An experiment was carried out in Monte Alto, SP to study the effects of different of weed control periods on the yield of onion (Allium cepa) bulb, 'Mercedes' cultivar, under the transplanting system. The experiment was arranged in a completely randomized block design, with four replications. The treatments were arranged in 4 x 6 factorial design with four initial weed removal periods (after transplanting until 0, 7, 14 and 21 days) and six final weed removal periods (from 28, 42, 56, 70, 84 or 98 days after transplanting until harvest). Two controls were adapted with and without weed control throughout the onion cycle. Lycopersicon esculentum, Portulaca oleracea, Eragrostis pilosa and Galinsoga parviflora were the most important weeds in the area. No interaction occurred between the different weed controls. Weed control should be maintained until 14 DAT and be restarted at 28 DAT to prevent significant yield reductions in relation to the control without weeding. Weed interference throughout the onion cycle reduced yield in 94.5%.
Allium cepa; weed community; competition
