Genotype x Environment Interaction for Yield and Yield Components in Some Sunflower (Helianthus annuus L.) Hybrids *

Abstract

A field experiment was carried out at two locations, New Halfa
and Rahad, in the summer of 2003/04 and winter of 2004/05 to estimate
the genotype x environment interaction for yield and its components
among 19 locally developed sunflower (Helianthus annuus L.) hybrids
and one introduced hybrid (Hysun 33). A randomized complete block
design with four replications was used at each location. Data on yield and
its components; namely, number of seeds/head, percentage of empty
seeds, 1000-seed weight and seed yield/plant, were collected. The
combined analysis of variance for all characters showed highly significant
differences at the genotype level. The genotype x season interaction was
only significant for 1000-seed weight and seed yield/ha. The genotype x
location interaction was significant for 1000-seed weight, seed yield/plant
and seed yield/ha, whereas genotype x season x location interaction was
only significant for 1000-seed weight. The number of seeds/head and the
percentage of empty seeds showed non-significant differences at the first
and second order interactions. The highest yield per hectare and per plant
averaged over the four environments was recorded for Hysun 33, whereas
the lowest ones were scored by Ka99x11. The analysis of genotype x
environment interaction for 1000-seed weight, seed yield/plant and seed
yield/ha showed that the major part of the variance was due to non-linearcomponents. The significant differences among the evaluated hybrids at
the different interaction levels suggest that seed yield of sunflower is
highly dependent upon the environmental conditions. Therefore, multi-
location testing for several years is needed to identify high yielding stable
hybrids.