Study on relationships among yield and some agronomic traits using Path Coefficient analysis in Pure lines Soybean

In order to study on relationships among yield and some agronomic traits, 141 pure lines soybean as well as three commercial varieties (controls) (Williams, Sahar, DPX) were arranged at simple lattice model (12×12) with two replications in Gorgan agricultural research station in 2009. Analysis of variance demonstrated a significant difference among pure lines at 1% and 5% probability levels for all traits. The results showed positive significant correlation among yield with percentage of protein, day to flowering, number of branches, pods per plant, plant height, and weight of seed per plant (g). Also the results indicated negative significant correlation among yield and oil percentage. Stepwise regression showed day to flowering, plant height, seeds per pod/plant, number of nod/plant and protein percentage could explain variations in yield. Path analysis showed that number of pod/ plant and plant height had highest direct effect on grain yield of soybean, and because their direct effects had agreement with ctotal effects thus for yield increase effective select lines taller plant with number pod. Using cluster analysis with UPGMA method and the squared Euclidean distance as a measure of the similarity, the cultivars were groped in to five groups. According to results, the number of pods/plant and plant height were criterion. index to improvement of soybean yield in breeding programs.