Introduction: Studying the genetic system that controls important agricultural traits is one of the most important prerequisites for choosing a suitable breeding method. The most genetic diversity of a plant species can be seen in its native genotypes and wild species. Investigating heritability, how genes work, and determining an effective breeding method to improve physiological traits, especially in conditions of moisture stress, is very importance. Drought stress is one of the most important factors that reduce the yield of plants, considering that the safflower plant is mostly grown in semi-arid and dry areas, nowadays much attention has been paid to the production of drought-tolerant varieties. Safflower is a one-year plant from the chicory family as an oil and fodder plant with various uses that can be cultivated under drought and normal stress conditions.
Materials and Methods: Therefore, in order to investigate the mode of gene action for some quantitatively important traits in safflower, a cross was made between two cultivars of Goldasht (resistant to drought stress) and American (sensitive to drought stress). And after self-fertilization of F1 plants and crossing them with both parents, F2, BC1.1 and BC1.2 generations were obtained. And the parents along with the F1, F2, BC1.1 and BC1.2 generations resulting from the cross were tested under moisture stress conditions in the research farm of the Research Institute of Plant Production Technology of Shahid Bahnar University in 2021-2022. The experiment was conducted in the form of split plots in the form of a randomized complete block design with two replications Moisture stress was applied at the stage of 10% tillering and until the stage of full maturity. The examined traits were 8 traits, including height, number of sub-branches, main stem diameter, number of bolls, single plant yield, photosynthesis, number of days to 50% flowering and SPAD were noted.
Results and Discussion: The results of analysis of variance showed that the mean square of the generations for all the traits in this crossing is significant, so the best genetic model was investigated for each studied trait. The interaction effect of generation and location (irrigation conditions) was significant for the studied traits. To check the appropriate genetic model, Meter and Jakens method and individual scale test (A, B, C and D) were used. The results of the above tests indicate the existence of additive and dominant effects, as well as epistasis effects (additive × additive, additive × dominant, and dominant × dominant) also played a role in the genetic control of most traits. In normal irrigation conditions, most of the traits except Spad, number of bolls and diameter of chi-square and their individual scale test were not significant and were affected by the three-parameter additive-dominance model. Also, in the conditions of moisture stress, most of the traits were significant in the control of inheritance, except for the main diameter of chi-square and scale test (A, B, C, and D), and additive effects, dominance and epistasis had an important role in their control. The existence of dominance × dominance effects compared to the parameter of additive effects in the boll number trait indicates the high importance of non-additive genetic effects (epistasis) and complex inheritance of this trait. Also, selection in the early generations for the above trait will not be successful. The additive effects played the most important role in controlling the inheritance of the main diameter trait under moisture stress conditions. Therefore, it will be useful to modify this attribute from the selection method. Also, additive effects play the most important role in controlling the inheritance of the number of sub-branches in normal conditions (no stress). But in the conditions of moisture stress, the mentioned trait was affected by the additive × additive epistasis. Therefore, one cannot hope for the success of selection in the early generations of diverging populations.
Conclusion: The genetic control model of all studied traits under both stress and non-stress conditions were not similar in terms of presence and absence of non-allelic interactions and were affected by stress.