grouping of alfalfa genotypes based on different characteristics using multivariate statistical analysis

in order to analyze genetic diversity and determine the most effective characteristics on seed yield, 49 alfalfa genotypes including foreign and iranian germplasms, were evaluated using a simple lattice design with two replications in researches station of natural resources and agricultural research center of east azerbaijan province. analysis of variance revealed significant genetic diversity among genotypes with respect to some of the traits. coefficient of variation (c.v.) was smallest for days to end of seeding and highest for seed yield. grouping of genotypes using ward's algorithms and all the traits assigned 49 genotypes into three groups. assignment of some landraces and foreign genotypes in one group indicated incongruity of genetic variation and geographical origins of genotypes. in grouping based on seed yield and related traits identified using path analysis, genotypes analyzed were classified into two groups. in this classification all of the landraces from east azerbaijan province were located in one group. in factor analysis, five factor with eigenvalues greater than one explained 77.02 percent of total variance.