GENETIC DIVERSITY OF SUNFLOWER BASED ON CLUSTER ANALYSIS

An evaluation experiment was carried out on 13 sunflower genotypes at the farm of Field Crop Res. Sta. of College of Agriculture / Univ. of Baghdad for two spring seasons (2005 and 2008). The experiment aimed to group similar genotypes into homogenous groups through cluster analysis and estimate genetic diversity between studied genotypes. Hierarchical clustering technique was used with respect to plant height, leafe area, head diameter, seed number per head, weight of seed and seed yield. Resemblance matrix was constituted then euclidean values were estimated .The genotypes were grouped into six clusters, where cluster I, III and IV included three genotypes followed by cluster II which has two genotypes followed by clusters V and VI which have one genotype for each. It was observed that genotypes Kws and Allstar, Flamme and Zahratalaraq, Macao and Florasol have similar response pattern based on minimum euclidean. Genotypes Shimoos and Ibis was found to be better in performance and should be involved in crosses based on maximum euclidean values that express high level of genetic diversity. This can lead to give a high heterotic response and wider segregation after hybridization. This statistical technique can be auseful tool when other molecular techniques are not available.