Evaluation of the Traits Impact Model on the Lentil Seed Yield to Determine Selection Methods (original) (raw)

Background: Lentil (Lens culinaris Medik) is a cool season seed legume and a good source of nutrients needed by humans, including protein, carbohydrates, vitamins, and minerals. The production of high-yielding and high-plant-height varieties is one of the lentil improvement goals. For this purpose, it is necessary to collect and evaluate germplasm as a base population, along with identifying and utilizing lines with high potential and other desirable traits. The environment has a significant effect on the crop production of this plant. Therefore, direct selection is important for seed yield. Since seed yield depends on the yield components, the yield and its components must be regarded as a group at the selection time to improve the yield. To properly increase yield and economic efficiency, we need to collect desirable lines with desirable genes and transfer these genes to cultivated lines to produce desirable cultivars. Consequently, sufficient information is necessary on accessible genetic materials, which is possible by evaluating different traits. Methods: To assess the model of the simultaneous effect of traits on the lentil seed yield for determining the selection procedure in native lentil lines of Zanjan province, an experiment was conducted in the research farm of the Faculty of Agriculture, Zanjan University, during two cropping years 2017-2018 and 2018-2019. In both years of the experiment, improved cultivars, such as Kimia, Sabz Kohin, Gachsaran, Maragheh, and Bilehsavar, were used as control cultivars. The first-year experiment was conducted in an augmented design based on a randomized complete block design with 200 lines. Each experimental unit included a 1-m row. The distance between the rows was 25 cm, the distance between plants in a row was 5 cm, and the planting depth was 5 cm. Two rows of Kohin-Sabz lentils were planted as margins at the beginning and end of each block. Due to obtaining a sufficient amount of seeds from the first year, the second-year experiment was carried out as a simple lattice design with two replications and larger experimental units for the lines selected from the first year. Each experimental unit included two 1-m lines. The distance between the rows, the distance between plants in a row, and the planting depth were similar to the first-year experiment. Two rows of Kuhin green lentils were planted as margins at the beginning and end of each incomplete block. The measured traits included phenological and morphological traits, yield, and yield components per plant and unit area. Results: Among the studied traits, the highest coefficient of variation was obtained for the number of seeds, biomass, straw yield, and seed yield. The coefficient of correlation showed that the number of seeds and seed yield were positively and significantly correlated with phenological traits, such as the podding period, physiological maturity, and seed-filing period, as well as morphological traits such as plant height and first branching height, respectively. A positive and significant correlation was observed between the number of seeds and seed yield with the number of pods per plant and biomass per plant. In regression analysis by the stepwise method, plant height was the first trait entered into the model, which could explain 46.8% of the variation related to seed yield per plant. Then, the number of seeds per plant, 1000-seed weight (TSW), straw yield per plant, and the seed-filling period were entered into the model, respectively, which could totally explain 67.5% of the variation related to seed yield per plant. The results of the path analysis showed that the number of seeds per plant had the most considerable direct and positive effect on seed