Abstract:
Seed production of 2 experimental winter- green varieties of Setaria sphacelata, EHB (released as cv. Solander) and EHC, was compared with the winter-green cv. Narok in a split-plot experiment with 2 levels of applied nitrogen (N) fertiliser. The experiment continued for 3 years (6 harvests), and 3 subharvests at 1-2-week intervals were taken at each major harvest to allow for possible phenological differences between varieties. Total numbers of basal tillers, and of basal and aerial tillers with inflorescences, were counted at each harvest. Ripe and shedding inflorescences were counted, dried, and threshed by hand, and the seed was sieved and aspirated. The aspirated seed was stored for 6 months, the percentage of spikelets containing caryopses was evaluated, and samples were germinated in petri dishes. Aspirated seed yields of Solander were mostly 50-100% higher than those of Narok, except where yields were generally low, associated with low levels of applied N, and in later harvests of the experiment. All 3 entries showed a progressive decrease in seed production over time, and no consistent phenological difference was apparent. The improved seed production of Solander was associated with a markedly higher tiller fertility, lower basal tiller density, and slightly longer inflorescences. Germination percentage of Solander was consistently lower than that of Narok, and it was postulated that this was due to a higher level of seed dormancy. Nitrogen fertiliser had a marked positive effect on seed production, inflorescence numbers, tiller fertility, basal tiller density, and inflorescence length in all populations.