to obtain models of the elements of cultural productivity. The calculated regression equations will allow us to predict with precision the various important economically valuable features of linseed oil and to increase the efficiency of further breeding work.