-The use of herbicides must be made in a rational way, giving priority to reduce possible adverse effects on subsequent crops. The objective was to evaluate the residual effect of pre-emergent herbicides on growth and initial development as well as the biochemical composition of snap bean seedlings. The experimental design used was the completely randomized, with four replications in factorial scheme 3x4, three snap bean cultivars (Macarrão Favorito, Macarrão Preferido e Macarrão Trepador) and three pre-emergent herbicides, plus a control (without herbicide). Was use the residual doses of pre-emergent herbicides sulfentrazone in 0.3 g ha -1 , diclosulam in the concentrations of 0.126 g ha -1 and imazethapyr in 0.05 g ha -1 . The work was developed in a seeds analyze laboratory with the purpose of minimizing possible adverse effects of the environment. The seeds were sowing on germitest paper moistened with solutions containing the herbicides, with each respective treatment. The following characteristics were evaluated: germination (GER), germination speed index (GSI), seedling length (SL) and seedling dry mass (SDM) and soluble protein content (PROT) in addition to the enzymatic activity of catalase (CAT), peroxidase (PER) and phenylalanine ammonia-lyase (PAL). The phytometric variables SL, SDM, GER and GSI reduced in the presence of herbicides, and for the levels of PROT and enzymatic activity of CAT, PER and PAL there was observed a variation according to the genotype. The sulfentrazone showed up as the product of higher toxicity, as well as the Macarrão Trepador as the genotype with lower tolerance.