Variation, genetic divergence and interrelationship analysis in buckwheat

Fourty four accessions of buckwheat gennplasm, of which 39 were ofIndian origin, 1 of USSR, 1 of Hungry and 3 of Japan were grown in complete randomized block design. A random sample of5 plants for each accession was used for data recording on 12 traits. The coefficient ofvariation both at the phenotypic and genotypic level was high for seed yield! plant, seeds/cyme, leafwidth and number of branches/plant. Both heritability and genetic advance were high for seed yield, seeds/cyme, number of branches and leaf width. On the basis of relative magnitude of D2 values, the 44 genotypes were grouped into 7 clusters each having 14, 10,7,7,4, I and I genotypes, respectively. Intra-varietal group clusters had genetic distance 0.00 to 12.81 and inter-varietal group clusters had 13.27 to 43.99. Cluster VII showed the highest mean for leaf length, leaf width, IOO-seed weight and seed yield. The genotypes fell in cluster IV showed good seed setting as the ratio ofthe number of flowers /cyme and seeds/cyme were highest, and showed earliest maturity and dwarfness. Of the total variation, the first four canonical roots accounted for 90.16%. The first root accounted for 40.47 % of the variation, followed by 23.13 % for the second, 16.51 % for the third and 11.05 % for the fourth. Seed yield/plant showed significant positive correlations with days to flowering, days to maturity, number of leaves, leaf length, leaf width, seeds/cyme and 100-seed weight. Path analyses showed direct positive effects of days to maturity, leaf length, leaf width, 100-seed weight, seeds/cyme and negative direct effects of number of branches, plant height, flowers/cyme, number of leaves, days to flowering, and number of inter-nodes on seed yield. On the basis of results obtained from various analyses it is suggested that selection ofbetter plant type in buckwheat can be achieved through days to maturity, leaf length, leaf width, IOO-seed weight, number of internodes and seeds/cyme