Genetic Divergence for Economic Characters in Gladiolus Under Different Environments

Multivariate analysis as measured by Mahalanobis's D² statistic was studied for 20 quantitative characters in 25 cultivars of gladiolus under three environments. The cultivars differed significantly for all the 20 characters considered collectively and were grouped into five clusters on the basis of relative magnitude of D² values at individual environment and pooled analysis. The maximum number of cultivars was in cluster I at E₁ & E₃ (16 and 11, respectively) and in cluster III at E₁ (II), whereas in pooled analysis it was maximum in cluster I (12). Based on cluster means, characters like days to 50% heading, first floret colour showing, first floret opening, last floret opening, number of florets per spike, average weight of a corm and propagation coefficient were the major factors for differentiation among 25 cultivars. Maximum genetic distance was observed between clusters IV & V (E₁ & E₃), II & V (E₂) and III & IV (pooled). However, no close correspondence is evident between geographical distribution to genetic divergence. The study also revealed that clustering behaviour and mean performance of cultivars of individual cluster were not consistent over the environments because of larger genotype × environment interaction. It is, therefore, suggested that for effecting hybridization and desired improvement, the cultivars should be selected randomly from divergent clusters possessing good economic traits.