Herbicides that present soil residual activity may become a problem in agricultural crops due to the possibility of crops successors intoxication. It is hard to detect residues of hormonal herbicides using the bioassay technique, due to the variation in the growth response of the indicator plant. The objective of this work was to select plants and variables susceptible to the herbicides 2,4-D and picloram for soil residue indication..To make it possible, two bioassays were performed,, where each of them consisted of 60 treatments in a factorial scheme with the combination of three plant species (Lycopersicon esculentum, Cucumis sativus and Beta vulgaris), two substrates (sand and soil sample) and 10 subdoses of herbicide (0, 3, 6, 12, 22.5, 47, 94, 187.5, 375 and 750 g ha-1) and (0, 3, 6, 9, 18, 30, 60, 123; 246 and 501 g ha-1) for 2,4-D and picloram, respectively. The experimental design for both trials was randomized blocks with four replications. Increasing doses of the herbicide caused reductions in plant height, total dry matter and visual intoxication of the assessed species. The visual intoxication was the most suitable variable for the determination of low residues levels of D 2.4 and picloram in soil. B. vulgaris was the most sensitive species to 2,4 D and L. esculentum was the most sensitive species to picloram,
test plant; Lycopersicon esculentum; Cucumis sativus; Beta vulgaris
