Residual effect and interval of herbicide application on soybean

In the burndown of <i>Conyza</i> spp., non selective herbicides are recurrent in soybean pre-sowing, and it is necessary to respect the time interval between the application of these products and the sowing of the crop, known as the plant back. Therefore, the objective of this study was to evaluate soybean phytotoxicity resulting from the application of postemergence herbicides commonly used for the control of <i>Conyza</i> spp., applied at different intervals between herbicide application and soybean sowing. A field experiment was conducted in randomized blocks, with four replicates, in a 6 × 4 + 2 factorial design, with the following herbicides applied: halauxifen + diclosulam (6.3 g ae ha ^-1 + 31.9 g ai ha ^-1); triclopyr (960 ae ha ^-1); fluroxypyr + clethodim (300 ae ha ^-1 + 210 g ai ha ^-1); atrazine + mesotrione (500 + 50 g ai ha ^-1); dicamba (480 ae ha ^-1); and imazapic + imazapir (78.75 + 26.25 g ai ha ^-1) at 0, 15, 30 and 45 days before soybean sowing (DBS). At 0 DBS, soybean phytotoxicities close to 70, 40 and 10% were observed for dicamba, atrazine + mesotrione and imazapic + imazapyr, respectively. For the intervals of 15, 30 and 45 DBS, percentages close to 50, 15 and 6% phytotoxicity were observed in soybean after the application of dicamba. For the other treatments and time intervals between herbicide application and soybean sowing, no significant phytotoxic effects were observed. However, notably, the herbicides dicamba and imazapic + imazapyr resulted in a significant reduction in soybean yield at the tested intervals. In this sense, dicamba was the herbicide that caused the most damage to the crop, regardless of the period, and the safest herbicide for the crop was fluroxypyr + clethodim, with a low phytotoxicity index at 0 days.