Although the use of glyphosate tank mixtures for managing weed species of difficult control is prohibited by the Brazilian legislation, it has been a common practice among farmers. This work aimed to evaluate the efficiency and selectivity of herbicide tank mixtures using glyphosate to control Commelina benghalensis L., Tridax procumbens L. and Cenchrus echinatus L. in soybean crop RR®. A field experiment was conducted in the Municipality of Maracai, São Paulo State, between November, 2006 and March 2007, using the genotype CD-214RR® in a complete randomized block design with 21 treatments and four replications. The treatments consisted of the following applications: glyphosate (180; 360; 540 and 720 g ha-1); glyphosate in sequence (180/360; 360/360 and 540/360 g ha-1); glyphosate + chlorimuron-ethyl 360+10; 540+10 and 360+5/ 360+5 g ha-1); glyphosate + lactofen (360+120; 540+120 and 360+60/ 360+60 g ha-1); glyphosate + cloransulam-methyl (360+30; 540+30 and 360+16,9/ 360+12,9 g ha-1); glyphosate + carfentrazone (360+4 g ha-1); glyphosate + imazethapyr (360+50 g ha-1); glyphosate + imazethapyr (177.8+30 g ha-1) and controls with and without weeding. Despite the similarity in grain yield between treatments with glyphosate alone and sequentially at doses 540, 720 and 540 / 360 g ha-1, the tank mixtures with chlorimuron-ethyl, cloransulam-methyl, lactofen and imazethapyr favored control of weed species tolerant to glyphosate such as C. benghalensis and T. procumbens.
Commelina benghalensis L.; Tridax procumbens L.; Cenchrus echinatus L.; herbicide; effectiveness
