Abstract:

Weed management trials in sunflower were conducted at the National Semi Arid Resources Research Institute (NaSARRI) in 2010, 2011 and 2012. Hand weeding started at 15, 30 and 45 days after planting (dap) and the weeding frequencies were once, twice or thrice. A herbicide called Select of the cloheaxanedione family and with clethodim as the active ingredient was also applied to some plots at 100 mls/15 litres of water once at 15 dap, and twice at 15 and 30 dap. Numbers of weed species and their counts were taken regularly before and one week after treatment applications. Data was recorded on sunflower growth parameters such as height, stem and head diameters. Sunflower was harvested from the four middle rows, threshed, dried winnowed, weighed and yields calculated. Costs involved in the various weed management treatments were recorded and weed control margins calculated. Data from this study emphasize that sunflower must be weeded, and weeding should start at, or around 15 dap but not exceed 45 dap. The earlier one starts to weed, the better the performance and more delays lead to reductions in yields. Starting to weed at 15, 30 and 45 dap increased yields by 58.8, 42.9 and 40.4 % over the non-weeded controls, while application of Select once or twice increased yields by only 10.9 and 10.6 % respectively. To maximize incomes and weeding margins in sunflower, weeding should be done once. The herbicide (Select) significantly reduced counts of grass weeds but had insignificant effects on broad leafed weeds. Yields, incomes and costs associated with its use whether once or twice do not justify its recommendation for sole use on sunflower in Uganda.