Improving the seed yield of cumin (Cuminum cyminum L.) genotypes using different selection indices under drought stress conditions

In order to improve complex traits such as yield, one of the most effective approaches to select the desired genotypes is the use of selection indices. In this research, in order to evaluate the efficiency of different selection methods under stress conditions, 49 top cumin (Cuminum cyminum L.) genotypes, selected in the second year of a two-year plan were evaluated. This research was conducted as a field experiment in a randomized complete block design with two replications under drought stress conditions in 2012 at the research farm of Shahid Bahonar University of Kerman. The selection indices Smith-Hazel 1 and 2 and Pesek-Baker were computed based on seven traits including plant height, number of lateral branches, umbels, umbellules, seeds, seeds weight and aerial parts weight (traits/plant). Also, the direct and correlated responses of these traits with seed yield were obtained. All three indices were evaluated in two forms, along with yield and without yield. The results of direct and correlated response of traits showed that genotypes with higher number of seeds and umbelluleshad higher yield potential. In the Smith-Hazel 1 and 2 indices, two traits including the number of seeds and umbelluleshad the highest response to the selection, while in the Pesek-Baker index, the number of branches was the highest response to the selection. The results indicated that the Smith-Hazel index had the highest selection efficiency and could be used to select appropriate genotypes in the cumin breeding programs under drought stress conditions.