Background: Drought, endangering agriculture, poses a threat to food security. Sorghum, with its low water requirement and high water use efficiency compared to other grains such as corn and wheat, holds promise in mitigating the impact of drought. However, due to limited cultivation of this crop in Iran and specifically in Bushehr province, there is a need for identification and introduction of compatible varieties.
Objectives: The project aims to (a) study the level and structure of diversity among sorghum germplasms collected from gene banks in Germany and Australia using multivariate statistical methods to harness the potential in sorghum breeding programs in Iran, especially in Bushehr province, and (b) analyze relationships between traits to understand the patterns and correlations among them, particularly focusing on 13 morphological traits.
Methodology: The research was conducted in an experimental design with 62 genotypes and 7 controls at the Faculty of Agriculture, Persian Gulf University, in Borazjan. Five samples out of ten in each row were randomly selected, and measurements were taken for days to germination, days to flowering, and morphological traits (plant height, leaf number, leaf width) as well as grain yield.
Findings: The results from correlation matrix indicated positive correlations between phenological traits (days to green-up, days to panicle emergence) and yield-related traits (cluster weight, grain number per cluster). Stem diameter and leaf area showed positive correlations with many other agronomic traits. Yield had a strong positive correlation with grain number per cluster. Stepwise regression analysis justified that grain weight, grain number per cluster, and stem diameter explained a major portion of grain yield. Path analysis revealed that grain weight and grain number per cluster were the most influential traits affecting grain yield directly. Principal component analysis identified five principal components explaining 76.18% of t