ABSTRACT The soil characteristics play a pivotal role in shaping the efficacy of pre-emergent herbicides in the context of weed control and their phytotoxic influence on the target crop. This study aimed to assess the effect of soil attributes on the efficacy of the S-metolachlor herbicide, as well as to determine its optimal dose in relation to soil attributes. The experiment was conducted under greenhouse conditions, in a 6 x 8 factorial design, with five replications, using 6 soil types [GMd (Typic Humaquept), NVe (Rhodic Eutrustox), CXbd (Typic Dystrustepts), LVw (Rhodic Acrustox), LVdf (Rhodic Haplustox) and RQo (Typic Quartzpsamment)] and 8 doses of the product [0, 1/8x, 1/4x, 1/3x, 1/2x, 1x, 2x and 4x (with x = 1,920 g ha-1)]. Dose-response curves were constructed and the outcomes showed that, for the LVw, RQo and CXbd soils, there was a control of over 90 % with less than half of the recommended dose, while, for the other soil types, lower reductions are possible. The base saturation and soil organic matter content showed a substantial negative correlation (-0.73 and -0.74, respectively) with the efficacy of the product. The S-metolachlor doses required to achieve a control of 90 % are contingent upon specific soil attributes, especially the variables base saturation and organic matter. The clay content did not present any correlation with the S-metolachlor doses for the studied soils.