Selection of agro-morphological traits related to spring wheat yield in dryland conditions as selection indices

Introduction Drought is the most common abiotic stress in crops. The study of agro-morphological traits and their relation to grain yield can be helpful in determining the most important agro-morphological traits affecting grain yield and their use in indirect selection. As the water resources available for agriculture become more limited, developing drought tolerant lines is significantly important. The purpose of this study was to identify the most important traits affecting grain yield in rainfed conditions and to use these traits in clustering to identify the best genotypes for introducing or conducting crosses to create new breeding populations. Materials and methods This research was carried out to investigate the relationship between agronomic traits and yield using 132 spring wheat genotypes in alpha-lattice design with two replications at the Agricultural Research Station of Gonbad-e-Kavos. Pedigree information is available on the CIMMYT Web site (http://orderseed.cimmyt.org/iwin/iwin-fb-table.php). Each replicate contained 12 blocks, and in each incomplete block, 11 genotypes were placed. Each variety was cultivated in four lines with a length of 1 meter and 0.2 meter lines spacing. Each plot was planted based on seed rate of 140 kg ha-1. Cultivation was carried out in rainfed condition. Characteristics of grain yield (g/m2), biomass (g), harvest index, number of spikes, height (cm), days to heading, days to maturity, grain filling duration, seed filling rate (g/m2/day), Flag leaf area (cm2), flag leaf specific area (cm2/g), leaf dry weight (g), 1000 grain weight (g), spike length (cm), spike weight (g) and grain weight per spike (g) were evaluated for all genotypes. Results and discussion The highest correlation was observed respectively between yield with seed filling rate (0.839), harvest index (0.838), biomass (0.858), 1000 grain weight (0.579), grain weight per spike (0.437), and weight of the spike (0.409). Yield had a negative and significant correlation with days to maturity (-0.566) and days to heading (-0.446). The negative and significant correlations between yield under rainfed conditions and days to maturity are a mechanism for avoiding drought stress at the end of the season. The genotypes that earlier complete their physiological maturity could run away from drought stress at the end of the season and do not face stress. Seed filling rate and days to heading were introduced into the model as important traits affecting grain yield. These traits account for 78.1% of the variation in yield. Path analysis indicated that seed filling rate had the most direct effect on yield (0.773). Also, the days to heading had a direct effect of -0.285 on performance. Cluster analysis in this study was conducted with the aim of grouping the studied genotypes in terms of performance related traits. Group 5 and 7 were superior to other groups in terms of traits. The yield of genotypes No. 38, 39 and 52 was 572.5, 578.3 and 535.8 g/m2, respectively. Genotype No. 20 earlier heading than other genotypes. This genotype ranked second among the other genotypes for days to maturity. Therefore, this genotype has the ability to escape from the end-season stresses. Genotype No. 39 had the highest 1000 grain weight and the highest grain weight per spike. Conclusions None of the genotypes in Group 5 or 7 were superior to all performance related attributes. Therefore, in order to aggregate the characteristics associated with performance in a genotype, crossing of several genotypes and the production of breeding populations using these crosses should be used. Genotypes No. 20, with the least days to heading and the least days to maturity, genotypes No 12, 38, 39, 52, 67 and 85 with high seed filling rate and genotypes No. 103, 85, 39 and 20 with the highest thousand seed weight for Crossings and production of breeding populations can be used. It is suggested that a research be carried out in a few years using genotypes in groups 5 and 7 to select the best genotypes among them more confidently.